Hello everyone, Welcome to this session
on Microsoft Azure. Today in this session, I'll be teaching
you guys what is Microsoft Azure end-to-end. So let's get started. Now
before starting off with the video guys, I would request you all to like, share
and subscribe to our channel so that you don't miss out on any new updates from
Intellipaat. So now let's get started. So the first topic that we'll be talking
about is what is cloud computing and then move on to discuss the various
cloud models on which the companies operate. Once we're done with that we'll
talk about the top three cloud providers which exist in the market among which
Microsoft Azure is also present. Once we're done with that we'll move on and
get introduced to what is Microsoft Azure exactly and then move on to
discuss the core Azure architecture on which Azure operates. Once we're done
with that, we'll move on and talk about what are subscriptions and what are
resources which are a very important part of the azure operation. Once we're
done with that we'll move on and talk about the core Azure services that you
should know off when you're working as an azure professional and towards the
end I'll show you guys how to architect an Azure application end to end so
that you don't have any doubts and you can actually implement all the concepts
that you're learning in the video. Towards the end, we have also planned a
quiz for you guys which you have to attempt and the answers you have to give
in the YouTube comments and we'll verify it for you and tell whether you're right
or wrong. Alright guys, so let's get started with
the first topic which is, what is cloud computing? All right guys, let's now start
this session by first discussing what is cloud computing? right. So let's go ahead
and understand first cloud computing by its definition. So guys by definition,
cloud computing is nothing but the use of remote servers on the Internet to do
your tasks. It could be basically hosting an application, it could be hosting your database server, it could be anything right. When you use
servers on the internet or when you use somebody else's servers on the Internet
rather than using your own servers or computers, when you use somebody else's
computer on the internet that is called cloud computing right and the person or
the company who provides you with that kind of server is called a cloud
provider. this is a gist of what cloud computing is now obviously cloud
computing has more things to it and we will discuss that as we move along right.
Now to give you an example as to how you can imagine how cloud works is like this.
Let's say you are a developer okay and what do you want to do is you want to
host an application. So you basically start up a company and you have an idea
which is similar to let's say Instagram and that idea is a billion dollar idea
right and what you want to do is you want to start off by just creating the
application by yourself, you don't want to hire any people right because you
wanted to keep the cost minimum. So you develop the code by yourself and now the
application is ready but you have to make this application available to the
world, so how do you do that. So the first thing that you will have to do is you
have to buy some servers and you'll have to put that application on those servers.
So once your application is on those servers, the next top task would be to
basically make that application available on the internet right and how
to make it available on the Internet. You basically connect Internet to these
servers and you assign a public IP to these servers and now your
application is basically available on an IP address. next step ypu would do is
you would buy a domain and the next thing that you know is now the domain is
ready and now your application anybody on the Internet who goes to your particular domain would be able to access your application ok. So
there is this is the normal life cycle of how applications were launched before
cloud computing. Now the problem over here guys is that you invested some
money on the servers and guys servers are not cheap. They are very very expensive right and as a guy who is starting up with their company, you might think that this idea is worth billion dollars but if you actually
think about it, there's a lot of risk and there's a lot
of I mean there's a 50% probability because you are just starting off now
that you might fail. Although, there's a probability of a 50%probability that you
might succeed as well but there are a lot of risks that are involved and
there's a lot at stake over here because you're investing a lot of money on these
servers right. So this is the general way of starting up a business before cloud
computing emerged. Now after cloud computing, it has become
very simple now. Let's say you come up with an idea and now you want to deploy
your application, all you have to do is you will go on cloud and you would say
hey I need some servers for deploying my application. So you would get those
servers which would already be connecting to the internet or everything
would be set up by the cloud provider you just have to use them right and the
only thing the cloud provider would expect from you is he would say like hey
you know what you just have to pay me for the time that you're using the
server's for and that's about it. I don't need anything else from you right.
Now, that's an awesome thing. Now the kind of
pricing that cloud computing gives you is a pay-as-you-go price and that
basically means you just pay for the amount of time that you're using servers
for. In case of azure, it counts in terms of minutes right. So
the number of minutes that you would be using the servers for, it will charge you
according to that right. Now if you again think about what all changes came about
over here when I was not using cloud and when I'm using cloud now. The first
change that would come about here is if let's say you know you were not using
cloud computing and you bought these servers, now obviously somebody has to
manage these servers for you as well right and now when somebody has to
manage them for you, there's always a chance of downtime as well. So you always
have in your back of your mind will always thinking of whether my servers are up and
running or not. Do I need more people to handle these kind of service? When your
application becomes big, you would also have to scale these servers. So all these
problems would be at the back of your mind and at the same time you also have
to develop code for it right. So that's a lot on your plate. Now what cloud does
for you is, it says you know you just pay me for the amount of time you're using
the service for, I on the other hand will hire the workforce that is required to
manage these servers. I own the other hand as a cloud provider, will also scale
you servers as and when required, when the demand increases on the application. You
just sit back, relax and you code your application that's all you have right.
And guys this is the biggest advantage that you get with cloud computing right.
You can focus on the work that you have to do. Everything logistical or
everything operational with respect to managing servers is handled by your
cloud provider and what is better than paying less for that right. You're
paying less, you're hiring less workforce and at the same time you're getting a
better quality of internet hosting available. Basically, like you imagine
and now if you have to secure an application, you'll have to buy, you'll
have to hire a security engineer who will basically secure your application, you
will also have to hire people who will ensure that all that your application is
up and running, how by basically scaling them up and down, that your number of
servers so you would need deployment engineers, you would also have to keep
some servers in your stack so that whenever new servers are required,
whenever the demand goes up those can be plugged and played as and when required
right. So these are all the things that you have to do but now with cloud
computing you don't have to do that because everything is managed by the
cloud provider, you just have to code your application. So summarizing the
advantages that you get with cloud computing is that while you starting up,
you don't have to invest a lot right that is the number one advantage. Number
two. you can focus on your app development,
you don't have to focus on you know what is happening on my server? Are my servers
running or fine or not? Are the security patches updated or not? Everything will
be managed by your cloud provider and the third thing is that it requires like
workforce because obviously you are not hiring anyone to manage your servers,
everything is being done by the cloud provider. If you wanna hire people who
can actually hire more developers, so that your coding work is distributed
among them right. So these are the advantages guys that you get with cloud
computing. Now cloud computing is offered in various types of services
and that we will discover as we move along but those types of services leads
to different kind of applications or different kind of products on the cloud.
For example, you might have heard about this you might have heard about Google
Drive, you might have heard about Netflix, Airbnb. All these applications are
basically hosted on cloud and you might not even realize it right. Now if we talk about Google Drive, Google Drive is nothing but it's a
kind of service, its software as a service that is given to you by cloud
provider. What is software as a service, we discuss that as we move along but just
understand that Google Drive is a cloud product. It is basically hosted on
Google's infrastructure right. You don't care about where if you are uploading a
file on Google now, you don't care about where on which server is it being stored
right. You don't care about which operating system is that server running.
You don't care about you know whether the hardware on that server is good
enough to basically handle the speed with which you want to work. You don't
care about anything right. You just care about your file being there on Google
Drive and you know it's gonna be there. Why? Because you trust Google with your
things, you trust Google that it will manage everything on the infrastructure
side and that is exactly what cloud computing is all about. Similarly,
when you talk about Netflix. Netflix is a very popular video streaming surface I'm
sure all of you must have used it that is itself also who should you want to
cloud right. So Netflix will rely on Netflix to be always up and running and
Netflix itself relies on the cloud provider on which it is hosted that you
know they will ensure that I am up and running always right. So that is how cool, you know the concept of cloud is that a popular service such as Netflix is also
not caring about how this servers are managed, all of that is managed by your
cloud provider. Similarly, you must have also heard about
Airbnb, you must have also heard about the e-commerce come a company in the
Giant which is Amazon then there is also Prime video, all of these products
they are basically hosted on cloud right. So if you have them in there, you would
also recognize Flipkart. FlipKart is again an e-commerce giant which was
recently taken over by Walmart right. Flipkart also is completely hosted on a
cloud provider. In Flipkart's case, it's basically hosted on Microsoft Azure right. So, we will be talking more about azure as we move along but guys these are some of the products that you might be
using in your everyday life which are based basically hosted on the cloud
right. Alright guys, so before moving ahead just
a quick update guys, this is not the complete Azure course. If you want to
complete Azure training, you can actually head on to our website. We have
a course called Azure master certification training which will not
only get you certified in one Azure certification which is admin or
architect or developer but all the three right. So if you're interested, the course
link is in the description box. Please go ahead and click on that here and you
will get the course details. So without wasting any time guys, let's go ahead and
continue with the session. Now moving forward guys, we were talking about cloud
computing services. There are various kind of services that cloud computing
has to offer us. They are basically the cloud computing business is based on two
kind of model and we'll discuss those models in this
section which is cloud computing models. Alright guys, now let's discuss the various
cloud models on which basically cloud companies are working and are
earning right. So basically, cloud computing gives us
two models. The first model is the deployment model and the next model are
the service models. Now what are these? Let's ask us them one by one. So guys,
deployment models are basically the way you can deploy your application on cloud
right. The various ways you can deploy your application on cloud are
basically called deployment models right. So that's a thing that you choose how do
you want your application to be deployed. How are they deployed? We'll discuss that
in a few moment. But first, let's discuss how is it different from service model?
So guys, service models are different like service models are how do you want
your service to be or how do you want the cloud provider to give you a
particular service? Do you want an infrastructure from the cloud provider? Do
you want a platform from cloud provider or do you want a software from a cloud
provider, right. Now, what does these three things mean. So, when you talk about
infrastructure as a service, you basically expect that you would be given a machine
to work on from the cloud provider that basically means you will get access to
the operating system. In case of platform as a service product, you get is a
basically a dashboard anything where you will not get access to the operating
system but you know you get access to a dashboard on which you can
upload your application and basically see whether your application is up and
running right. So a software installation everything is handled by the cloud
provider and the next thing is software as-a-service where in you do not get
access to the operating system. You do not get access to a dashboard where you can
upload your code but what we do get is a software that has made by the cloud
provider which you can use to cater to your needs. For example, there are various
CRM software out there which are hosted on cloud. For example, Zoho CRM, it's a cloud product right. Google Drive on which we used to basically use for solving your
storage needs is again a software-as-a-service. Similarly you can think of a lot of things which are basically software that
you use in your day-to-day life on which you cannot upload your code but you can
actually use that software. For example Facebook is also a software and it's
hosted on cloud so that also falls into software service. Instagram is a software
that you use that also is hosted on cloud that also falls in the
software service. Platform as a service is a little different basically because
you will be uploading code on it and that will be deployed on a server that
you would not be aware of right. So that's why it's a platform, it gives you
a platform to code and that's why it's a platform as a service and on the other
hand when you talk about infrastructure as a service like I said you get the
access to the operating system. Similarly, when we talk about deployment models,
public cloud is nothing but when you share your code or when you share your
machine with other people on the network as well right. More about
that I'll talk as you move along right. But let me first tell you the
differences between private, public and hybrid cloud. So private cloud is nothing
but when you are, when you do not want to share your machine with anybody else
right. You just want, you all go to exist on the machine and then you have something
called a hybrid cloud which is basically a mixture of both of them right. So let's
go ahead and discuss each of them in detail as we move along right. So let's
first talk about the deployment models. So the Diployment models like I said there
are public, private and hybrid.The first one is public cloud. So what basically
happens is that when you deploy a machine on cloud, when you deploy virtual
machine on cloud, it is not a separate machine which gets deployed on cloud. So
what basically happens is that each cloud provider, they buy machines which are humongous in size right. It could have around 30 petabytes
of storage and it can have around let's say 200 GB of RAM or 300 GB of RAM or
even more than that right. So each machine has this kind of specification.
