Introduction To Google Cloud Platform Fundamentals Certification | Simplilearn

[Music] hello and welcome to the google cloud platform fundamentals course offered by simply learn this is lesson one introducing google cloud platform in this lesson we're going to take a look at the very underpinnings of the google cloud platform what is today about today is an introduction to the google cloud platform and you may be coming from another cloud platform or you may be just curious about what is cloud and in particular what is google's offering of the cloud and this that is what this course is all about the the course will cover a lot of material in terms of breadth and we will not go into a lot of detail on the particular items that's reserved for some other courses so my objective for you is to explain what the terms are and what the abstractions are and then we'll have some labs where you can do some additional exploring and get a feel for things so you're going to hear about the different parts of the cloud and you'll experiment with some of them in in some labs what i have discovered and i've worked in various clouds is that even the terminology is not settled so one particular vendor may use a word in a certain way and another vendor may use it in a different way so one of the things i hope that you will leave with is an understanding of what google means when they say particular things for example platform as a service or infrastructure as a service now we're going to take a brief look at the overview of the google cloud platform this will include a brief tour of the hardware underpinnings names that they give to the various abstractions including things like what does it mean for a resource to be global what does it mean for a resource to be regional and what does zonal mean these abstractions live on top of the hardware level and it's important to understand that there is a distinct difference between the two so why choose a cloud platform to begin with well it's one of the things that google's mentioned recently in a presentation they said that if you look at the cost of a computer if you look it over its four year life that the majority of the cost is energy and google is very very focused on making their data centers very energy efficient and they've invested a lot in security and extremely high-speed networking and i'm convinced that companies and especially small companies just can't match that investment so there is a broad movement to the cloud people are realizing that it's going to be better to rent instead of buy computers and if you look at the google cloud platform they've had years and years of experience building out a cloud platform for their search for their gmail for their very you know youtube for their various offerings and they decided a while ago that they wanted to make this available to all of us so they're offering this so that we can build and test and deploy applications on their platform that scale just as the google application scale computing solution there is no one ideal way of storing information so google offers a lot of choices several choices in computing their choices in storage their options for big data one of the things that google is really leaning on now is machine learning so by opening this up we can work on this highly scaled global compute infrastructure and build our own applications that run alongside of google's apps and as you know google is in the business for a long time so they've had all that time to perfect their technology so so there's a couple words here that i wanted to find so that we're all clear a data center is a facility it's a building or multiple buildings that have their own power substations they have high security there's perimeter fencing there's all sorts of codes that you have to use to get inside then very very few people actually ever get inside a data center it's highly secure and this is where all the computers are this is where the networking is this is where you know the the low level compute power of the cloud is is residing they have different data centers and they're opening many more we'll talk about that a little bit later another really critical thing about the google infrastructure is they have a very well designed very well thought out networking underneath all of this they have a global meshed redundant fiber optic network that spans the globe and i don't know you may have noticed in the in the news recently that they announced a connection with asia a new one that they've just put the fiber down so they're actively building new connections a point of presence is the edge of google and the rest of the internet so if you're have a computer and you need to have packets go to google they'll transit from your computer through the internet and they'll hit a point of presence and that's where your packets your connection will get on the google fiber and find its way into the data center and they have over 110 edge points of presence now in 33 countries just a little while ago it was i think 80 so they're expanding those quite aggressively and if you look at information flowing out from the data centers to you or your clients there's edge caching so these are also residing right at the edge of the google fiber and they will cache your information to make it a faster interaction with your customers and we'll talk about caching a bit more when we look at some of the particulars of the compute infrastructure this is actually one of the more important slides of the whole session this is the first one where we're going above the physical so when the physical we talked about actual computers data centers the fiber cables the switches the network cables and so on we're going to leave those behind and now we're going to raise ourselves in terms of abstractions and the three terms that are crucial are global regional and zonal so these are essentially the lowest level of idea that we're going to be working with so let's start with zone and a zone is roughly equivalent to a data center and zones are where you have your compute power you have your disks and so on in terms of designing a system especially for a high availability system you need to consider that a zone could fail so it's best to have compute power in multiple zones and balance the request across them and and we'll talk about that particular detail later the important thing to know is that some resources in the cloud are anchored in a zone for example a compute engine instance is a zonal resource the next abstraction is region and a region is a geographic area and it may it will contain multiple zones and an example of a resource that's regional is a network load balancer so you can set up network load balancer and then it will send the request to different zones and it will be aware of the health of all the computers so if one does fail it will stop sending requests there and it'll send it to other computers in different zones and the third concept is global so there are some resources that are global that span all the different regions for example when you create a network it is a global resource so your compute engines in a zone will connect to a network which is global and essentially they can talk to any other computer connect to that global network so you can easily configure a system where everything appears to be on a local area network but in fact it's a global network so and that points out another thing that's kind of important about moving to the cloud is some of the things that you know based on your experience with the real world you know i say the real world the physical world of computers and so on some of those ideas you're going to need to leave behind because it's different some things are quite different when you get into the cloud and i'll be sure to point those out as we as we work through these these slides so again let's look at cloud regions and zones and global zonal resources operate in a single zone an example again is a compute engine instance and its associated disks a regional resources are in a geographic area for example u.s central or asia or europe and a region will have multiple zones and then global is spanning the whole planet so you may want to have a single ip address for your web application that would be a global resource and google provides load balancers across regions and you can have regional load balancers that balance things across zones google is really working hard to pioneer efficiency and performance in their data centers they as i said that they have determined that the majority of the cost of computing is energy so they are focused very very diligently on reducing the cost they've been carbon neutral since 2007 they're investing heavily in wind and solar and they have been they're pioneering efficiency so they're a really good engineering company that's one of the things that you need to keep in mind as you learn about the various parts of the google cloud platform they're good engineers they're very proud of their engineering and they do incredible jobs with what they're focused on recently they've applied artificial intelligence to optimize each data center for energy use so they'll measure ambient temperature and other factors and adjust all the parameters for a particular data center so it has it consumes the least amount of energy possible there's also a trend that we're very very aware of and that is that the price of computing is constantly going down and google is very focused also on passing the savings along to us and they've come up with several ways of monetizing the savings one is sub hour billing so the idea here is that if you ask for a computer i'll give you a specific example you can ask for a 32 core 208 gigabyte computer and run it for only 20 minutes you know they don't round up to the next hour and the cost for renting a computer of that size for 20 minutes maybe you're checking out some code maybe you've got a caching algorithm or something like that you want to test well that will cost you about 50 cents so it's really really economical to use for experimentation and getting things set up and and the so and so on on the other side once you get your software stable and you're ready to deploy it google has what's called a sustained use discount so as you use the computing infrastructure throughout the month they will give you an increasing discount that basically goes up to about 25 percent so they're rewarding not just temporary use but they're also rewarding sustained use there's another feature they came out with which is custom machine type so you can dial in exactly the amount of cpu power and memory that you need for your application and we're hearing reports that that will save customers about 10 or 15 percent and the fourth innovation they've come up with is what's called preemptable instances so these are this is like scavenging the for free essentially an unused compute power so if you build a system that that anticipates a computer going down and your software can understand that and reschedule it on another computer you literally can save fifty percent the price is half of the same machine in non-preemptable mode so again there's four ways that they are currently providing uh they're passing on the savings that they're finding from declining prices of computers google has made numerous contributions to cloud computing in fact they've been pioneers of cloud computing since the very beginning all of their major products including search maps youtube and other features like that run on their own cloud computing infrastructure they've been running containers for years and now they want to provide that understanding that intellectual property and the ability to scale applications to google scale to us google's well they're they're focused on innovation and what they have learned is that it's it's to their benefit and it's to