Now what they do is, they install hypervisors on these machines and what
do hypervisors or virtualization software is do is that they launch
multiple machines on this very same system right. So, let's say you launch
five machines on Microsoft Azure. So that could mean that all your five machines
which are launched may be hosted on one particular server as well that's a
possibility right. So this is how cloud provides you with different kind of
specification machines. They basically buy one kind of machine, they buy all
their servers with one kind of specification. But according to your
needs, according to what kind of specification you want, they virtualize the machine from that very server and obviously they launch
more virtual machines also from that very server if the resources are free
right. This is how cloud works. Now what happens in case of public cloud is that
let's say you are launching a virtual machine right and you have launched it
let's say on Microsoft Azure, so if there is one more company let's say
you are Amazon okay and there's one more company let's say it's a Flipkart or
let's say it's Walmart. So if they come up to Microsoft Azure and they say that
you know what I also want to launch my website on your cloud platform it could
be that both of your websites are actually being hosted from the same
server right. Now this might sound okay to you because you believe in your cloud provider and you know they guarantee you security to your applications, they guarantee you
that no data of yours will be hacked from the other people who are also
sharing the same hardware as you. But some companies, they basically say or
they have some confidential data which they do not want to share with anyone
right. So we will come to that but if you are okay with this kind of first setup
where and you know your application could be on the same hardware on which
other applications are also there or other people also using it, then you term
it as public cloud because you are on the cloud which where all the public can
basically come up and they can launch their own servers and it could be that
you are existing on the same server as well but you don't care about that that
is public cloud. Next type of cloud is private cloud, so private cloud is for
those companies that I was talking about which basically want confidentiality and
they want security in the utmost manner that they can right. So let's talk about
some government agencies right. So because if some government agency have to
host their data on cloud and let's say they turn on to Microsoft Azure for that
matter, so they would say okay so we don't care about the money that if you
have to pay you right but we want security in the most highest regards
possible. So what Microsoft Azure tells them is you know what although we have
this system in place which is unhackable, it's technically unhackable, no other
person who is hosted on the same server as you would be able to get into your
website but if you are still concerned because you know in technology actually
there is nothing called as unhackable. Everything can be hacked right. It's just
Hardware right. If earlier people can like if you would have heard
about people jail-breaking iPhones and Apple every
time it comes up with a new phone it says it's the most secure device on the
planet and still people managed to hack it right. So same is the situation here
as well. Even though your cloud provider is guaranteeing you that you know your
hardware that you own and if there are other companies also on that hardware,
they will not be able to hack into your data still you know the companies they
do not want to take any chance so they offer private cloud. So I was taking the
example of government agencies which have very you know would volatile data
with them data which is very very confidential. So they don't take any risk,
they also opt for a private cloud. A private cloud basically means is that
they would give you one unit of their server for your use right. You use that
server as you own it, no other company will be hosted on that particular server
that the cloud provider guarantees you right. So that is termed as a private
cloud one other way that you can create the private cloud is then you can buy
your own servers right and you can set up set them up in our datacenter and
what I can do is you can install the hypervisor, the virtualization software
on top of it and you can use those servers as if you are using any other
cloud provider for that matter. Like if you use Microsoft Azure you can deploy
machines with various specifications similarly we'll be able to do that in
your data center as well but the scene is over there you have bought your own
servers right on the other side of the second type of private cloud that I
discussed with you is when your cloud provider that is a company like
Microsoft Azure they would say ok you know what I can provide you a server on
which I will not be putting any other company and you can use that server as
your own and is also termed as private cloud okay so
these are the two prong bodies first we discussed private cloud also a public
cloud and then we discussed private cloud the third type of cloud which
exists in terms of deployment in terms of how you can deploy your application
and cloud is called hybrid cloud now what is hybrid cloud as the name
suggests it's basically a mixture of public and private cloud now why would
you use a deployment model like this so one use case that can I can tell you
guys is let's say you know because I told you private clouds are expensive
let's say you have a distributed kind of an application on which storage is done
on one server the website hosting a let's say done on the other server so
what do you do is because your website is already facing the internet what you
do is on the private cloud you basically just store your files all
right and you know all the confidential data is stored on the private server on
the other hand the website is actually hosted on the public server because
anyways it has to be exposed on to the Internet you create a private connection
between your server that is your website server and your back-end storage which
is basically on private cloud and then they interact so it basically gives you
an advantage of saving some money because now your website exists on the
public cloud so if there's anything related to website anything related
scaling the ADA to website or anything for that matter
everything you will happen on the public cloud and if there is any extra space or
extra resources required on the private cloud you will actually have to set up a
new server which will be storing the extra data for you you do not have the
option of choosing the cheaper option which is public cloud in this case
because now private cloud is very expensive and
only that in beforehand then if I if you put your application on private floor
I'll give you one cool server for your own use and nobody else will be deployed
on it right so if Kaling and actions has to happen
but only your data will be stored on it and you do not or you cannot have a
control on how much specification do you need on that computer because obviously
you will be buying that own buying that computer also for yourself right on the
other hand if you deploy it on a using a cloud provider that is you buy a private
cloud or you host or you ask your cloud provider to basically give you a private
cloud on their own premises so they might give you a server which is
configurable but again that is more expensive than what is there on your
public cloud so that is how a hybrid implementation guys works moving forward
guys now we have discussed about the various deployment points the various
ways you can deploy your application on cloud now let's discuss the various
service models which exist on cloud right
so we already because there are three kind of services that can be offered to
you on cloud you might get an infrastructure service platform as a
service or a software service nowadays infrastructure and service like you
already know I have already explained the ways when you get access to the
operating system right of an application or off your server that is you can
configure or you can install anything on that server
I have fill joints you can install a database you can install web
applications you can install any back-end application whatever you want
in soil you can install and it's you also do whatever you want with it right
that is called infrastructure as a service on the other hand when you come
as to platform as a service in that case we do not give you an access to the
operating system in fact they do not give you the access to the machine as
well they just say we'll give you a dashboard you upload
your code over there and I will voice your application for you let me handle
all the operations let me install the required software needed for your code
let me do everything able to scaling let me do everything for you right so in
that kind of a scenario that kind of a service is called platform-as-a-service
okay and the third type of software in the service where we have already
discussed that when you get access to software rather than a platform or the
whole machine as a service you call it a software service and the examples were
Google drive's the examples were Facebook Instagram Gmail anything of
that sort origin you do not have the option of
uploading your code to host your application you basically can just use
that application that's termed as software as a service now moving on four
legs let's talk about who provides these kind of services right now
this was cloud in general right but today's session is about Microsoft Azure
so how does Microsoft Azure fit into this kind of a story if let us explore
that so that is the top three cloud providers in the whole world are
basically AWS and then you have Microsoft Azure and then you have Google
cloud so all these three companies they provide the cloud business model on
their own infrastructure right so every concept that we studied is provided by
these three cloud providers and let's let's put it out there that AWS is the
most used cloud provider in the world right it's the top cloud rider with the
largest market share but that does not mean that Microsoft Azure and Google
cloud a little nests operational or are not that great
as AWS it's just that AWS was launched way back in 2006 and that's why most of
the companies who got you here didn't you know okay so the undershoot cloud
and they said okay so it's actually going to save a lot of costs for us they
moved on to AWS and because there was no other cloud provider at that particular
time they are now still on AWS because now of course those companies it's
shifted at that time they are now completely dependent on the cloud
provider and if they'd like to migrate that will actually cost them a lot right
and that is right AWS has the largest market share reason
being it came here first and hence there are more and more people who are
actually hosting their applications from a The Bluest the second-largest cloud
provider is Microsoft Azure right so their Microsoft reserve was recently
lost I guess it was launched in 2012 right and since then it has grown a lot
right so it has thirteen or fourteen percent
of the whole market share right off of the whole cloud paradigm and it is
growing exponentially every year right while is turning about Microsoft Azure I
will give you a lot of facts you know the upcoming slides but what I want to
point out over here is the growth of Microsoft Azure is more then the growth
of AWS right so at the growth rate at which AWS is growing Microsoft Azure is
growing at a triple speed then the growth of AWS so that just shows that
people have started adopting Microsoft Azure people have started using
Microsoft Azure and that but that there is no points that you know people have
stopped using AWS that basically means no this shift that I see in the ministry
right now is everyone wants their application up and running they do not
want any discontinuation in processes which are running 24/7 in
their organization so for having that kind of availability
so basically Essen is when you deal with SAS SAS are basically service and
agreements so when companies they get in partnership with cloud providers and
they say you know you lose our cloud service for us they also sign service
level agreements where they cloud why this is you know this is the amount of
guarantee that I can give you that you don't have service would be up and
running so the this SFA's are usually 97 98 99 % it basically means your cloud
provider is guaranteeing 99% of your uptime throughout the lifecycle that you
will be on that cloud provider but you know companies they still do not like
the number 99 they feel you know why should I take a risk for that one
percent so what the do is they actually rely on other cloud providers as well
right so if let's say our application is hosted on Microsoft Azure and what you
want to do is here your Microsoft is always saying you know your application
is going to be up 99% of the times so what you do is you replicate your
architecture and you deploy it on a secondary cloud provider as well because
whenever your cloud provider will be down let's say your Microsoft Azure
cloud computing service goes down right so if that goes down you at least have a
backup which will serve as your main website again or your main application
again which is hosted on some other cloud provider and hence you know even
if Microsoft Azure is down your other cloud provider
does not get affected right and hence your business still continues so this is
a shift that I'm seeing in industry that people are not only on cloud but they
are now on multiple cloud providers as well so that you know the application
never go down so guys like I said the three top top riders are AWS Microsoft
Azure and Google cloud but now let's go ahead and talk about why are we learning
or Microsoft Azure today when I mentioned that you know AWS is the
largest role player in the market why am I not showing about either blue is why
am i showing about Microsoft Azure so if we go ahead and tell you guys why right
so the reason for that is that Microsoft Azure in terms of Microsoft products
like you have both know that at some point of your life in IT or at home you
would have used a Microsoft product that's how where they have integrated
oil or that's how well they've designed their product that each and every person
on this planet if they are into IT they must have used Microsoft products at one
point in their life right and Microsoft if the gives you a benefit in in terms
of using Microsoft products that it see is that if you are on cloud and if you
want to use any Microsoft product like Windows or SQL for that matter
AWS whatever price you the place gives as your can give you a price which is
five times cheaper than that price the reason for that is very simple that
Microsoft owns the licenses and ends a license which is basically being bought
by Amazon would be you know cost you to them and when they
give you that license on a shared basis that is on a timely basis obviously
they'll charge you more but when it comes to Microsoft they have those
charges in hand because it's a old license right so what they tell you is
that dude you know what whatever AWS is charging you for that particular
instance or a product for a software which is by Microsoft which is running
on that server I'm going to give you that same server in one fifth of the
price so this is the first advantage that you get with Microsoft together if
you are into Microsoft products if your application is using any kind of
Microsoft product you will find it one fifth of the price when compared to AWS
the second thing is that if you have Microsoft licenses obviously if you are
let's say in a business and you are using some Microsoft software you
obviously would have bought some licenses and when you want to go onto
cloud you know you do not want to pay for the licenses as well you just won't
pay for let's say the server that you are using why should you pay for the
operating system when you have actually paid for that operating system before
when you purchase it for your office laptop or your office server right so
what you can do is or what option Microsoft Azure gives you is the bring
your own license option so you bring your own license you get the license key
to them and they will not charge you anything for their licensing even the
1/5 cost that they were charging we will not charge you that if you have your own
license if you have your own Microsoft license and you bring that to the
Microsoft Azure cloud ok so that is that is second advantage that you get with
Microsoft Azure and the third advantage alligator's that more than 95 percent of
Fortune 500 companies are actually using Microsoft Azure now that could be either
in the case of a primary cloud provider or could be using Microsoft as well as
secondly cloud provider or it could be using some products of Microsoft Azure
but if we talk about the companies the fortune 500 companies 95% of them are
using Microsoft Azure right so if you get yourself certified in Microsoft
Azure you have a 95 percent probability of hitting a company if you go 200
companies you have a 95 percent chance that you will be actually landing up in
a company which will be using some kind of a Microsoft Azure service right so
that's a pretty huge number and that actually gives you a higher probability
of getting a job as well in these fortune 500 companies so these are the
three main advantages as to why Microsoft Azure is being preferred by
businesses for their work right and that's what companies have now started
to shift to Microsoft Azure as well and that is why it is seeing such a expert
shoot exponential growth rate when compared to AWS now now that you know
that you know why are we learning Microsoft Azure the reason for that is
obviously the chances of getting yourself a job in Microsoft Visio is not
right now let's move on and talk about what exactly Microsoft is journalist
right alright guys so before moving ahead just a quick update guys this is
not the complete SEO course if you want to complete as your training you can
actually head on to our website we have a course called as your master
certification training which will not only get you be certified in one as your
certification which is admin or architect or developer but all the three
right so if you're interested the post link is in the description box please go
ahead and click on that here and you will get the course details so without
wasting any time guys let's go ahead and continue with the session
so guys Microsoft Visio is basically a cloud service which is developed by
Microsoft right it is owned and managed by Microsoft and it offers you its
services in a pay-as-you-go model and it is a second-largest cloud
provider in the world right so this is what Microsoft is your s so whatever
I've taught you with regards to the concepts of cloud all of them are given
to you by Microsoft Azure right and all the advantages that will do all that you
get with Microsoft Azure and all you pay for that particular service is based on
the number of minutes that you're using its services for right that's how
amazing cloud is and that's how amazing even Microsoft Azure is right now let's
move forward and talk about how Microsoft Azure basically works right so
we have discussed enough about what cloud computing is what is Microsoft is
you're we are we all clear with that why we using it why we learning it today we
are all clear with that now let's get into the technical aspects of Microsoft
Azure and understand how does basically Microsoft Azure is structured in terms
of its architecture alright guys now let's go ahead and understand the as
your core architecture so Kay this is the core architecture for Azure as you
can see there are four ways of accessing your right these four components are
basically nothing but four different ways of accessing the azure resources
right so the as your resource manager is basically a mediator between the as your
resources and the external agents which can interact with the azure resources
now to interact with the azure resource manager you need these four ways now
what are these four ways the first way is the azure portal now what is here's
your portal the azure portal is nothing but the GUI website that you get
provided with right for example if you go to portal dot as you're calm once you
have signed up on a jar you can basically go to that portal and you can
deploy any kind of resource that you want right so you can control and manage
the resources from that interface so that interface is called the as your put
right the next thing is Azure PowerShell now what is the zero power shell so
there is a thing called Microsoft Publisher also that exists in a Windows
system so if we're using a Windows system you can just type in a start by
PowerShell and you would be able to see what PowerShell is so it is some native
Microsoft product now what Azure or what Microsoft's provide you
with is it also gives you an extension for making your PowerShell interact with
Microsoft Azure right and once you install that extension once its
installed on your system your PowerShell would be able to connect to Microsoft
Azure which will basically give you the ability of controlling or deploying
resources on Microsoft Azure using the command line so that's Azure PowerShell
similarly you have something called as as your CLI now PowerShell is like I
said is a product from Microsoft Azure CLI is basically done on the DOS prompt
all right so dos is all together a difference of
fear which has been there since ages on Windows right so if you want if you are
more inclined towards using the DOS prompt you can actually install the
necessary software for Azure and then you can use the DOS commands to use or
manage or deploy as your resources so this is a third way of accessing your
reservation sources and then you have something called as rest lines now what
a rest lines rest lines are nothing but they are api's which you can include in
your web applications right for example you can create our own web application
through which you can create or use your as your account for example let's say
you create a website wherein you know you take the user input you take an
integer number from the user and then you have created a button which is
called deploy and what that does is whatever number you have mentioned it
basically deploys that many number of servers on Microsoft Azure right but you
are not basically doing that directly it's from the azure portal oh you're not
doing it that so the PowerShell or you're not doing that through as your
CLI what you have done is you have created your own application and you
have linked your application to your Azure account and you're making or
you're making that website or you're allowing that website basically to
control your as your resources and this is
possible using ApS which are also called rest lines right now these resources in
case of Azure PowerShell an azure CLI they would interact with the help of
SDKs to connect to as your resource manager your Azure portal any red
screens can directly interact with the azure resource manager and what is it as
the resource manager like I previously stated it's nothing but a mediator so
basically what happens is all the resources which are deployed on Azure
they're they are not directly accessible by the user even if he is using the
website or using the PowerShell or using the CLI or using the rest lines every
request has to go to the resource manager and what the azure resource
manager does is it basically authenticates your request it will check
whether you as a user of a Microsoft Azure account have the necessary
permissions to do a certain tasks which are trying to do with your application
for example when you create a Microsoft Azure account you will have the option
of creating multiple users in that same account so that multiple people can work
on your own account for example let's say you start a startup or you open a
startup right and now you have higher detecting now what do you want to do is
you want your CTO or the person who's going to be managing your tech team the
admin access of here's your account and on the other hand the developers who
probably let's say if one of the developers is going to work on the Azure
storage right so you won't just want him to access Azure storage you don't want
them to access as your VMs or any other kind of as your service so what you can
do is you can create a user and you can give definite permissions as to what he
can access and what he cannot exist okay now how will this be helpful now let's
say your technical manager he signs in as a user right and he signs in let's
say through the azure portal and then he tries to deploy a VM
resource or was a machine or he tries to use his toilet good you would be able to
do so why because he has the administrator access you have given him
and he'd be able to do everything that you can do on the other hand let's say
now your developer comes in he also logs in through the azure portal he also sees
everything which is there right but the moment he tries to deploy a virtual
machine because he has to go through all his requests would be going through the
azure resource manager when the azure resource manager authenticates this guy
and sees whether he has the permissions to launch over the machine or not the
basically the authentication fields so the Authenticator will basically say no
this guy does not have that permission so as your resource manager will not
take a request for word it will basically revert you back with the error
message saying you do not have the sufficient permissions to carry out this
task right and this is the sole reason we need a mediator between our as your
resources and the and the various ways through which we can use as your now
this would be in the same case with Azure PowerShell this would be the same
case with the zero CLI when he uses your PowerShell with your as your account you
obviously have to enter your credentials in case of a developer he will be
entering his credentials that you have given to him right and those basically
just have the access to the storage and again you would also get the same kind
of message in Azure PowerShell similarly you will get the same kind of messaging
as a CLI and even in the rest lines right so when you connect your website
through your Azure account again you will get a key and a secret access key
right or in other case or what you will have to do is you'll have to include
some metadata on your our server on which you are hosting that website and
that metadata would basically authenticate you to the user account so
for every user you get a different kind of credentials for your metadata in case
of a developer you will get a different set of credentials and those credentials
would basically when they'll be authenticated they would
get or they will inherit the same properties or the same permissions that
the developer account has and accordingly your as your resource
manager will authenticate and check whether you do you do you have the
required permissions or not right and accordingly you would be able to do the
operations so that is summing up or just revisiting what we just studied so there
are basically four ways to access as your you have as your portal as your
partial as a CLI and you have less lines right all of these they interact through
the user resource manager and there is your resource manager based on what kind
of permissions you have it basically routes your request to the necessary
service and carries out the tasks for you
right so this is a gist of how as your works in the backend now let's move
forward and talk about the front-end the middleware and the services right so the
front-end as we discussed there were four ways of accessing as your services
which were this is the azure portal this is your PowerShell this is your zero CLI
and then you have the rest lines the middleware was nothing but it was you as
your resource manager it be discussing that and then you have the services that
we'll be discussing in a few moments right so let's start off with the
front-end part of the front-end components so what is your portal so
guys the azure portal looks something like this right it's a GUI component
that you get on your web browser and on your web browser you can basically you
can see the on the left hand side you have things as all resources you have
web apps you have SQL databases you can just select the service that you want
and you can subsequently launched it will be discussing more on this as you
move along I'd be showing you guys a portal I'd be showing you guys how you
can use it to deploy resources for now is the Sun designing what the other
portal looks like and this is what it looks like right on the right hand side
you get all kind of widgets so these are visits these are called widgets which
show different enough information right so you have the option of including
these visits on a dashboard more on this we will talk as we move along in the
session so as this was the azure portal this is the Azure PowerShell so I told
you guys Microsoft has launched its own command
line which it now comes with Windows integrated which is called PowerShell if
you want to use that PowerShell with your Azure account what is after ways
you'll have to download some libraries which will help your PowerShell to
basically connect to ensure and once you do that you would basically see
something like this Windows Azure PowerShell and now you can enter the
post PowerShell commands which would basically help you to connect to your
account and carry out whatever you want in a command line fashion right
similarly if you are into DOS you can actually go ahead and use as your CLI so
dos commands are different from PowerShell and there could be a scenario
wherein you're better at DOS commands than your PowerShell commands and if
that is the case you can actually go ahead and just install the libraries for
us your CLI and then you can actually use it in the DOS environment right in
this case also if you want to manage resources like you were managing it
through Power Cells through command-line statements you can do a similar kind of
way in as your CLI as well but the commands that you would be entering over
here they'd be more in line with DOS commands on the other hand when we talk
about PowerShell we would be using PowerShell commands to control as you're
in that case okay so this was a saucy Li guys in the nest and the last topic is
rest clients so rest lines as I already told these are nothing but API is that
you use on your website so you'll have to authenticate them using the metadata
or credentials and these credentials would be available for all the accounts
that are created under root Microsoft Azure account
so each credential would be different from each other
based on if there are more users for example in our case we created a user
for our IT manager and then we created a user for a software developer for a
software developer we just giving to storage access for our IT manager we
basically gave him the administrator access so when they put their
credentials down the code of their web application and they try to
connect to azad based on the permissions at their they would be able to access
the resources in case they do not have the permission you will get a
corresponding error to that thing this particular user does not have the
necessary permissions to access that particular racial or against so these
were the front-end components of the azure core architecture now let's talk
about the middleware the component which basically acts as a mediator between
your touch points that is your front-end components and the core components which
is the azure resources so it's acts as a mediator the as your resource manager so
as your resource manager it basically helps you to deploy and manage the azure
resources right it also helps you to organize resources in a cell for example
it will help you in grouping resources together let's say I'm launching a web
application okay and I'm also at the same time launching some other
applications let's say I'm also launching my android app back-end
service so what I can do is let's say if I'm launching a web application I would
need a certain components I would need let's say I need a separate storage from
a web application right so I can create a storage account I would create a web
app and then I would create a database so now if I create it just like that in
Azure and you try to see all the resources you would not be able to make
out as to which resources are there for my Android application and which
resources are there for my web application what you can do is probably
you can define nomenclature and then you can say probably you can name your
instances like let's say for Android database you can say Android underscore
database for your web application you can save web app and the score database
so that's one way of sorting it out but as your resource manager makes it easy
for you how does it make it easy it basically gives you the ability to
create groups right so if you are creating one kind of an application what
you can do is you can create a group and inside that group you can basically just
map the resources or launch the resources that are specifically there
for that particular application for example for a web
application I can launch all the resources that I want and I can group
them together in the web application group right I can name the group
anything it's just like a folder guys if you were to understand it more simply
understand it's just for sorting out resources let's say you have a deck
store full of files and those files the PDF txt and Doc documents so what you
can do is you can just create three folders and inside those three folders
you can put your dog folders in the inside the PDF folder you can put PDF
files and set the text files you can put the text files right so this is the way
of sorting on a desktop similarly if you want to sort your resources in Azure you
can use the ability of creating resource groups using the as your resource
manager right and at the same time the third point or the third feature of as a
resource manager is it or ten decades it checks what all permissions you have and
based on that it gives you the ability to control your resources right
so it also authenticates your calls to the azure resources and only through the
as your resource manager would your calls go and if they are accepted by the
user resource manager only then they will be forwarded otherwise they would
be rejected and you would get an appropriate message for that all right
so there this was all about the azure resource manager so I was talking to you
about what are the source Krupa's so as you can see let's say this is a resource
group of mine and what I want to do is I want to group all these three resources
so my VM instance that I launched that is termed as a resource if I launch a
web app that is also termed as the resource don't worry if you don't
understand what is the virtual machine what is the web app I will talk more
about this as we move along in the session but for now understand that any
service in Azure any instance that you launch in Israel that is basically a
resource right and when you group one or more resources together or I should say
two or more resources together for a particular thing whichever so I give you
the null energy of folders if you group them together and that symbolizes
something that basically can be done using resource groups and that is
exactly what is specified in the figure as well right so you have some virtual
machines you have app servers and you have SQL database and you have all
grouped them together in a resource group and probably you can call it let's
say the production environment you can create a test environment call the
resource group a test environment and deploy the necessary resources inside
that as well all right guys so before moving ahead just a quick update guys
this is not the complete as your course if you want to complete as your training
you can actually head on to our website we have a course called as your master
certification training which will not only get you be certified in one as your
certification which is admin or architect or developer but all the three
right so if you're interested the post link is in the description box please go
ahead and click on that here and you will get the course details so without
wasting any time guys let's go ahead and continue with the session so guys with
this we come to an end of because your core architecture so by now you
understand the nitty gritties of how Azure works in the backend right how
things are structured insiders are moving forward guys now we will be
talking about the azure services right so we'll be talking about the resources
that we just saw the resources that we were grouping inside the resource flow
or the resources which basically there's your resource manager was managing we're
going to talk about those and those are nothing but services in a zero right so
it will become more clear as we move along so let's just start off now with
core is your services okay so for starting of the services guys there's
again simpler way to understand all the services in is your what does your has
done is let's there are around 300 plus services in us all right now it is
difficult to actually remember all those services or to basically know how one
service is different from the other so what does your has done is it has
actually divided its services based on what the services does or what the
services do right so the major sections that it has divided its services in are
DS right so these are not all the sex that Azure gives us the services in but
these are all the important sections or the important domains in which as your
provide you services and as a cloud engineer or as a as your engineer when
you'll be working in companies mostly you would be working on services which
are included in these domains right we'd be covering these guys there is no much
your engineer out there who would be knowing each and every service in a job
right but what do we as learners can do is we can learn all the important
services which are basically used in your everyday life when you become a
cloud engineer right and as and when you apply to companies probably you can tell
them you know all the basic services all the important services on Azure and if
there is any special requirement for a particular service that they use it is
obviously going to be an easier task for you since you understand how is your
works so understanding one more service in Azure would not be a great deal but
most of the times I would say 90% of the things that you'd be doing as a cloud
engineer or 90% of the services that you would be using or working on as a cloud
engineer would be covered under these terms so now let's go ahead and start
off with the first domain which is compute but before that let me give you
a brief about all of these domains right so the first domain is compute so in
compute you have compute intensive resources wherein you get the raw
processing part more on this we will talk about as we move along then the
second kind of domain is networking so this particular domain would include all
the services which provide you with networking capabilities right then you
have the storage domain which basically gives you all the services in Azure
which can basically give you the capability of storing some files right
and obviously there are a lot of services and stories and each service is
targeted at a different kind of use case so we're going to discuss some of the
storage services in Azure which are very prominent then you have services related
to database and analytics right so if you want to store actual data which also
you want to analyze using graphs or using flowcharts right you can do that
energy or in the domain data please plus analytics then our next domain comes out
to be AI and machine learning now this is a very important to being honest and
I've seen actually people who are data scientists people who are researching on
data science so basically when you're working on
designs or when you're working with AI you need a lot of computing power right
now the good thing about cloud is that it
offers you a pay-as-you-go model which basically means you only pay for the
time you use the resources for right and nobody can afford you know a high spec
machine if you need an i7 and octa-core processor an hour 32gb ram it's very
difficult for a layman to get hold of such kind of a machine so what
researchers and what data scientists can do is they can actually launch a machine
on a jar with a similar spec that they want for their use case use it for the
time they want and shut it off and you'll be charged as low as around 0.05
or 0-6 dollars I guess for if you use it for half an hour or so right so I guess
half an hour of qualifying minutes should be enough if you want to do a POC
or if you even do resourcing or if you're into testing I guess this would
be the cheapest option for you rather than getting a full-blown server or
full-blown laptop with high specs and all that'll be more expensive for use
it's better if you use services from cloud for your compute intensive needs
but that is not it what you get in a and machine learning domain is you get a you
know you get a pre-built dashboard kind of a thing so you don't have to feed in
algorithms everything is pre-built in Azure you
basically get a drag-and-drop kind of a push functionality when you are
particularly using Azure ml because I've used that I am telling you it's pretty
drag-and-drop if you compare it with its counterparts let's say if you want to do
a regression test in our you would actually have to write a long script to
do a particular thing but the the kind of services that azure has launched for
example like I mentioned as your ml all you have to do is drag and drop and you
can get your right so that's how convenient it is and
that's why people I'm more and more to save their time to save on their costs
they're actually using the AI and machine learning capabilities of the
cloud and particular we are talking about a gorge so as your also has a and
machine learning services which also we will be discussing as we move along in
the session right then the next domain is identity so identity to me domain
would basically include services which will help you in authenticating users it
will help you to basically get the metadata credentials that we discussed
earlier for your website so all the authentication part the authorization
part whenever you want to give specific permissions to particular users all of
that can be managed with the identity services and lastly we would be
discussing the domain which is management which would basically include
services such as monitoring it would include services such as infrastructure
as code right don't worry about these big words guys as we move along
everything will be clear but without wasting time let's jump on to our first
domain which is compute and let us understand what are the different
services that were gonna understanding that right so in computers you basically
have four prominent services you have virtual machines you have function apps
you have app service and then you have as your cube innately service
so the first service that we're gonna discuss about is the Azure virtual
machine service now what is that nobody's as your virtual machines and
nothing but it's it's basically a server or I can say the most layman terms
possible let's just say it's a computer just like a laptop which has just
installed windows on it and nothing else is there right for example let's say you
buy a laptop what is the first thing that you see you see your windows there
or you see a Mac OS there or you see your Linux operating system there and
basically no no no software is all it's just a bare minimum operating system
which is given to you and whatever you want to do on your laptop you install it
and then you work all right so similarly virtual machines are their virtual
machines are nothing but the machines that are launched by your
cloud provider which in our case is your and they launched it for you so you
don't have to worry about very allocating it from they launch it for
you and what they do is they give you the remote access to it right so it's
basically like working on your own laptop but working on it from a remote
location so similarly you get your own machine on Azure which is freshly made
right you have a fresh operating system on it there are no software is installed
and whatever you want you can install on this particular server so we call it
server because basically you'd be using it for deploying applications which
would be basically be available to the user so that's why require a server
right so basically your server all computers that are launched by us your
for your own personal use but you would be accessing it remotely if it's a
Windows machine in a GUI case you can basically do an RDP and use it if it's a
Linux machine you can do basically do an SSH and you can work on the command line
for it but remember there are no software installer and there are no
extra software installed on it so anything that you want this over to do
for example let's say you want to create a database server you can ingest install
database on it and we'll be ready right and it will be ready for accepting
requests for the database service if you want it to be a web server you can
basically install Apache PHP etc and then you can put your website on it
it'll become a web so anything and everything that you want or that you can
do on your laptop you can do on virtual machines so this is what as your virtual
machines is guys now the next kind of service is similar to virtual machines
but it's actually an advanced version of virtual machine so what our function
apps so those function apps is like I said it's an advanced version of virtual
machine in which you do not get the access of the operating system okay so I
let me give you an example for example let's say you are let's say doing some
work on a website and that website basically let's say it does processing
for you for example let's say you are on Facebook and
what you do is you upload a profile picture and once you upload a profile
picture let's say you want to drop that profile picture to a particular size and
then you want to upload it as a image on your Facebook so what will happen so you
will upload your image on the Facebook app you will do the cropping part let's
say you want to add a filter as well you add that filter and then you click on
save and then you want it to be uploaded as a profile picture on your Facebook
right so there is a minimum amount of processing that would have to be done on
your image some processing has to be done on your image right and basically
how that works is that processing is actually not done on your phone or on
your computer on which you're accessing Facebook that is actually done on the
Facebook servers but the way we handle it is they basically would have a web
server in place they would separately have a database server in place which
would be interacting with your web server and for the processing they have
a separate server ok and that separate server is basically called up back in
server and what is that back-end server do all the requests which comes in it
basically processes it and gives out the result that's the sole job of a back-end
so and why is it separate why is it not integrated with the web server the
simple reason for that is let's say there are a hundred users who are using
your website right and they are constantly exploring a website going
through the UI doing uploading images doing processing and everything now your