the industry's benefit to make things open and as they have designed innovative technologies to to support their own infrastructure they have released technical papers describing what they've done so they've had names like dremel and colossus and things like that and once they've published these papers the open source community has noted them has read them and frequently will implement the ideas uh as an open source uh product and the the google has really taken note of that and in fact they are releasing a lot of their new innovations completely open source a notable example that we'll look at later is something called kubernetes so they've they've learned how to run containers very efficiently on their infrastructure and they decided to rebuild all of that add some new intellectual property and release that as a open source project that we can all join we can all use open source licensing again that follows with not just kubernetes but a lot of the low-level libraries for connecting to the google cloud platform they release those open source and again it helps everyone it helps google and moves the industry forward the publications i've mentioned already for example you can look on papers on how they've built their their low-level data stores what dremel is about dremel is about high efficiency querying of data and because of this interoperability and the fact that they've released things open source we can now think about even building things across multiple platforms multiple cloud providers so again kubernetes is the open source container management software that google's released google has a product called container engine that uses kubernetes and adds some special google features but you can run kubernetes on your own machine you could run it on amazon you can run it on openstack you could run it on digitalocean you can run it anywhere and so by doing that google is opening up the interest of the whole industry about containers and google's gamble i mean it is a gamble they're they're releasing their key intellectual property they believe that they will be able to provide the best platform the best product for running containers and by reducing the risk to all of us by having kubernetes be open source everybody wins and google will be in a good position google is not standing still and this slide is meant to show you a bit of where their vision is going so the first wave we've had is was co-location and i literally on my desk have a computer that i sent to a co-location place a long time ago and i've since requested it back so co-location is where you either send your equipment and it's located in a managed environment or you rent computers that are in a managed environment and you can imagine your computer sitting in a rack and you can connect to it and the advantage is that you've got that vendor's security and you've got their power and so on the next wave that happened is virtualization so here we're getting into virtual machines and we're starting to get the virtualization of various parts and this is essentially where we are right now we're right at the end of the second wave and we're moving into the third and the third is a completely global elastic cloud and once we move here we'll we'll not worry about virtual machines we'll just ask for a bit of compute power and we'll get it you know and we'll probably get it very very quickly so we'll be able to scale up we'll be able to scale down and we'll be able to match the compute resources that we consume to match the the needs of our software as requests go up and the requests go down the fact that you can have a computing environment that tracks with your usage is is actually it's it's more significant than you might think and from a technical perspective it has huge benefits from a business perspective just think about how you have to in the old days you'd have to invest in a large number of servers if you had a startup company and get your software ready get it all hosted and you'd have to size your servers to en to the anticipated peak load and if you don't get the peak load then you're basically wasting your investment but when you have an elastic cloud and you use compute resources as you need them you're only going to pay for those so from a business perspective you don't have the initial investment the huge upfront cost you can just have things track with your usage and it turns out to be a much much easier environment to manage okay in terms of definitions this is another one of those slides that's significant in terms of this you know kind of defining what is infrastructure as a service and what is p platform as a service these are both completely uh valid things to be using and you can even use them at the same time but they're different philosophically so on the infrastructure as a service side the main offering from google is called compute engine and this is where you have raw compute power storage networking and so on and it's the closest analog to what we're used to you know in our physical world so if you think about a machine a computer that you've got and maybe you're building it yourself or you're ordering the parts or you you order the complete computer in the computer is the motherboard is the chip is the ram are the disks and so on and so compute engine and infrastructure as a service is the cloud equivalent of that so you literally can go and say hey i want a particular computer i want to have this much memory in it and all we're doing is going to a web page and clicking options and we're building an abstract computer this way and because we're essentially taking things off the shelf we're saying okay i want a particular disc it's the size nobody else can use it it's ours well i want a cpu with so much power it's our cpu and nobody else can use it we're taking it off the shelf essentially so the fair thing to do is for google to charge you for what you've requested and so that's the pay