back-end component is doing the image processing and the front-end component
of yours which is the website component of yours is basically serving the
website to the users right so there are two tasks happening over there and let's
say any one of them they get overloaded with some work right
that would hamper the performance of the second component let's say if there are
more users on the website right so obviously your web component or the
website component of your server would require more processing power so let's
say it takes up 60% of the processing and let's see what happens is all these
people who came in they upload their profile pictures and they put some
filter on and they now want to process it so one
is your server is already busy serving these hundred people your website and
the CPU mark is at 60% now because of the processing load which is coming on
to your back-end component that also requires around 80% CPU to work
efficiently but now because 60% has already been taken by your website
server there is only 40% left so what will happen in this case your
server will inadvertently become slow right it'll become slow and it it might
even crash in case a deadlock situation happens right and this was usually the
case when we even not aware of distributed computing but nowadays what
happens is we are dealing in a world where we cannot afford downtime and that
is the reason everything has been distributed right so if there is more
load on my back end server my back-end server is going to handle it it will not
hamper the performance of my website server or my database server so every
software component is now distributed and why are we discussing this we are
discussing this because we are talking about function app now what is the
function app function app is nothing but a back-end service on which you do not
have to deal with the operating system like you do in virtual machines you do
not get an access to the operating system you cannot install any kind of
software on it whatever you choose let's say you have an option of choosing what
kind of back-end code you want to run on the function app let's see your image
processing whatever code you have written for the image processing app
lets you have written it in Python so when you will be deploying a function
app what will be choosing is what kind of code would I be uploading on the
function app let's say you choose Python the next thing that you would be doing
is you'll be clicking on next and then it will make your function app ready and
then it'll ask you for the code for the image processing you given the code you
click on save and now you're back in server is ready to accept requests
that's it so now all your website server has to do is bring the back-end server
with the required information with the image it will basically take the image
it will run through the code see what it has to be done do the processing on the
image and give the image back the websites over or to the Facebook
servers which basically upload the profile picture right
so energy status function app a function app is basically a service which will
not give you access to the operating system it will basically give you one
dashboard on which you can upload code and it can do any task you all you have
to do is give it to work it will do for you and that's about it you don't have
mess around with anything else you don't have to worry about what machine says I
should give to my function app you don't have to worry about how many machine
should I run so that my function app never gets overloaded everything is
managed by us you're all you have to do is given the code and see your
application working so these kind of applications are basically called
platform as a service and the earlier service as we discussed was virtual
machine is basically called infrastructure as a service why
infrastructure as a service because in that case you got access to the
operating system right you can do anything on that server you can stall
anything you can actually delete everything also and you can make that
server hang or you can make that server not working you can actually uninstall
the installation files or the operating system as well that is something that
you can do with your server but what will happen is it will not cause any
harm to us you're basically you would not be able to use this over in in the
far end time you'll basically have to delete it because it won't be usable
that's the most extreme thing which could happen but the case with function
app is because it's a platform as of which it's giving you a dashboard on
which you can upload code and hence it's a platform right it is not giving you
access to the operating system it is not giving you access to what software you
can install you just choose the environment in which your poor can run
and that's it that's what a function app is all about it's a back-end server
without choosing or worrying about the underlying infrastructure moving forward
the next service that we have is app service app service is yet another
platform as a service kind of resource that we have in Azure and what you can
do with app services you can basically launch or deploy websites now you would
be wondering that in function app also I can just give my code and probably I can
give my website code and deploy the website for me know your function
I can only give outputs based on the inputs it cannot deploy a web
application for you if you want to deploy a web application you'd have to
use app service in app service you would find a resource called web app we'll
have to deploy that and on that you will basically get again a dashwood since
it's a platform SSO which you will again get a dashboard where you will be able
to upload your website files and once you do that once the process is finished
you will basically get a link and if you click on that link you can see your
website right so again we did not get access to the operating system we do not
have control on what softwares are installed the basic way the app service
works is or in our case with since we are discussing board websites our web
app works is it will ask you what kind of code is your website written in let's
say my website is written in node.js so I'll just say I'll just select no js'
and I'll click on next and the next question will ask is do you want to auto
scale when the CPU increases or when the memory is low do you wanna also Oracle
you will say yes I'm gonna scale you click on next and it does everything for
you and that what at the end what it get is again a web UI on which you will have
a button called upload on which you just have to upload your website and
everything would be set for you right there is also an option to just directly
mention the github link it will pull the code from github and deploy whatever
code was there on github but the key word here is it deploys application for
you in function app so this is a very tricky thing that what is the difference
between function app and a web app so function app only does the backend tasks
for you you give it code it had taken input it will give you the output that's
it it is not used to deploy an application app service on the other
hand it is only used to deploy an application okay so this is the main
difference between a function app and an app service a next service is probably
the very important service when you're working in a containerization
environment so I'm sure most of you know outward talker is if you don't know just
type on YouTube docker tutorial by Intel apart go through that video and you
would be able to understand you know what docker is so docker is nothing but
a containerization platform on which you can deploy application so we were
talking about distributed computing right so containers are nothing but they
act as separate virtual entities which are isolated from each other
so I can launch an ubuntu container I can launch a center waste container I
can launch different flavor of Linux container as well right in them I can
install any software I want then exactly whatever code files I want I can put on
those containers and these containers can then interact and may make a
distributed application kind of a scenario right now what Kuban it is is
it manages these containers for you it's an automatic service which manages these
containers what what does it manage let's say you know I deploy three
containers one is my website container one is my back in container and one is
that database container now for some reason my back in container is not
working okay it stops working now what will happen
what should ideally happen ideally my if my back-end server stops working I need
to get an alert I will go there I will see what the problem is I will fix it
and then my back in container or back in server will again be ready and my
application would serve normal but what happens is you cannot monitor
application 24 by 7 so what Kuban ities does for you is it does all the manual
tasks for you so it automatically detects that a fault has occurred in a
particular container and what it does is it deletes that container and launches a
new copy of it automatically right and this is just one of the tasks that Kuban
ities does automatically for you you can also configure communities to scale your
containers for you or descale your containers for you and it can do a host
of other things and we would disperse it probably when we move along and we just
when we just are focusing on Azure communities for servers we can actually
talk more about it because there are loads of feature in it but this is a
a very important service so most of the startup companies not not MNCs but most
of the startup companies have now adopted the container architecture or
have now made their code in to fit into containers and now they are using the
community service if I were to give you a little background of the humility
service the Kuban ID service was actually developed by the Google company
later they made it open-source and now it's available to the world anybody can
use it or download for free and can actually install it on their system
right but when you install it on your system you actually have to manage Kuban
it all by yourself so what does your does is it has created platform as a
service again and it says I will handle the Cuban ities installation I will
handle the Cuban IDs configuration you just tell me what you want and I will
deploy that in a Cuban ities cluster for you right so all the advantages of Cuban
ities you get and you also don't have to deal with Cuban at ease so this is the
power of cloud this is what Azure is giving you as a service right and this
is what as your Cuban 80 service was all about all right guys so now let's get
down to the interesting part let's now start off with a hands-on where you will
get an taste of how as basically as your portal looks like and also how things
work out how you deploy resources on Azure and basically in the most essence
we learned about what a VM is so we're going to see how we can actually launch
a VM inside Microsoft Azure alright so just give me a minute I'll just switch
on to my portal and then we will start on with the demo now before moving over
there guys these are the three things that we're gonna do so when we were
learning the architecture I told you guys that basically all the resources
that are deployed on an azure can be managed or they can be grouped together
using resource groups so the first thing that we're gonna do is we're gonna
create a resource group which will have the name demo environment okay then what
we're gonna do is in this demo environment basically we are going to
launch a virtual machine which will have a Linux operating system on it now once
it is deployed we will to connect to this virtual machine using
the putty tool that is available for doing SSH on Linux so basically when you
whenever you're installing or whenever you're launching a Linux Linux instance
a way to connect to a Linux instance is using SSH right because it's a command
line operating system and all everything in anything that you want to do on this
operating system you can do it on the command line on the contrary if you
deployed a Windows machine in that case you have to use the remote desktop
protocol so it's basically called RDP and there's a software for that which
can basically connect using RDP so that is in built-in Windows so if you are on
a Windows machine just type on type remote desktop right and you'll get that
application the only thing it'll ask is the IP address click on connect and then
we'll ask for the password all right guys so without wasting any more time
let's go ahead switch to my Azure portal and let's see how it looks like right so
guys this is my zero portal so let me just go to home and let me show you how
it actually looks like so this is how your portal looks like guys this is the
web's UI that we were talking about now there are some resources that are
already launched in my Microsoft Azure right so as you can see I have named my
resources something like this but let's say I am going to deploy a Linux VM run
right that latex VM is only for demo purposes now this will actually give you
a take as to how you can or how these source groups basically can help you
guys now if I go to all resources right now you can see these are all the
resources which are deployed on Azure right now but I do not know which
resource is doing what or which resources contributing to what
application on the contrary if I go to resource groups you can see that there
is a group which is called Intel apart which exists on my Microsoft Azure
portal and if I go inside this resource group I would be able to see all the
resources which have been deployed so now what I'm going to do is because what
I am doing right now is just for demo purpose so what I'm going to do
I'm gonna add a resource group so let me just click on add and let us create a
resource group by the name demo environment so I will type in demo - env
right and let's say the region that we have to choose always choose a region
guys which is nearer to you so what do you mean by region so region basically
means now Microsoft Azure is so huge that it has deployed or it has set up
its data centers in different countries the reason they have done so is because
let's say you have your application and you have a start-up and you have
actually started it up in let's say India right and your application is
actually only accessed by the Indian audience right so that's whom you cater
to so what you can do is rather than deploying so it makes more sense to
deploy your application or your website on in a region which is nearer to your
audience because in that case it's the internet route it'll be far lesser when
you compare it with launching your application in some other country for
example let's say we launched this resource of a create this resource group
in let's say us right but my target audience is basically in India now what
is gonna happen now when my target orders audience is in India and let's
say I'm going to the Intel epic on website that is my application every
time the data is being fetched from the US and it is being transferred to the
Indian servers and then from your ISP you are getting the data on your system
now if we were to make or if you were to deploy our application in India itself
what will happen is that the distance that the data has to be covered eatable
covered will be less and that will significantly increase your response
time of the website right so that is why choosing a region is very important you
can choose any region you want but I would suggest you choose a region which
is more nearer to you right so in our case let's scroll down and see if we
have any region which are nearer to me so I can see that we have a region in
India is called South India so let us select that right all right so now the
that my resource is getting deployed is in South India right so that's what I
have selected the name that I have given is demo environment and now let's go
ahead and click on review plus right so this is what is gonna happen the
subscription is going to be pay-as-you-go if you are on the free
tier you will see if retail the resource group this is the name that I have given
and the region on which this is being deployed is South India let's click on
create so now my resource group is being created right and if I do a little
refresh over here I can see that my demo environment resource group is present
here now what I want to do the next step that I have to do let me just go back to
my slides is to deploy a Linux VM in this demo environment resource group so
let's do that let us come back to our job portal and
now what we want to do is we want to deploy a virtual machine so you will see
on the left hand side that there is a service called watch of machines let's
select that and now what I have to do is I have to click on add right so because
I want to create a virtual machine so I click on and now it will ask you which
resource group do you want to make this server available in so I want to
basically make it available in demo environment so this is what I have
selected what do I want my virtual machine machine name to be let's say I
wanna know what's a machine name to be new VM - Linux okay this is what the
name is the region I wanted to be nearer to me so I select South in there right
availability ions options if you go ahead and choose it you will have an
option between availability set or no infrastructure redundancy required so
what is redundancy redundancy basically means do you want a copy of your server
somewhere else as well so that let's say in the data center where this server is
going to be deployed that datacenter goes down right or something happens in
the data gets corrupted so if you choose that you know you want to replicate your
server what will happen is even if there is a downtime in a particular data
center in which your VM was deployed your redundant server will come
to power and that is from where you will get your data so basically it increases
the availability of your application and even if there is a problem from the
other side your application will not go down but since we are doing a demo what
we will select is no infrastructure redundancy required what is the image
that you want so by image we mean the operating system which was what kind of
operating system do you want to connect to right so you have all these operating
systems over here I can actually launch the windows 10 pro I can launch a
Windows Server I can draw on Ubuntu 69 launch the Bay 9 Santo is Susie Linux
RedHat Ubuntu server right so for our sake let's consider over to server is
what we want to deploy and now it will ask me what is the size of the machine
that I wanna give right so what I can do is you can actually change the size so
by default it will say you should deploy a to CPU and ia GB machine you can just
click on change size and you will have all these options available to you right
so you can see that because what we are doing going to do is a demo so what you
can see over here is 0.5 GB RAM and a 1 CPU will basically cost me around 383 to
Pease per month right so let us go ahead and try to deploy this machine right so
this makes more sense because we are only doing a time and probably deleted
after this session right so I've selected this and I click on select now
so you can see it's 1 V CPU and point 5 GB memory great next thing it is asking
me is what is the authentication type that you want to give on this particular
machine you can either choose the SSH public key or what you can choose is
password right so if you want to choose SSH public key what you'll have to do is
you'll have to generate SSH public key right if you choose password what it
will ask us the username and the password if you and this is basically
the most standard way of doing it but if you want to give security to your
instance the suggested method would be to create a public key now how can you
create a pub there is a software called put ijen
right and the way you can download it is just type on Google put the download
right and then you will get a link from put T dot org just click on that and you
can say you will see this link which says download putty just click over here
right now you will get all the versions on which you can basic using which you
can basically installed putty now what we're basically looking for is put each
n right so you will find it over here based on your CPU architecture you can
basically select put each n and then you can download once put each n is
downloaded it will look something like this so as you can see it's putting each
and the later what I want to do is I want to generate
a public private key so let's click on generate now what will Alaska is you'll
have to hover your mouse here for some randomness and it will create a key
based on that so let's hover our mouse in this region and now it has created
the key for me now it says the public key for pasting into OpenSSH authorized
if I so this is the public key basically right and you will have to copy this
let's copy it right so this is my public key now let's
come back to my portal so I can just choose SSS public key and I can paste
the public key over here and this is how it works now if your key is verified
what you will get is a tick mark over here which basically means that a key
has been verified and now you can go ahead right so my key is now verified
what that is the next step that I'll have to do the next step that I have to
do is I'll have to save the private key okay so this is the public key which has
been generated what I have to do is I have to save the private key which I'll
be using to connect to my instance so in this case you basically do not have to
use you know any password if you have this particular file which will
basically be saved once I click on yes let's say I save it on the desktop or
let me actually save it somewhere else let me save it in the C Drive and let me
actually create let's say in the app folder I'm just creating this private
key file in its name it has as your - key okay and now I'll click on save so
my PPK has now been saved if I want go there I can just go to C I'll go to app
and I can see that as your - key PPK is now present over here great so this is
done now I have to specify the username so let's specify the username as Azure
you can specify any username that you want right so we are specified it has
sure now the next thing that it is asking is do you want any inbound ports
to be enabled on this particular instance now we will be connecting to
this instance from our machine right and the protocol that we'll be using is sss
so we'll have to open the SSH port so we'll have to click on allow selected
ports and what port do I want to allow I want to allow SSH port okay
let's select HTTP also why I'll tell you that in a few moments once you have the
Linux machine up and running right in case this was a Windows machine I will
also have to enable the RDP port which you
can just take over here in my case I don't need the RDP port
I just need SSH and I need HTTP do they use for it I'll tell you as we move
along all right so I won't SSH an HTTP to be enabled and
what I can now do is I can just click on review plus create and now it should
basically just show me all the configuration for my VM which is going
by default try it I can just review it and that will be it right now I will
show you where is something very interesting over here let's um once this
loading is complete so as you know cloud computing it doesn't charge you per
month but basically the pricing model is a per our kind of a thing right so as
you can see the machine that I'm launching right now is right now 50
paise upper this is the rate or this is basically the pricing of my machine that
I'm gonna launch so that's awesome right the next thing is the OS which is being
launched as Ubuntu server 18.04 that is also great and if you want you can
actually look at the username which is a bun to the soy as your what are the
public inbound puts it's SSH and HTTP though this type is SSD is it a managed
this yes the network it's basically creating a new network for me which is
which goes by the same resource group name which is demo - env so what is what
happen is guys when I launched this particular VM a lot of things will be
created it will create disks it will create networks and everything related
to it will actually go inside that particular resource group right so I
will show you how to do that just click on create once you feel everything is
right and now my machine is basically getting created ok so let's wait for a
few while if it will take like a minute or so to deploy a machine so let's wait
for that time and let's hope this deploys soon so as you can see it says
your deployment is underway once it's complete you will see a different
message over here so let's wait for that message to appear alright guys so my deployment is
completely as you can see the message is being told over here so the next thing
that I'm gonna do is I'm gonna go to resource groups and now I can see that I
have something called as demo - env I'll go inside that and over here I have all
the related resources to my VM ok so guys as you can see my virtual machine
is this particular resource so let me go inside that and now what I have to do is
I'll have to select this IP address this is the IP address on which the instance
is available so let's copy this IP address and now let's open the put T
tool now what is the potato we used put ijen before to create the
key now for connecting to the instance you will have to use the putty tool now
how do you download the putty tool again back to your tab where you opened the
download web page now over here you will have to select put T dot exe it says the
SSH and telnet client itself you'll have to download this according to your
architecture and then you will have a tool which looks something like this
now what I want to do is I want to connect to this IP address which I've
just copied from the azure portal and I will have to go inside SSH and I'll have
to coincide auth and now I will have to select the PPK
which will basically be used to connect to my machine which is as your - key
let's click on open and now everything is ready guys so I have selected the PPK
I have inserted the IP address and now let's click on open so it will ask so
whenever you connect to a new IP address it will give you this morning
don't worry about it just click on yes and now it is asking what do you want to
login and so remember the username that we gave was assured so let's select that
and hit enter and that's it that's about it guys so
now you're inside your Linux machine which is the Ubuntu machine and now you
can do anything with it as you feel necessary right now you will ask me one
thing that you know we deployed we allowed HTTP connection why did I do
that just to show you show you one very awesome thing now since this is a bun to
the commands that I'll be entering now is basically for Ubuntu and if you want
to copy these commands you will also have to install or you will have to
launch Ubuntu OS right so what I'm going to do is I'm gonna make this server as a
web server so in order to do that I'll pass on the commands sudo apt-get so
let's update this machine first sudo apt-get update and this will now update
the machine all right so my machine is now updated
now let's go ahead and install Apache so sudo apt-get install Apache to
so guys what is the pachi Apache is basically a software for web server and
what is right now happening is it is being installed on my open two machine
once it gets installed on I want to machine all I have to go do is I'll have
to go to the IP address of my VM instance and then I will be able to see
a web page over there right so it's almost complete guys so let's
wait for this to complete and then we move forward
all right so my apaches has now been successfully installed and now if I go
back to my machine or my portal manager pool I just copy this IP address and now
let's paste it in a new tab and hit enter
can you see you this is basically a website which has been created by Apache
2 it's basically a default page but you see when you have installed a patch a -
but nevertheless guys if you guys go to this IP address given that I would not
shut off this machine you will be also able to see this particular page which
is basically the Apache a to Ubuntu default page now what I can do is I can
also go ahead and edit this web page right so let's go ahead and go to the
location where this web page is basically stored so it's stored inside
where www.h tml and this is the index dot HTML file which is being shown over
here so what I'll do is I'll just change the name of this index.html file to
let's say 1 dot HTML file and I'll have to give sudo in order to do that and now
let's create one more index dot HTML and let's say this is the HTML page whose
title would be demo website right and the body of this would basically have a
heading which would go like welcome to intelidox as your
okay so this is what I specified I closed the headers and I closed the body
and I closed the HTML as well okay so I've closed the HTML I've closed the
body I've given the h1 I've given the title as demo website let's save it and
if anyone LS now I can see that there is an index.html present now if I do a
refresh you can see I'm getting this welcome to and telepaths as your
training and the title of my page is demo website right now you would have
also noticed I made the Apache page as one dot HTML so if I go to one dot HTML
this is the Ubuntu default page and if I simply go on the IP address it will open
in extra HTML which is basically this so guys we have successfully deployed our
Linux VM on Azure and you're successfully hosted a website also on
this particular so now guys remember that this is infrastructure-as-a-service
now what I've done is I had the access to the operating system
I basically installed Apache 2 and then I put a website over there in that
particular folder which is Val www.h tml and then I caught the website right so
this is infrastructure as a service you're getting the whole operating
system and you can do anything with it I can also install MySQL on it and I can
configure my website to basically talk to the MySQL server right I can do
anything with this so now when you compare it with app services which we
saw earlier that app services are basically platform as a service or
function as its functions app is also a platform as a service now how is it
different from the hands-on point of view you will basically not be using
putty you will basically just deploy that app that is the web app and you
will get a URL instead of an IP address you will get a URL and if you go to that
URL you will see a default page so let's let's say you saw a default page when we
install apache2 you will see a default page from Azure
when you go to that link that this is a default page for the web app that
deployed right and you if you find the dashboard you will see a button called
upload and what you can do is like I created this index or an HTML file you
can also create yours and you just have to upload it now as your will take care
where that file has to go inside the file system right and it will take care
that on the link when you go now you will see your website it will take care
whether what software is it has to install whether it wants to install
apache2 which is basically the web server or there are several other web
servers as well you have nginx you have you know Tomcat so anything that you
want or anything which is configured in result would be automatically installed
and all you will get is a button which is ask you to upload the website so you
would not get direct access to the operating system and hence it is called
platform-as-a-service similarly with functions are in app services basically
if in deploy applications in function up you cannot deploy application it is
simply a place very quick just put your code and that code will be done right so
if if in short I'll have to tell you what functions happens it basically does
processing ok guys so guys this was our demo this was the first demo as to how
the you can launch virtual machines on Azure now let's go ahead to our next
topic alright guys so next in our list are the
networking services of azure so let's go ahead and start with them
so guys basically there are five core services in networking that you should
know of the first one being the very important one which is virtual networks
now what a virtual networks what your networks are basically isolated
environments or isolated networks on the azure infrastructure for whatever
machines or whatever the VM steady launch in Azure and if you want them to
talk to each other for example I deployed an application where probably
able to have a database I would have a back-end server I would have a front-end
server now all these servers have to interact
with each other so that the whole application can basically function now
how can they interact with each other they can interact with each other only
when they are on one network right so that is very important otherwise they
will have to interact with each other other
over the internet but that is not a secure thing to do right it is a cure
thing to do is when instances or when servers interact on their own private
network which is basically not accessible to the outside world it is
only accessible to the administrator who has probably launched those instances so
in those cases we need virtual networks and also one more thing that many are
using virtual networks or if you are using private networks in a jar you
basically get a bandwidth of around 1 GB per second right that is the kind of
bandwidth that you get when you are dealing with instances which are in the
same network that is in the same virtual network but if instances have to
interact over the Internet obviously the bandwidth will go down and
at the same time it is not secure right so whenever you want to launch any
resource in Azure which basically can be launched inside a network they are
launched inside a virtual network and you do not have the option of launching
a instance or a server without a virtual network you have to launch it inside a
virtual network now that if you don't want to use it and want to have
instances talk over the internet that is your choice
but as your does not give you an option of deploying an instance without it
being a part of a virtual networks alright so this this is a virtual
network moving forward now let's talk about load balancers now what are load
balancers now this is a very important service reason being when you deploy
your application on cloud one of the most popular reasons of deploying
applications on Azure is because you can get high availability which basically
means you can launch your application on multiple servers so that even if one
server fails the other server can basically be their replica and they can
serve the application right so this results in a high availability of your
application now what is a load balancer a load balancer basically sits in front
of multiple virtual machines which has the same application running on them
right and it sits in front of it now obviously if you are a customer you do
not realize whether a application has been made redundant on multiple servers
right facebook.com/ but if you go to Intel
about comm you wouldn't realize how many servers are actually working in the
backend so how is it possible right you have to know basically servers they have
IP addresses right so you as a customer should know which IP address you should
ping but all you do is you go to a particular domain let's say Intel about
comm and you get the website so how is it actually working so basically the way
it works is that load balancer has its own IP address or domain name and what
the load balancer does is it basically randomly spreads the request out to the
n number of servers it has let's say my intellibid website resides on five
servers wherein one is the primary server and for other redundant servers
but all these five servers are basically serving my website now when a customer
comes on to Intel upon his request is processed in this fashion the bet that
whenever he comes in telepods com basically he is routed to the load
balancers domain name or IP address and once a request reaches the load balancer
the load balancer sees what which server basically has less load right and the
server which has less load it sends the request to it all right now your parents
will is also an important concept you must have heard about auto scaling right
or if we talk it in terms of Azure it's basically called VM scale sets so what
happens in that particular scenario is we specify a threshold limit we specify
that whenever the CPU usage goes beyond 80% of the aggregate number of server
let's say there are four servers and the average CPU load on all these four
servers has gone beyond 80 percent right so what happens in that case it will
basically launch a new instance now when it launches a new server with the same
application your load balancer should also trout traffic to it and that is
when load balancer plays a key role it plays a key role when you are doing auto
scaling and at the same time it plays a key role when you have multiple servers
which are basically serving the same application and you want to equally
divide the coming onto your servers or basically
coming onto your balancer on your servers right so this is what a load
balancer is as the name suggests right but the load balancer in its most basic
sense it it basically does it randomly right it would randomly equally
distributes the traffic among all the servers it does not find follow any
protocol a rule as in how it has to basically divide the traffic right so
this kind of process or this kind of procedure when it is followed it is
basically called a round robin fashion of distributing traffic right the next
kind of load balancer is the application gateway load balancer now what does the
application gateway load balancer the application gateway load balancer
basically works a little different from the normal load balancer that we
discussed earlier so in the application gateway load balancer basically the load
is distributed based on rules now what are those rules those rules are
basically paths now for example if I go to in telepods comm slash blog right
what will happen I get to see the blogging website and if I go to in
telepods comm slash all courses I basically see all the courses which are
there on the website now how does that work now you if you are from an IT
background and if you understand how servers work you might say that there
must be two folders inside the root document of the web of the server
wherein inside the all course folder you have the code for displaying all the
courses and inside the blog folder you have the code for showing all the blocks
but that is not actually how it works right so basically if you if I were to
talk about our infrastructure the intel parts infrastructure we have a separate
server for blogs and we have a separate server for all the courses that we have
right now the load balancer that we use is an application gateway load balancer
it's kind of like that so basically what happens is whenever it sees that in the
URL is a path which says slash blog it basically routes the traffic onto the
blogging server and whenever it sees that the path is slash all
forces it routes the traffic to my courses server so this is how look this
is what path based routing and basically it works on the layer seven so if you
guys are aware of the OSI model right so according to the OSI model the
application gets a load balancer works on layer seven right now if you are not
from CS power and I think and if you didn't understand what what I just said
it is okay so in the most basic sense what the applications say the layer
seven that it works on in the most basic sense what that means is that
application get with load balancer does path based routing right whatever path
is there in the request based on that path it basically routes the request to
a particular server you can also define a default rule in case anything which is
there except slash blog and slash all courses if there is a default server
that you want to send your request to that is also possible
like for example you would have seen that some websites if the link is not
found on that website they basically route to a 4:04 page right now how does
the typical 4:05 for page look like right if you look a 404 page it
basically says error for zero for content not found right but some
websites like if you if you would have visit flipkart.com or amazon.com or even
in telepods dot-com in that case if you try to search for something which is not
there on the blog for example let me just switch to my browser right and let
us go to let's say and telepods calm and let's say slash as your right now what
will happen it'll basically give me a 404 page and let's see how that looks
like okay so there is a slash is your thing let me do one thing let me type in
some gibberish content then let me just search for that so this should give me a
404 all right so you can see I've got an oops page which says we were unable to
find the page you are looking for and if I change the content also over your like
let's say we enter some random text again I will again see this page can you
see it says page not found and similarly the way it happens for other
website Israel for example if I go to amazon.com thing gibberish let's see
what happens it says four zero four document not found and this is a
basically a custom page right and if I again go to anything other any other
gibberish UI I again get this page the heart is how's that working
so that is working because they also have using an application load balancer
which basically says anything other than the rules that we have specified if
someone goes or tries to go to a particular URL like the one that we have
specified route them to the four zero four page which is this right and that's
how the application load balancer is bringing you to this particular 404 page
right so guys this is the application gateway the next service that we have in
networking is called the DNS rooms now what is DNS zone now any website that we
go to guys you do not enter the IP address for that website right you
basically enter a domain name and that domain name basically gets you to the
website right you never enter the IP address so similar is the case with DNS
zones as well so what DNS zones helps you in is basically it helps you to
route your domain to the azure resource where your application basically resides
for example you go to any domain website and you buy a domain let's say you buy a
domain which is a personal website dot XY said let's say you buy this domain
now you want to route what whoever is going to this domain you want to route
their traffic to one of the virtual machines that you have launched near as
your portal now how would you do that for doing that you will have to go
inside the DNS zones service officer and you basically will get some name servers
right those name servers are basically DNS servers that as your owns and those
DNS servers you would have to specify in your domain so very bought your domain
you would basically add a dashboard wherein you can put theirs
DNS servers which the domain that you have bought will ping on right for
example whoever will be going to person web site dot XY said would basically be
routed to those DNS servers that you will be mentioning over there right and
those DNS servers you get from DNS zones the next thing that you do is you
specify in the DNS zones that whatever traffic is coming from the domain that
are that that basically has specified route them to this particular VM
instance I'll have to specify that in DNS own service and that is how it will
work so whenever you need use or whenever you have a use case like you
have the domain with you or even if you want to buy a domain you can actually
buy the domain at the is your dashboard as well but let's say you have the
domain itself right and you want to point it to any as your resource which
is existing in your dashboard so the way to do that is going inside DNS one write
a next service guys is again a very important service which is called CDN
now what a CDN CDN basically means content delivery network now how does
that work what is a content delivery network a content delivery network
basically it basically improves the time taken to basically serve you a website
for example if you go to amazon.com or if you go to facebook.com let's say do
you happen to see the speed with which the website loads if you have a good
internet connection it grows very fast right now where do you think the way
Facebook website is actually hosted right now what happens is now Facebook
is a multinational corporation right it's a it's a big website which is
athletes being used by the whole world right so they have different data
centers I cross the globe from where the requests are basically served but let's
talk about a smaller scale website let's let's talk about in Telecom right we
don't serve the whole world we basically might have around 400 to 500 people at a
particular time on our website right now what happens is we get traffic from
across the globe right we want everyone to have a speedy experience when it
comes to them using our website right now what is
what are the ways that I can ensure that let's say my traffic is coming from us
right the traffic that comes from us is like let's say 60% of my overall traffic
so what I can do is I can set up my data center in the US so that the US people
will actually get the website fast right but let's say someone is accessing the
website from Japan so in that case what you will have to do is whenever you goes
through in telecom the request will basically be sent to the US servers and
from the US servers the request will come back and that's when his website
will start to load right so this basically increases latency latency is
basically response time of a particular application in our case it's a website
in telecom now if you want to reduce latency what are the different ways to
do that the different ways to do that is the first way that you can do it you can
basically open the data center in Japan as well and load all your scale up the
number of servers that you have and then also put your website over there and
whenever people from Japan will basically be accessing your website they
should be seeing the Japanese version of the website and the people in the US
they should be seeing the US version of the website the people in India using
the Indian version of the website right that's how it works when you use an
e-commerce website like Amazon so you have Amazon or ten you have amazon.com
if for Australian people it's Amazon dot type so for different people countries
it has the different extension which basically means that that particular
website is being served from their home countries data center right but there is
another way to basically serve your website faster and that way is basically
called CDN now what I see the end Sirian basically caches all your state a
static data right for example what kind of static data are we talking about
let's talk about videos let's say my intellibid website also plays videos in
the self-paced courses now if you are a person who is in Japan and let's say my
servers are there in the US so what will happen is the moment you try to play a
video on my website then video will get downloaded to the nearest
location often as your data center right so while you watching that video that
video has been downloaded on the data center of the Azad premises in Japan and
what happens in later is basically whenever any other person in Japan would
try to exit that website he will get access to it through the Japan server
rather than the USO so what I have done is you have set up your servers in u.s.