for what you allocate even if you don't completely utilize it you can build up a computer and they'll charge you per hour or actually per minute for all the resources that you've requested platform as a service is a different philosophy here you are supplying your business logic and google is taking care of a lot more of the details they'll give you run times it's a kind of a sandbox environment and they will scale things for you automatically so the fair thing to do is to charge you for the resources you use at that moment so if you need a hundred app engines to support your your request or if you need just one google will scale that up and down and you'll only pay for the amount of compute power and resources that you're using minute by minute okay so across the whole cloud platform these are the major categories there's compute power there's storage there's a bunch of products focused on big data and the new innovation area and where google is working very very hard is machine learning and we'll learn about that near the end of the course okay so talking about compute here we've even got three options so we've got app engine we've got container engine and we have compute engine app engine is the most constrained environment you do not have much flexibility in terms of what you can do but the benefit is by constraining your application to only the business logic you really only write the business logic google takes the responsibility for everything else making sure the software is updated patching scaling taking care of all the infrastructure again on the far right side this is the infrastructure as a service you're going to be building your computers you need to worry about how big they need to be and do the scaling and so on but google still takes care of a lot of things they will actually move a virtual machine to a different bit of hardware as it's running in case they need to do maintenance so there still is a huge benefit even at this level of you know defining computers to have it run in the cloud container engine is kind of in the middle it's a level above virtual machines uh container engine is all about very lightweight docker containers coming to life very quickly or scaling up or scaling down so again there's no one ideal choice and also in storage you can see the same thing there's no one ideal choice for storage we've got big table we've got cloud storage that we're going to look at in detail we're going to look at cloud sql a relational database and cloud data store a very unique data mechanism that scales alongside of app engine on the big data side and again these are all completely managed services there's bigquery it's extremely high performance fast sql language data processing system it's great at reading log files and giving you an idea of what's going on in your application it's great at processing all sorts of information and it will even ingest information as it's occurring so you can think of internet of things devices sending information to bigquery and you can be doing near real-time analysis of all that data pub sub is a very simple idea it's about sending messages to people who are interested in what you're saying and it's google scale it'll take you know million messages per second and it's also a great decoupling tool in terms of architecture data flow is very innovative this is essentially generation two of map reduce i'm told that people at google don't actually use mapreduce anymore they all use dataflow and dataflow incorporates the idea of mapreduce plus a lot of other things and again we're going to have some detailed slides on all of these coming up dataproc is essentially hadoop managed by google the value add by google is it can spin a cluster up in 90 seconds and it can resize it as your workload is going so it minimizes your cost and it's designed to be compatible with hadoop processing outside of the cloud or in other clouds so you can just bring your hadoop processing run it on dataproc and you can take advantage of the google platform datalab is a notebook like presentation of big information so you can have it execute queries you can do graphics you can put it into a notebook like format and then give it to someone who wants to understand the results of all this processing and the machine learning side this is really i think this is particularly exciting we've made some major advances in machine learning you probably know that google has a machine learning system that beat the world's expert in the go go game and what you do is you build up a set of definitions in machine learning and then train the system to do a particular task so google has done the training already in some areas they have a vision api that can recognize uh things in a picture it can tell like if your pet is a cat or a dog which is pretty interesting there's general machine learning so you can build your own models they designed a speech model so that where it can recognize speech and there's another model for translation between languages so you can build your own models or you can take some of these existing ones and incorporate them in your application okay so now we're going to have a lab this first lab is going to guide you through the steps to set up a relationship with google cloud you're going to set up an account and you're going to create a project projects are the container for everything that you do in the cloud so these will take you through the steps you're going to need to have a credit card and the credit card is not so that you're going to be paying for things it's for your so that google knows your identity it wants to know exactly who you are and then once you do that you will select an option for a free trial and upgrade your account you're going to get three hundred dollars of credit that's good for 60 days so the the course that you're taking now probably will use maybe two i i'd be surprised if you more use more than five dollars