you have not set up your servers in Japan but what I have done is you have
basically enabled CDN on your website and what happens in that case is all the
static content is basically loaded on to edge servers of azure
so whatever servers are used to cache data cache static data in the CDN
services are called the edge servers right so whoever is accessing my website
from whatever part of the world if they have as your datacenter near them my
content will basically automatically get downloaded over there and whenever there
is a next person who is trying to access it that already downloaded data will be
served to him directly and that's how I reduce the latency of my application
without setting up servers in different locations for different kind of traffic
that I get right so this is how it drastically reduces costs because I'm
not launching new servers and at the same time it serves the purpose of
reducing your latency of your application so this is what CDM profiles
is gied so then these were the core services in networking of jihad right
now let's move forward and talk about a next set of services which are basically
the storage services all right so guys next in a list we have the storage
services of azure so let's go ahead and have a look at them so guys essentially
there are six services that you should know of while you are using Azure in
terms of storage the six services are blog file storage tables queues derelict
storage and data box so let's discuss them one by one the first one being blog
now what is blog blobbs it basically means binary large
objects right so if you have binary files and you want to store them on as
your blob storage is the answer for you right so it can store anything from
music files or it can store video files it can also store text documents any
kind of file that you want to store on Azure can be stored on blog now blob
also can be used with your in conjunction with your websites and it
can basically act as a storage server for you right it also enables you to
host content which is publicly accessible over a link it can also host
static websites on its storage so that is what blob is guys right so most of
the time you would see that companies if they have a website all its static
content is actually picked up from the blog right so this was about the blob
guys our next service is file storage now what is file storage file storage is
basically a shared file storage that can be used with multiple computers that
basically means if I have five servers right and let's say they need a
particular file for their working and all these five servers basically need
the same file so what I can do is I can basically create a drive or I can create
a storage point on file storage and all all these file servers can be used or
can be used to basically mount the file storage drive on them right and any
change that is done from one server on the central repository of file storage
would all those changes would be reflected on all the next four servers
the use that I can think of for file storage is let's say you have an
application which writes data onto a particular file right and basically this
application of yours is spread across multiple servers to ensure it is highly
available right let's say you have an application which basically writes data
right and this application is cos Phi servers now what happens is you
we've already discussed the role of a load balancer you hit the load balancer
and you're randomly assigned a server and you work on it let's say you work on
server one as a customer you would not know bit server you're working on
correct let's say you work on server one right you saved your changes and now
those files are now saved on server one but the next time you hit the URL now
you are unser three if you are on server three you would not realize it but your
work you would expect it to be there on server three as well
now here comes a problem if there is no central storage for all these files
servers they will be asynchronous in nature in the sense that they will not
sync data with each other whatever changes are done on the first server the
third server the fourth server and the fifth server will be unaware of it right
and that is when the arise the need of a central storage and that is what file
storage is for right you can it basically uses the SMB protocol SMB so
basically SMB 2.1 and 3.0 is compatible with file storage and recently even
Linux platforms like Ubuntu sent to us in even Mac OS
have started supporting SMB right so it essentially would recommend you to use
SMB 3.0 but there are some OS versions which could have a SMB 2.1 for example
if you are using ubuntu 16 or a12 14 point o4 they are basically using SME
2.1 right so it has the backward compatibility as well for older OS like
1 to 14 0.04 so if you wanna mount drive on Ubuntu which basically is hosted on
as your file storage you can do that using the SMB protocol there are lists
of software that have to be installed more on that we can discuss when we
study file storage in detail but for now because we are in a tutorial session we
are basically I just told you what file so it is useful so if you want to ask or
if you want to think of a use case where you need a central storage kind of a
thing now you know that you have to use as your file storage moving on guys our
next service is as your table now as your table is basically a no sequel data
store right and it can help you to store structured data aright it it basically
can store structured data it has table tabular columns and table rows in which
you can save data but it is no sequel in nature which means the data does not
have to be in symmetry right your next data could have eight columns the third
the third row could have two columns the fourth row could have just five columns
it could take any kind of data but the only condition is that the data should
be structured among columns right so if there is any kind of need for story the
data of this kind you can use as your table and next a storage is as your
queues which basically is used with stateless systems right systems which do
not know of what all jobs are executing on the counter part or on the replicated
servers of their application for example let's say there is an image processing
website and what it does is the moment you upload an image you can process the
website or sorry process the image according to your need right so what
happens in this case let's say there are 200 people on the website and they have
all pressed the process image button together and what will happen there are
200 images that have to be processed and obviously not all 200 images can be
processed at the same time so what happens is the images are processed one
by one and let's say the in the backend server where the processing is done
there are five servers who are doing the processing okay now all these five
servers they pick up your image at random from the queue so let's say the
queue is first in first out whichever image comes first goes out first as well
so whenever the job of the first server is done let's say it was doing the
processing on an image and the image processing is done it picks up another
image from the queue the second server will also do the same the third shovel
would also do the same right so what happens is let's say the first server it
processed image more now the second third fourth and fifth server
would not know whether the first image has been processed or not and they can
process it again so for those kind of scenarios and for
solving those kind of situations we had cues which basically streamline all the
content which has to be processed and whichever content is taken up by a
particular server it deletes it from the stack and then puts the second image or
the second object which basically has to be processed right so for these kind of
scenarios you use as your cues an external service is data like storage
now data like storage is similar to that of a blob not a blog but tables as your
tables so data like storage can store data but it basically used to store data
for big data analytics it specializes in that particular cool segment so if you
have a big data analytic use case and you want to store data for that data
like storage is should be the choice for you right so this is about the lake
storage lies the next service that we have is as your data box now what does
your data box now you might think of your company let's say let's say you're
working for a company and you have all the servers on from ice right you have
been working since 15 20 years in the sector that you are I'm in your company
and you did not have cloud at that time and what I have done is your body our
own servers you have hosted your own applications but now the cost incurred
for those servers is too high when you compare it with the cost which is there
in cloud computing when you opt for cloud computing so now what what I have
decided is basically that you want to migrate all your applications to Azure
data box ok or sorry to Azure so let's say there are five Pete there's five
petabytes or let's say thousand terabytes of data that you have to
transfer on Azure now in businesses time is very important guys
time is money basically you must have heard of that phrase so everything is
required at a fast pace and if you want transfer a thousand terabytes of data
imagine the data transfer which would incur to you and also it
would take a lot of time to upload thousand terabytes of data wire Internet
to Azure right you would agree on that now to solve problems of this scale when
you have a large amount of data that you want to transfer onto as your cloud what
has ever does is it basically gives you a physical device on which you can load
your data right so as already the box is a service in which you can you can
basically request a box kind of a system from Azure which basically ships through
your company you can load all your data onto it and then that box is again
shipped back to Azure cloud and all your data is again uploaded to Azure cloud
from their data center directly right so this not only reduces the time taken to
transfer data but it also hugely reduces the cost that you would incur in the
internet charges because thousand terabytes is not less data is now
thousands otherwise is very less when I talk about a company which is existing
since 15 years and when you talk about multi national corporations the data can
be in petabytes right and if you have a petabyte scale data that you want to
transfer to Azure it would take weeks if not months to transfer that kind of data
so then that is where data box comes in then data backs a data box as your data
box can basically support around five petabytes of data at one time right now
if five petabytes of data is less for you you can make request multiple boxes
from Azure to and multiple boxes will reach your location you store your data
onto it I mean you copy your data onto it and then request your back that you
can take over your data boxes and they will take the data boxes back to their
data centers and transfer their data directly right this will then use your
time to around one week of time so your data of petabyte scale can be
transferred to Azure in less than a week and at the same time you would also save
on huge internet costs right so this is what the data box service of azure is
alright guys so now let's go and do this hands-on on Azure blob
storage so let me quickly switch onto my Microsoft Azure portal which is this
alright guys so this is my Microsoft Azure portal what I want to do is I want
to show you guys how the blob storage works in Azure
right so the first thing that I'm gonna do is I'm going to go into storage
accounts and over here you will find that there is a storage account called
demo environment why is it here because I created virtual machine with the name
demo environment right and obviously the virtual machine requires some storage if
you go inside this you will see that you have the four options that we discuss
that a blob files tables and queues what we are interested in is blob so we'll
just click on that and you can see that this has boot Diagnostics for the new VM
Linux that we launched earlier right so all the logs for that VM are basically
going inside this container so basically if I go inside this container you can
find that these are the two text files which are present we'll basically we'll
have the log data of that VM that we launched right so but we're not here for
that what we basically want to do is I'm
gonna show you that is how you can use blog now no matter what you are working
on if you want to create a blob so it's for yourself you will have to create a
storage account right in our case this was the VM storage account if you are
starting afresh on Azure you will not have any storage accounts over here
right so let's create a so account first LAN let's try to name it something
relevant which might go with the demo that we are doing so the resource group
that we want to add it in his demo environment and the storage account name
that we more give is let's say in telepath - Chris your let's say this is
the one that we want to give okay where is the location that we want this toy is
account to be in as discussed earlier we'll choose a nearest location to us so
in that case the nearest location would be
asia-pacific South India let's try to choose this location right now how do
you want the Poppins it could be just under the premium we
can discuss this later probably when we'll discuss this service in detail now
for the sessions but for now let's let it be head standard right put everything
a default and now let's click on review plus create when you do that it will ask
you all the basically show you all the information with respect to the storage
account that we are creating if everything looks good just click on
create once you have clicked on create guys your store account will take some
time and then it will become ready for use and like earlier I specified all you
have to do is go to resource groups let's say this is the demo environment
that we created earlier I'll go inside this and now I can see all the things
that I had launched a present over here even these stories account that I'm
creating that is that is basically being launched right now I can see that also
over here as you can see it's in telepods as your this is so is the con
that I created and now it is visible inside demo environment resource group
on the other hand if I go to all resources I'd be able to see all the
resources that are there as your account but this looks a little messy and that's
why resource group Cyprian focus because I know that my resource that I've
deployed is sits inside a specific resource group and that is also
basically I can name on the basis of what tasks that particular particular
resource is going to do since hours of the demo session what we have done is
all the resources that we are basically deploying are inside the demo
environment resource group right and right now we just deploy this tools
account which is in telepods azure and as you can see it has been successfully
deployed now let's go inside the storage account and like we saw earlier you see
a similar user interface over here you have four options either to go inside
blob or to go inside files or tables or queues let me demonstrate to you how
plops works so we'll go inside blob skies and as you can see there are no
containers over here water jános containers are nothing but root
folders right for example on Windows you have the dekstop folder so dextra folder
becomes the root folder or to give you a better example let's say I go to the C
Drive of your computer write whatever is inside the C Drive is the root of the
drive if you create a folder inside that drive and it then put files inside that
folder that is no longer the root directory right so in whenever you are
uploading files on blobs the first thing that you need is a root directory and
that prove directory is nothing but containers right so let's go ahead and
create a container so I click on a new container and let's name this continue
let's say as Intel apart ok and now let's click on OK alright so my new
container is now ready now you would have noticed that there is something
called as access level and right now the access level is private more on this
I'll explain to you as we move along so right now we have not changed it to
anything it's private and I will be at private now what I'll do is I'll go
inside this container and it says no blobs formed right so what I can do is I
can upload some files onto this container and for doing that I'll have
to click on upload and now I'll have to select the file so let's so these are
some images that I have on this system so what I'll do is I will just select a
random image and now let's upload this image onto my flow alright so this image
is now uploaded on the blob and what I can do is I can just click on this image
and I can see all the information with respect to it it's in it's 88 KB in size
this is the last modified time this is the creation time now here is something
interesting guys I have the URL to the file that I uploaded right now this URL
can be embedded anywhere it can be embedded on a website it can be embedded
on a post 30 making to social media this basically is a file which is now hosted
on the and anywhere anyone who has access to
this particular link would be able to see the image that you have just upload
awesome isn't it right so what I'm gonna do is let me copy this link and let me
open it in a new tab right and I'll hit enter so right now it says resource not
found this specified resource does not exist now why do you think this happened
this happened because we did not give access to the container for general
public use right what do what does that mean if I click on the container and I
click on change access level it's right now private which means I can only
access this blob through the azure portal all through you know CLI or
through some other way but only that way would be working now if I want everyone
and anywhere who wants to access this file I'll have to go ahead and choose
anonymous read access for blobs only right so when I do that I just select a
blob over here and I click on ok now the permissions for the files inside this
container has changed and what does that mean that basically means if i refresh
this now I would be able to see this image over here right similarly if this
link is copied by you and pasted on a browser you also would be able to see
this particular image right so guys this is what a blob storage is you can use
blob storage to dump data such as logs such as images such as videos that could
be either private if your application does not want to use it and you just
want to keep there keep it there for you know just keeping it there or what you
can do is you can provide it a public read access on the blob level that is
for all the files will have public read access and those files will then be able
to use by just going to their link let me upload one more file free ways so
that you are clear the concept of defining permissions so I
define the permissions on the container so I don't have to do anything now I'll
just upload the images and this is a new image which has been uploaded and now if
I go inside the image and I copy the link and if I paste link over here you
can see I am able to see the screenshot even here right the moment I go out and
I change the access level to private and I click on OK now if i refresh ok so it
did not change the permissions of the files not visible ok so that basically
took some time for the permission to get allocated on the container which
basically means like if I go out and if I change the access level again to plob
and if i refresh this files yes i am able to see them right and
again if I go out and if I change the access level to let's say private again
and if i refresh the first refresh it is there the second refresh it is there but
if you wait for some time like a minute or so this access will basically be gone
as you can see it now says resource not fund right so with a click of a button
you can basically control all the files permissions which is there inside a
container and grant them public or private relaxes right guys so guys this
was a brief demo about blob storage on a job let us come back to our slides
right so the guys I hope now you clear with what blob storage is now let's move
forward and talk about our next set of services which are the database and the
analytical service in Microsoft Azure all right so basically there are five
core services that you should know off when you're dealing with Azure in terms
of databases and analytics so the first service is the SQL database service
which basically is a database as a service right you do not get an access
to the operating system on Lister's database is installed
you basically get access to only the database and basically if you remember
what we have studied earlier for a service which does not gives you access
to the operating system but in turn gives you access to a platform kind of
thing where you can interact with the service and probably upload something
onto the software being used is called platform as a service so as your SQL
database is basically a platform as a service which not only gives you the
independence from the infrastructure side but at the same time it is highly
scalable and it can provide you up to 212 percent of return on investment that
means that whatever money you are going to spend on Azure SQL database the
benefits that it is going to provide you is going to give you back two hundred
and twelve percent of the money that you would be investing in using Azure SQL
database isn't that interesting right so this was the first service among the
database and let analytic services in Azure a next service is cosmos dB now
where is cosmos DB is basically a fully managed database service again just like
SQL Server but in this case what happens is your database is extremely highly
available which means it is distributed throughout the world using Azure regions
and it is highly available right so basically what happens is when you
launch a cosmos DB cluster it distributes or it creates a replica of
the database and multiple regions as you specified right and the cool thing about
cosmos DB is whatever region you want to close or want to stop and probably you
do not want your database to be replicated in a particular region you
can do that with the click of a button if you want it again in some other
region you can again do that with a click of
so everything is fast everything is at one the whole control is at one central
place and you can replicate your database accordingly with that one
platform or with one that one cosmos DB dashboard that you get right now again
you have multiple models that you can implement in cosmos DB for example you
can implement multi master replication or you can implement multi write regions
replication which basically means usually when you have a distributed
database which is highly available you have one region where all the lights
take place but the read can actually be done from any of the regions where the
database has been distributed but in case of cosmos DB you also have the
option of configuring multi region region multi write regions which
basically means that let's say I want to Emile India and my and I am a user who's
using your application right now if I am using your application probably the
website is being served from the nearest edge location of my location for example
you must have enabled content delivery network on your website if you want to
increase also it decrease the latency on your application right now to decrease
the latency of your application you have implemented CDN so I am being served the
website from the nearest location of my home from where am i accessing your
application and at the same time the application itself is also accessing the
database right now imagine the website is fast but the processing that it does
that is is the data that has a it has to fetch from particular database it exists
and altogether a different region which is on the other part of the world right
on the or on the other side of the world right that'll again increase the latency
of your application so it so you don't only need your front-end component to be
highly distributed you also need your back-end and your database to be highly
distributed and highly available right and that is the sole reason when you are
dealing with the audience which is spread across globe
you also have to take care that the processing that these that each of these
guys are doing would also be available to them in the lowest time possible and
for that what you have to do is the backend processors or the backend
servers and the database servers also have to be near them so that whenever
they're writing any information that also information can be written faster
now if you have a right one and read anywhere kind of a distribution in that
case you will have to write at a central location let's say the center location
is the US and your databases which are basically the read replicas are actually
distributed throughout the blue what are the replicas real replicas are basically
the copy of the central database from which you can read the data and these
read replicas basically get synced or basically are synced with a central
database with the so whatever you write on the database is automatically and
quickly replicated on the read replicas as well so it's very fast right so in
terms of reading the data you can get a very low latency but in terms of writing
the data if you want low latency as well then you need to have multi write
regions and that is exactly what this feature is falling so in a nutshell
cosmos TB is basically a fully managed database service which provides you with
a large-scale deployment of databases throughout the globe and it also gives
you the option of deleting or replicating your database on a new
region entirely with a single click of a button right so that is what cosmos TB
is for you guys right the next set of or the next service that we have is the
data factory service now what is the data factory service it is basically an
ETL service which is an extract transform and load service which
basically can take data from multiple sources and then transform the data
according to what has been coded in the application using regular service which
can be used in conjunction with data factory and then
load the resultant data into a bi service which can be used to analyze the
data right so usually when you are doing this kind of an infra setup you need a
lot of planning you need a lot of mediators between the technologies which
are being integrated for example the data can be coming from multiple sources
such as social media websites it can be coming from emails it can be coming from
chats it can be coming from reviews it could be coming from anything right
so all these data sources are basically first of all aggregated and the data is
first put in the raw form in one particular place then what happens is
according to the logic that you have specified for the transformation of this
data for example you probably want to see the data in a particular view so
that basically means you would have to first transform this raw data that you
have got from multiple sources into a structured format so that is what
transformation is so once you have the resultant data that you feel is the
correct way of getting the data or representing the data the next step that
you need is obviously if there are a million rows in your data set you won't
be able to find what you are actually looking for or basically it might take a
lot of time for you to analyze a million rows of data that you have just created
so what you can do is or what what the industry came up to this particular
problem was that they created a business intelligence application what is the
business intelligence application it takes in the data set and it quickly
creates patterns it creates quickly creates graphs which can actually help
you to better understand the data right so data factory also happens to have the
integration capabilities with a lot of BI tools the most prominent one is
Microsoft's own power bi tool but it also can integrate with tableau and
other BI systems all right similarly it can be integrated
with multiple data injection systems as well so it's a total ETL system that you
can deploy on Azure and connect your various sources and
outputs and get the resultant output okay so this is what data Factory is our
next service is event hubs now what is the event UPS event hubs is basically a
place where you can again take in a lot of data just like data factory you can
extract data from multiple sources but in this case you're not actually
extracting data but the multiple sources from where the data is being generated
the data is being pushed on to event hubs
okay so the data is pushed on to event hubs and the event hubs job is to
process this data and see where the stage data has to go next right for
example let's say the data is coming from a social media website right so you
analyze okay so this data came from a social media website it has to go to
this particular service this data is coming from this particular website it
has to go to X or Y particular service similarly you have millions and millions
of data coming every second or data packets coming every second and the sole
job of event hub is to basically analyze each packet and directed to the
corresponding consumer of that packet right so this is what event Apps is for
you so this is done very fast so the speeds claimed by azure is around 1 Gbps
that that is it can process 1 GB of data per second so the message is coming to
end event ups can be processed that faster and it also does parallel
computing of each data packet right so this is how what this is whatever event
hubs is for guys you know obviously event ups cannot be just used alone
right you have to use it with in conjunction with a host of applications
for example the producers that is the data producing resources have to first
combined with innovator and then you have to define the rules in event hub as
to what data goes where our what data goes to which consumer right and then
your consumers are defined on and they are basically connected to event hub
right so basically event ups in the most basic sense is a single point of data
ingestion so you don't have to worry about where should I send my data send
all your data to event hubs and event hubs will decide very Oh data packet has
to go based on what rule you are defined inside that system okay so guys this is
what event hubs is our next service is data link analytics right so we already
saw what data leak is so data link is basically a storage for big data
analytics right so data lead analytics is basically a distributed cloud based
data processing infrastructure right so it is basically architected to perform
data processing on the big data I mean the data that is stored on data Lake is
has to be processed by something right and that processing can actually be done
by data Lake analytics so it is basically the architecture is based on
yarn so yarn is basically a component of the Hadoop ecosystem and it can all and
basically pairs with Azure data Lake store or where basically we have stored
all our data and can perform the analysis on the scale of bitted analysis
that big data is huge amount of data with variety of data in it right so all
the data has mixed and matched fears right the data is in the rawest form
possible and the processing can be done only by using selected tools which can
actually make that happen if you use simple tools or if you use traditional
tools the time taken by those tools to process this model data will be huge
right and that is why you need loop tools to basically make this process
faster and as your have used the same technology that loop uses and has made
it even faster whew right so data link analytics is used in big data analytics
it will basically use for processing data and what data the data which you
store on zero data like store right so this is what data link and I'll take
guys moving forward guys now let's move on to a next domain which is the AI and
machine learning domain in Azure alright guys so in Azure you basically have
three core services in the AI and machine learning domain the three core
sources are cognitive sources then you have bought services and then you have
the awesome machine learning studio so let us understand the each of these
services one by one the first service is the cognitive services now what are
cognitive services cognitive services in a server are basically API is or SDKs
you know which have been developed for a developer which can be integrated in his
application and these API is and SDKs basically interact with the machine
learning models which have been created in Azure right so cognitive services
would include services like vision which basically gives you image processing
capabilities it also has text analytics services which can basically do natural
language processing for you so all these services all these services in Azure are
ready to use and to use them you will have to integrate them with your
applications and those integration can happen with the help of api's can happen
with the help of SDKs and once your application is integrated with these
services you're charged on the basis of the number of requests that you make to
these particular service for example the vision service and the cognitive sources
office your it's pretty awesome any image that you upload it will tell
you each and everything in the image whatever is present and how is that
possible basically as your has processed thousands and millions and even billions
of images already right and they have trained their image processing machine
learning model in a way that now it gives accurate results right so instead
of creating your own machine learning model for image processing you can
actually use the azores machine learning model and you can get your work done so
it might have used deep learning it might
the only machine learning we do not know but the results are very accurate and if
you want to use AI and machine learning capabilities office your so the models
which are trained by Azure if you want to use them directly you can actually
use the cognitive services of azure and to integrate them with your applications
you have seven lay PS and SDKs available to you so this was about cognitive
services guys now there's a derivative of the cognitive services that roger has
launched which is basically called the bot service now what if the bot throw is
guys the baud service is basically a chat bot which has been developed by
Azure which is totally based on AI and it basically makes uses of the natural
language processing service or the capabilities which are basically there
in the cognitive services of a Searle it makes use of that right and it is very
to the point and you don't have to train the model to become expert in the
conversation that you're having but it can actually learn from every
conversation that you have with the AI chat board of Israel and you can
actually tweak its setting so that it becomes custom design for you so that
your customers get the right answer every time they ask a question from the
bot service officer right so in a nutshell as your bot service is nothing
but a chat bot service it's a pre-built chat pod service which you can integrate
in your application and this chat bot service has already been trained by
Azure and it gives accurate answers to questions and but obviously you're you
will have to tweak it according to your applications you'll have to do some
settings so that it answers the correct way of answering the answers would be
according to you and according to your domain of application okay so there's
this was the bot services and next set of service is the machine learning
studio now what is the machine learning studio guys it's basically a very
simplified version of using machine learning for example the most simple
language that I think for starting off in data science is our right so as a
beginner if you start using our or start learning
you will have to spend some time to first of all understand the syntax how
it works etc and then you'll get a hang of how to create mortals how to train
them how to test em etcetera right what machine learning studio says is you
do not have to learn any programming language it has drag-and-drop interfaces
so all you have to know is their science AI or machine learning conceptually and
if you understand or you feel have the principles of the data science concepts
understood play well what you can do is you can start off by creating your own
machine learning model by just doing drag and drop from the UI of the O's
your machine learning studio and you'll be up and ready with the first model in
under five minutes of course for a specific set of data set obviously you
will not be taking millions of billions lines as a result otherwise that will
take a little more time but yes you can create your own machine learning model
with Azure machine learning studio without knowing any programming language
right so that is the power of synchronous video and most coolest
feature of it is it has a drag-and-drop user interface so everything is
drag-and-drop you just drag and drop whatever you need whatever data set you
need whatever recorded them you want to implement whatever thresholds you want
to implement you just I can drop everything and it will work like a charm
okay so guys this was the machine learning studio application for you the
next set of services that we are going to discuss guys are the identity
services in Azure alright guys so now let's go ahead and understand the code
of your services in the identity domain right so that is the probably the most
important service when you are dealing with identity in Azure is as your Active
Directory now what do you mean by identity identity basically means when
you want to give a particular person some access to a particular resource
that care is being used on Azure cloud right so as your Active Directory is
basically a fully managed multi tenant service from Microsoft that basically
offers identity and access to particular users
in your organization now if you are acquainted with how Microsoft Windows
server works you would there is a server active directory as
well right there's a Microsoft server active directory as well in which you
can specify for your on-premise applications which users have access to
what extent right for example some people can just have read access or some
people can have read and write access or some people can even have admin and
access to a particular application right and most important part about as your
active directory is that it can integrate it with your on-premise server
Active Directory and can also authenticate people who are there on the
on-premise infrastructure and want to use the compromised resources like or
on-premise software can also be authenticated using Azure Active
Directory and if those optimized users then you want to use resources on Azure
that also will be authenticated using Azure Active Directory right now how
does all this works so it's a very simple concept guys what you can do you
can actually add users in Azure Active Directory right now you can add
individual users and you can assign individual roles to each users for
example if there is a SAS application which you are deployed on Microsoft
Azure giving access to that particular SAS application to a particular user
would be possible by just adding a user and giving him the access of thus that
SAS application right and you can also do a mass a location of a particular
permission for example let's say you want to give administrator privileges to
a set of people now how would you give that so one way is whenever there is a
person who comes into a company whom you think can be given admin access we will
add as user and provide them the admin access for what you can do is you can
actually define a group on which you have specified admin resources also
admin privileges and whoever users will be added to this particular group
they'll get those permissions or will get those privileges automatically right
so in this case what you will be doing is whoever is coming into the
organization you are just adding him to the group and the permissions
inheritance takes place automatically because
that group has been assigned the admin privileges similarly if you want to
create some other group for example you want to create a group for some SAS
application you can also do that right and this concept is a part of Azure
Active Directory and again in a nutshell Azure Active Directory is nothing but
it's basically a directory in which you can add users and specify what these
users can do on Azure or on on to my software right and these users can be
managed in two ways either either you can manage them individually by creating
a user time and again all whatting in two ways you can add these users to
already existing groups that you have created with the respective permissions
and they'll inherit all the properties which have been specified to that group
okay so there's this wiz the azure active directory in this will about the
identity domain in Microsoft Azure moving forward guys now let's talk about
the management tools in Microsoft Azure that are very important for you to know
while you're working as an azure professional all right so let's go ahead
and understand the management services of azure so the first service is log in
analytics guys now what is log in analytics so basically the full name of
this services login analytics workspace now what you can do with login analytics
workspace is you can dump all the logs for with respect to what is happening on
a particular as your resource on to this workspace now how do you do that the
first thing that you do is you basically go to your management console and you go
to log in analytics workspace over there you will have to create a login
analytics workspace and specify the name anything right it's please
straightforward once you specify that the next step would be to add a data
source so if I will show you quickly I can just jump on from a management
console I can go to log narratives workspace now here is the sample log
analytics workspace that I've created I'll just go inside it and as you can
see it says connect a data source so you can create or you can basically connect
as your virtual machines over here or you can specify some other sources
you want to and you can also specify as your activity log so stream over here
alright so this will start accumulating logs in this workspace once that is done
we will have to specify monitoring solutions as to how you want to read the
logs and what do you want to do with respect to a particular level of logs
for example you have info level you have critical level in logs right this
basically tells you that a particular command failed or a particular command
passed so if the level or the level field of a log effect let's say info
that means it's fine it can go in but let's say the level should in the log so
what do we mean by level field a little filled with nothing but you can specify
in your log notation that whenever there is an error
I would specify a term such as level okay and I would specify that is equal
to critical but if things are okay it's just for know-how as to which command
has been executed and that command executed successfully in that case I can
specify the level as info okay and then I can specify in my monitoring solution
that whatever log comes in check the level field in that log and if the level
field is critical you can flag that log basically so all that can be configured
inside the monitoring solution so if I were to click on view solutions you can
actually check out what all monitoring components you can actually add on your
log in a textbook space so as you can see you can add as your security sensor
you can add optimize catalog so there are a lot of things that you can
basically add on to monitor logs right you can choose according to your will it
is available in Azure marketplace so once you specify the monitoring solution
god log starts to flow in and in your monitoring solution is after specify
which log so flag and that's how it's gonna work right so this is the log
analytics service energy-wise in and it helps you to manage your Ezio resources
effectively pinpoint where the problem is
go over there and see the logs see on which command problem occurred fix a
problem and then go ahead with your daily routine so this is one log
analytics is guys the next service is cost management and billing so this is
basically a native application of a service which helps you to manage your
bills in a journal right so for a few things which I can tell you on the top
of my head which comes in is that he can manage a budget you can specify that
might as your budget or my as your bill should not go over a particular limit
you can specify that or you can so when you when you say it could not go over a
particular limit that means if it goes beyond that all your services will be
stopped but when you are in a business you actually don't want that so what so
this particular limit is called a hard limit that is you want to limit it it to
that particular cost there is another kind of limit which is called the soft
limit which basically means it will alert you that your budget has been
crossed and you can actually make forecasts also in this particular
service based on your uses lets you have used as you're heavily four five days so
it gives you a forecast that if you happen to use as you're in this similar
fashion this is what your bill is going to be at the end of the month right and
then you can actually see what kind of tweaks you can make in your services so
that your bill comes down so all of this is possible in cost management and
billing service officer now this also is under the management domain reason being
it helps you in managing your resources in a better fashion the next service
guys is a very important service which is an automation account now what is an
automation account is it is nothing but it's it's a way of deploying as your
resources so how do you reply as your resources so we have seen few of the
resources in this session today therein we deployed a virtual machine we
deployed a storage account so all of that was possible one by one I mean I
had to go manually into that service I had to click on add I had to specify a
name and that's how you know my resources were deployed but in
automation account what happens is you can create run books now what are the on
books you basically specify a code where and you specify all the resources and
their names and their configurations that you want to deploy and then you
upload that run book onto the automation account and then you run that run book
then what it does is it automatically creates the resources for you you don't
have to do anything just put that run look over there click on play button it
will create all the resources for you with the exact same configurations that
you have specified in the code it is particularly helpful when you're setting
up a large infrastructure wherein you'll have to deploy thousands of to thousands
of a scene where each where the two thousand machines are basically divided
into multiple sets it could be back in Somerset it would be database services
it could be front-end server side all of which will have different kind of
configurations and if these this is the kind of a scenario then run books would
be of great help and that's where automation account also comes in which
basically helps you to play those mangos okay a next service is matrix guys no
matrix is a service which helps you to visualize what all is happening on your
as your resource for example what is the network throughput what is the CPU usage
right what are the memory usage all of that you can see inside matrix now you
can actually check a 24-hour log as well in matrix and see at what point of time