of that 300. so at the end you're going to have a nice account you'll have money left and you'll be able to do a lot of experimentation in the cloud so i'm going to bring up another window here and i'm going to go to my console okay go to this little icon in the upper left hand corner with the three horizontal lines if you click on that that pops out a tray the different areas of the cloud and what i suggest is you go to billing and you should see that you've got a billing account see here billing management for cp 300 there's it's linked to an account called randolph kale your account your billing account will be named something else and you can then manage billing accounts you can see that there's a billing account the account id that it's active and here it says i've got nine projects associated with that and i can say my projects and see exactly what they're connected to so again i'm already in a relationship with google so i i don't see exactly what you're seeing but see if you can go through those areas and see if it's if it's active and so on is it necessary to write down the credit card you need to enter the credit card yes the credit card is required by google not so they can charge your credit card but it's used for identification and you what you do is you enter your credit card you set up you say yes please bill me that means create a billing account but then you immediately say i will accept the free trial you get 300 of credit which is good for 60 days if at the end of 60 days or when you've used up to 300 your credit card will be charged so do understand that so you may want to turn things off we'll talk about that how you can make sure that you're not going to get charged the easiest way is simply to delete your project you can just go in and say i need this whole project deleted and then there's nothing active so you will not be charged okay activating the cloud shell let's take a look at that i'm going to go to my home area on my screen and the cloud shell right here is this icon right here so you notice in the upper right hand corner there's a little picture of me that identifies the account the relationship with google so that's me with a particular account i actually have about five accounts with google so i use this little photo to remind me which one it is and if you go to the left you'll see this little icon that's the cloud shell so let me pick um cp 100 class let me switch over to that and then i'm going to cl oh that's interesting the cloud shell is not working okay they they made a major change to the the user interface yesterday in fact this dot down used to be over on the right um and i'm not sure what is going on i'm going to go off mute and see if i can figure it out first let me explain what's going on the cloud platform is comprised of many different services each of which has an api structure so that you can access it in fact everything is communicated through a rest api and they don't turn on all of the apis when you create your project and you first set up a relationship because they they just want to save the the compute power to set up credentials and so on so i'm guessing the thing that you need to do is to go up here and click on this icon and bring the tray out and then go down to compute engine and click on that and what you should see then is not what i have you know i see create an instance you will probably see please wait we're setting up the compute engine api and so once you do that they google will set up the infrastructure they'll set up the credentials they'll set up a bunch of things like that so i notice that once i did that notice my cloud shell is now open that's what's happened they've made a they've reset some things so you need to go into the compute engine area just by going into that area should turn on the activation of the api it should turn on the creation of credentials and then it'll give you the cloud shell that i'm now running that you can see in the window so dale please try that and let me know if that works and i also want to add do not get concerned about the amount of time we're spending on this this is an absolutely crucial lab this obviously gets everyone set up to run all the other labs so we will take as much time as required to make sure that everybody's up and running you have the accounts properly set up and so on and we'll work through the details with you right so the last thing you should be doing is is creating a project we're just creating a project called cp100 i'm showing you on the screen that's the last step of this lab this lab is overview and then introduction and then register for the free trial which is five steps and then create a project which ends with the third with step three on the final slide let me go bring up a console and show you so you should be here there should be a drop down right here notice that first of all there's the the icon that brings the tray out and then i have some of these pinned i can tell tell you about those later but right here i've got a project and if you click on that it says create a project and this is where you put in your your name and you get your id and so on so that's what you should see let's review the key takeaways for this lesson [Music] this brings us to the end of lesson one the next lesson is getting started with google cloud platform

Channel: Simplilearn

Views: 167,985

Keywords: simplilearn, training, tutorial, certification, course, curriculum, Free resources, google cloud platform, google cloud platform tutorial, google cloud courses, google cloud platform training, google cloud platform videos, google cloud platform fundamentals, google cloud platform introduction, google cloud platform pricing, google cloud platform machine learning, google cloud platform api, google cloud platform sql, google cloud platform php, google cloud platform demo, 2017

Id: VUWB807TJEg

Length: 36min 41sec (2201 seconds)

Published: Thu Aug 25 2016