does your resource or your or the particular metric that is checking who
BCP usage will be memory using is actually rising and then you can plan
accordingly as to how to plan your or how to scale your infrastructure at that
particular time right so this is what metrics are useful for it basically
gives you an overview of how things are going so if you think about it it is
nothing but metrics sorry it is nothing but logs which is basically basically
getting visualized in from of graphs but nonetheless it's very helpful and that
is a service which is being provided to you by a sir by the name metrics all
right guys so these were all the services in
management domain guys are next up we'll be discussing a hands-on so enough of
theory I guess we have discussed probably most of the important services
which are there in the azure portal now what we'll be doing is I will be showing
you an application which basically exists on my local host and what we try
to do is we will try to migrate that to Azure and we'll choose the services
based on the knowledge that we have gained today we'll choose the services
accordingly then we will set up the infrastructure and then we'll see how it
works on the cloud so there is let's go ahead and do that alright guys so let's
go ahead and let me show you what all we are going to do so the first thing that
we're going to do is basically I have a website that I've already created on a
local host which basically can upload data to Azure blue blob storage through
the website so I'm going to show you guys how I have configured it and how
can it basically upload data to us your blog right so the second point is create
a new container and upload files to this container from the website right so we
will be creating a container and our website we're going to configure to
upload files to this particular container third thing is we have to
create a MySQL database on as yours right now what is happening is my
website is basically storing data on the localhost MySQL engine what the hands-on
expects us to do is we will have to deploy MySQL database on a jar and the
website should basically push the data on to the MySQL database which will be
there on the portal right so this is the third step and the force F finally we
want our final website to be uploaded on as your web app and basically we have to
deploy it using the local git method what it is don't worry about it as you
move along you get it right so let's start off at the first point right now
which is demonstrate the website on the localhost so let me switch on to my
browser so guys this is my azure portal and if I have to demonstrate to you my
website it basically exists as your one right so this is my website guys and
what it does is it basically can upload data onto
Microsoft Azure but right now what we are not basically considered to connect
to my Microsoft Azure storage account I will show you how we can configure it
right so once a file is uploaded on the blob it also makes an entry of it in the
database right now whereas database the database exists on my local host so what
I'm going to do is I'm going to just open the mash in the console right and
my MySQL console basically has a database called images so if I do a show
databases you can see that there is an images database so this is the database
that my website will be interacting to so let me just use images and in this
database what I have is a table so if I do a short tables command you can see
that I have a table called names right but right now there is no data inside so
if I do a select star from games as you can see it is empty set right so when
the data when this website will be uploading file when it will be
successfully uploaded next step would be to basically put the data inside the
database as to what is the name of the file okay so let me show you the code
guys let me show you how the code looks like this is my index of PHP let me just
open it alright guys so guys this is this is how my code looks like so the
first thing that I'll have to do is basically I will have to enter a
deployment key over here which will basically be my connection string and
the second thing that I'll have to do is I have to enter the container name which
I'm going to create in a little while right so let's first jump onto our job
portal and what we want to do is go inside a storage account so guys this is
my storage account and what I want is the deployment key so in the storage
account you will have two to access keys just go inside that and
here you will find two keys you can take any of these two keys so what I'll do is
I let's say I take t1 and this is the connection string so I'll have to copy
this and I'd have to come back to my code and I'll have to replace it over
here so let's replace the deployment key and now I will have to remove this part
out of the planning this is not required right
but this we will remove and this is my deployment Keeler's all right next step
is to basically enter the container name now the hands-on basically expects us to
create a new container so let us do that so let's go inside overview and let's go
to globes so I have a container in telepods over here let's create one more
container and let's call it new right and let's give the exercise globe and
let's click on ok all right so my container is now ready and it's called
new let's name the container over here so my container name is new ok
everything else looks good so right now this is this is basically the connection
information for my MySQL database right now it is going to upload the data onto
my localhost - kale alright so let's see if I can upload a file now let's save
this in extra PHP let's come back to website let's refresh it once and now
let's choose the file which we want upload let's go inside let us try to
upload let's say database so this is an image which is called database let's
select this and let's click on submit so it says blob updated uploaded
successfully and new record created successfully so let us verify that in
our database so I'll do a select star from Nimes as you can see there is a new
entry over here which is 6 8 0 3 4 double 5 7 3 this is the name which has
been assigned by my website so this is basically a random name the reason for
that is that it could be that you are uploading duplicate files
and if you upload duplicate files you don't want the names to coincide so what
I'm doing is I'm assigning a random name to every file that I'm uploading all
right so this is the file which has been uploaded and we can check it over here
if I click on check list this is the file which has been uploaded and if I
click on this file now I can download that file basically from it says blob
not found and the reason for that is that I have to change the URL inside my
list website which is basically if I click on check list this is the list dot
PHP I'll have to change some code here I'll show you that but first let's go
inside the new container and as you can see there's a PNG file over here
now this PNG file I can directly download from a list file but in order
to do that what I have to do is I have to click on this file and I have to copy
this URL and I have to see what is the prefix over here so the prefix is this
so I'll have to copy this prefix and now I will have to go inside my code and
open the list of PHP code and this is the URL that I'll have to replace let's
replace it all right great now let's save it come back to website
this is a list or PHP let us refresh this points and now when I click on the
file you can see that I have downloaded the file automatically let's click on
this file and this is an image which has been downloaded now let's try something
else let me show you the image first and then
I'll try to upload it so what I'll do is I go inside downloads and let's say I
want to upload this machine file so this machine file looks something like this
ok now we're going to flow this machine file let's go inside our website fill it
through the file and now let's select the machine file which is basically this
this we will upload will click on open and we'll click on submit so it says
blog uploaded successfully if I go inside my container and i refresh it you
can see there's a new entry over here even in my database if i refresh there's
a new entry over here and now what I can do is I can
basically just go to my list refresh it this is the new entry click on it fire
gets downloaded click here and this is the file which was a pure so what is
basically happening over here is this file is now posted on plob and any of
you if you click on this link you will be able to download it but the problem
is that this website right now is only available on my local host and i have to
make it available to the world now how can I do that so first thing first let
us go back to our slides and see what is the next step so our next step is that I
want to create a MySQL database on as yours right now
my database which is being updated is on my localhost but what I want is that it
should be updated on my as your maestro database now let's see how we can do
that so the first thing that I will do is I'll have to go back to my as your
dashboard I'll have to go to services and I'll have to go into databases and
this is the database that I want to launch as your database for MySQL so
let's select it let's add a database I want to include it inside my demo
environment resource group the server name let's specify it as in internal one
right the admin user name let's specify it as enter the password let's specify
it as until at the rate 1 2 3 ok it says your password cannot contain all a part
of the login name no shoes let's name it as as your at the
rate 1 2 3 same is the case over here alright and where do I want to launch it
I want to launch it in South India the version is 5.7 which is great now let's
change the amount of specification that we have for this server let's click on
configure server I want a basic configuration I basically want one code
and I want the least amount of storage which is 5 GB Otto growth no I don't
want it backup retention period the lease is seven days which is fine and I
think that's it let's click on OK now server has been configured now let's
click on review last create okay so our username is Intel my password is 0 at
the rate 1 2 3 and now let's click on create alright
so my deployment is underway which basically means my MySQL database is now
getting deployed let's wait for some time let's wait it will wait for it to
be complete and then we'll proceed with our demo or I guess so my deployment is now
complete so I can just click on go to resource and I am in so this is my
database guys now in order to see or in order to access my database this is the
server name that I'll have to use so let's go ahead and select this so my
server name is this let's copy and let's come back to our command prompt and what
you want to do is MySQL - edge this is the connection string or the server name
next thing is I'll have to specify the username which in my case is this so
let's copy it paste it here next thing that I want is the password so password
is your at the rate 1 2 3 now this will be really it says client with IP address
is not allowed to connect to this MySQL server so let's solve that let's solve
that so we will go inside connection security and what I want to do is I will
specify my client IP which is this alright that's been added also guys turn
this enforce SSL connection off because right now we don't want to get into
making an SSL connection since this is the demo all right guys everything looks
good now let's click on save and once this rule has been said what we'll see
is that we should be able to connect to database
alright guys so it says it has successfully updated the connection
security and now when I go back to my command prompt and I try to enter the
same command with the password as your at the rate 1 2 3 you can see I have
successfully connected to my MySQL database on Azure now I'd like to do a
show databases right now you can see there are only the default databases
present so let us change that so i'll just exit or before exiting let us just
create an empty database which is images so we'll use images and let's create a
table so create table names and the names will be named space back at 20 so
we have created it and now let's just go to our code so over here there's my code
cards now I will have to specify the server name so server name in my case
would be this let's copy it this is my server name a user name is this and a
password is your at the date one two three okay everything looks good let's
save the file similarly in index dot PHP let's do the
same changes my surname is this let's copy it my username is this let's copy
it and a password is your at the rate one two three alright let's save the
file and now let's go to a website let's try to check the list let's see what it
shows us it shows us empty because there's no data in the azure SQL table
great now let's choose the file and now let's try to upload some random file
let's click on submit it says blog date it successfully a new record created
successfully let's check that so I just do a select star from names and I can
see that there is a new entry over here if I do a checklist this is the new
entry if I click on this I can download the file and this is the file that I
basically great now my website is updating data on to your
database and at the same time it is uploading data on to my storage account
inside the container which is new right phones as blobs new and this is the new
file which is basically just uploaded great guys now what I want to do is I
want to make this website public right this website is working fine over here
but I want this website to be used by everyone in the world now how can I do
that in order to do that I will have to go on to and we studied or we saw a
service which can basically just upload our website and not ask us to install
any software launched asked us to login into the operating system do some
configuration nothing it will just give us a dashboard through which we can
upload our website so let's go ahead and do that so the web the service that I'm
talking about is app services so let's go inside app services and let's click
on add so now it will ask me with resource group so I'll say demo
environment name of the instance let's specify it as until demo alright ok it's
already been taken so let's specify into alright this seems to be available what
do you want to publish I want to publish the code the runtime stack it's actually
PHP 5.6 this is the one what is the region that want to deploy it in we want
to deploy it in South India so that's like that one great let's change the
size of the server which will be launched so basically we are in dev or
test environment right and this is the minimum configuration is that we can
launch all right let's click on apply and now let's click on review and create
so this over here we can basically do all the change all the necessary review
as to what all we are launching so we just selected PHP 5 point 6 and we
specified the ah that's all we did right now let's
click on create now there is what will happen now is basically deploy so on
which it will install PHP 5.6 install Apache and then it will give me a URL
and when I go to that URL I will basically see a sample app right it is
not giving me access to the operating system it does not ask me to install any
software all that was done over here was we selected the runtime stack which is
PHP 5.6 we specified the server configuration and that is all that is
all we need to configure and that is all we get access to now once it gets
deployed I will show you there is how you can upload your code onto this
particular web app so as you can see it is already deployed now let's go to
resource and this is the resource guys the status is it's running right now and
if at browse this web app right now you'll basically show me a sample web
app which has been deployed right let's wait for that website to appear since
this dev app has just been launched it might take some time for the web app to
show the website but nevertheless it will show the website let's wait for it
to actually show us let me try stopping it and refreshing it once all right it's
going to take its free time so let's give it that alright so as you can see
hey app serviced well opposed your app service is up and running trying to take
the next step and deploy your code great so this is what you want to do now to
deploy your code you will have to go inside deployment center so let's go
inside that so once you're inside deployment center it's asking me where
is my source control right so my source control is basically on my local right
so but it my local is not a great repository as of now so what I can do is
I can just go on my coal which is over here 0 1 this is the folder and let me
just right click here and click on git bash right let me just make the text a
little bigger for you so that there is visible alright so guys now what I'm
going to do is I'm gonna initialize an enthe git
repository git init and now I'm gonna add all the files in the get filesystem
I'm going to stage it for the good filesystem so once this process is done
I'm going to commit it git commit - then and in that first now let's say
the files straight to the files are saved now what I want to do is I want to
upload this particular code onto my as your web app right so going back to the
portal I do have a local gate where I am a code checked in let's select this
option and click on continue now it's asking me which build provider to choose
don't worry about here don't worry about what is as your pipelines and what is
kuru engine basically if you select kuru engine you don't have to do any
configuration more on this we can discuss in the further sessions but
right now just select the kuru engine and click on continue and now it says
your local git repolish you are will be generated upon completion branch would
be master app so this is build service okay let's click on finish so now it is
basically going to set up the get environment for me on the web app so and
what I will get is basically this URL which is my gate URL so this URL I will
be using to upload my code but before that I will also have to set the
credentials for my gate system so to set that I click over here FTP your
credentials and I go to user credentials and I can specify anything over here
let's specify a username let's say the username telepods and let's say the
password is e which is saving credentials for the success great now
what I can do is I can just close this and now I just have to go to my resource
I have to go to deployment center and now I will just take the kit URL
over here right this is a get URL let's copy this and now get the move and let's
try to paste this link now yes so this was the syntax git remote add as your
right and now what I want to do is I want to push my code on to a jar so git
push is your master I'll hit enter and I will ask me the credentials so
credentials is Intel apart and the password is Intel at the rate one two
three Oh indication field I think I forgot the password no shoelace so I can
just specify it again so Intel at the rate one two three and one two three
inches okay inches are saved let's go back here it pushes your master
and now it should basically push my code on to the web app now it will take some
time over here guys don't worry about the time it will take around a minute or
so to applaud your code so let's wait for that time and let's hope everything
goes well all right the process has started so it
is updating the branch now copying all the files great
so my code is now uploaded on the app and it says it will restart in ten
seconds great so meanwhile what I can do is I can go to my web app and I can just
refresh this and as you can see my website is now available over here
alright if I click on check list you will see an error and I will tell you
why that error would be there it says connection failed client with IP
address is not allowed to connect to this MySQL server how can you solve that
just go home and go to your database go to connection security and allow this
allow access to your services just click on on and as a security feature just
remove the IP address that you used earlier and now let's click on save so
now my security settings we get upgraded and then I
would be able to use my website on the web app alright so my settings are about
to be abated let's wait for that time and then we'll go ahead
alright guys so my data is has been updated successfully so if I go here
back and I click on refresh you can see the list is now being generated let's
try to upload a file guys let's try to upload let's try to upload this word
file ok let's click on open and let's click
on submit it says blob updated successfully let's check that lets go to
resource group let's go to our tour it account let's go you know and this is
the docx file which was just updated and if i refresh the list i can see the docx
file over here as well if I click on it I be able to download it in guys with
this we have successfully completed our demo but let me show you a very awesome
thing that comes with connecting get to your data right so now I can just simply
go to my index of PHP and let's say I want to change this heading okay so
let's try to change C heading I'll just go down upload towards your blob and
let's add a little bit of text over here let's say welcome to telepath ok let's
save it come back to work it terminal let's do a bit status you can see that
index dot PHP has been modified great let's scale this file let's commit this
file updated index great and now let's push it
so get push will again take a minute or two and then our files will get updated
on the web app automatically alright so my code has been updated let's go
ahead and check it so let me just refresh this website and as you can see
my code has successfully updated over here it says upload to a job blog
welcome to in telepath and this is exactly what we changed all right so
guys with this we have successfully completed our demo let's summarize what
all we did so let me come back to my slides so we demonstrated the website on
the localhost fine then we created a new can
and then we uploaded all the blob files over there after that we also deployed a
MySQL database on Azure and the localhost website was then able to
insert data onto the mysql database on Azure and then finally we also deployed
an is your web app and we deployed our website using the local get method and
finally we also checked if we updated anything and we pushed through a get
method the files were successfully being uploaded onto mine as I better right
so with this guys we've successfully completed our demo our next topic is
basically quiz so let's go ahead and see some of the questions that we can answer
after attending this session okay so our first question is which of the following
is not a platform as a service and you options are MySQL database 4 is your app
service as your VMs or is the answer none of these okay so you wanted to
pause the video and think about it so did you guess the answer so yes the
answer is your VMs the next question is can we deploy as your VM without a
virtual network so the options are yes/no or none of these so what do you
think so yes you get it right the answer is no right let's now move on to a third
question which is what is Azure Active Directory useful is it a monitoring B
Identity Management C automation or D none of these now guys I will not give
you the answer for this particular question so you'll have to answer it to
me in the comment section below right alright guys so I hope this video was
useful to you and if you found it helpful please like and share this video
and also subscribe to our in telepath channel so that you don't miss out on
any updates from us right if you have any queries with regards to what was
taught in the video today please complete it down in the comment section
and we'd be happy to help you out right so keep learning with Intel apart I'll
see you in the next video have a great day ahead and good bye
