Taste of Premier - Azure and the Internet of Things

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
[Music] [Music] so by the way a nice beard there i see you got a beer seemed like a thing to do these days this is my covet 19 beer yes i was gonna say yeah yeah as long as i'm working from home and uh i don't have to go into the office i'm not shaving unless i have to nice nice so so man internet of things that's uh just for those that aren't aware of that term and you all should be by now i'm hoping um internet of things is just that it's it's devices connected to to a network that aren't necessarily standard computers right not things that you would think of as computers yes right yeah typically there's a microprocessor in some way shape or form involved but it's uh i mean it could be anything from thermostats to your vehicle um you know that to something a little bit more sophisticated which in in some in additional complexity involved when it comes to talking to things um like an assembly line and think about you know barcode scanners and robots and industrial controllers there's a whole nuance of of um things that you have to or things have to be asked about that you need to to be able to plug in and and extract data from those devices to be able to do interesting things or you know and anticipate whether or not you're going to have a machine breakdown or alert somebody if there is a machine breakdown um you know because all of these pieces of equipment have different protocols they talk different languages sometimes some of them still today use serial cables so so so we have some some additional capabilities that you can actually put um attached to something that might be old that can make it appear to be new again when you when it comes to data collection and ingestion using using some capabilities we have that can run on uh you know any piece of computing hardware whether it be a raspberry pi or a laptop or something a little bit more powerful than that so yeah and i was just thinking about all of the internet of things devices that i have in my house right right i've got a ring doorbell iot yeah i've got a nest thermostat that's iot right my security system has webcams that's iot um i've got uh a couple of alexa devices that's iot i've got cortana devices that's iot um a buddy of mine just bought a lg refrigerator uh that's certainly iot yeah you know they they can not iot isn't just involved in you know consumer devices but think about you know power plants um utility generation you know gas distribution lines you know these environments that that um you know are are large in scale and scope you know the utility grid in a particular you know any additional or particular municipality would be massive and it would be cost prohibitive to put a you know an i5 based laptop or computing device every spot that there's a gas valve or switching station so you know it's common for these or if you know for utility companies to be use these little power devices to capture data um you know maybe you just periodically wake up and send you know some some information um you know likewise um you know you see it being used pretty heavily today in a smart building space where you know in today's climate we have a lot of commercial buildings unoccupied and you know think about lighting and hvac um you know control systems to be able to try to reduce the not just the carbon footprint but the energy usage for for financial reasons of of commercial buildings and in real estate so so used everywhere um i work with a lot of customers in the industrial front and then also um i have some customers that use iot devices um particularly we'll talk about today and then the edge space um actually on very large pieces of equipment so thank you uh you know dump trucks with tires is you know taller than you so uh you know they you know these types these customers are actually putting iot edge devices underneath the seat of these vehicles to be able to capture telemetry from the vehicle you know warn the driver if they're doing something that could damage it or cause you know harm to a multi-million dollar piece of equipment and then also you know helping the owners of these pieces of equipment be able to optimize the efficiency of it because you know they're expensive their downtimes expensive so it's it's um you know it's a it's a good application for iot stuff to be able to help a real world application of iot stuff to be able to help a customer you know save money ultimately yeah i think we'll quickly get to the point where there are more iot devices on the internet than there are computers we might already be there yeah yeah i mean it'd be interesting to see some counts on that but i bet you're right um and in fact i wonder if that's to a large degree why they're uh you know then ipv4 you know just thinking about all the ip addresses getting gobbled up but i have to imagine that that's not helping yeah yeah but that is certainly an issue that uh ipv6 solves right yeah yeah all right all right so what's up a little stuff to show us yeah so let me uh go ahead and share my screen here i got a couple of slides i know slides are boring but they do illustrate some points and some spots that there are some things we can talk to before we kind of dive in and i love your slides how many internet of things devices are there on the millennium falcon you know i i don't know you know i've got uh get quite a few phasers and those are all uh all iot based the entire weapon system yeah yeah pretty much all right so um there are a couple different ways that you know we talked about smaller devices like thermostats and um you know ring cameras and those are i don't want to say they're unintelligent but they have low power inexpensive compute microprocessors in them so they can be sold to the likes of you and i at low price points but there's not a whole lot of capability embedded in them to be able to do processing so they tend to send data up to some place you know typically some cloud provider to be able to allow you to control them so like my thermostat i can pull up an app on my phone see what the temperature is and and remotely control it or toggle it and the messages message flow goes from my phone to the cloud and then down to the thermostat um you know so that solves a couple of challenges and we talked about the ipv4 challenges so it allows uh those devices to be installed behind you know network address translation devices and so it kind of solves you know some security challenges and some technical challenges using using that mechanism but that doesn't work terribly well is when you think about um uh you know a a mine site you know the middle of nowhere with pieces of equipment running around on it that need to have some uh the similar types of intelligence running on them to be able to inform the operator about things or you know make the vehicle do things or in a automotive plant or a parts plant or you know widget plant um you know that has some pretty strict security guidelines on on what can access the network that very expensive and dangerous you know potentially life-threatening pieces of equipment you know with robots swinging around in the cell you don't want them to come back online when an operator or you know maintenance personnel is in there so that's where we we here at microsoft have taken some steps to take a lot of capabilities that we have in the cloud onto what we call the intelligent edge and intelligent edges really comprise a number of different services that can be deployed on on compute devices and i'm using the word compute devices generically because we have a lot of options here including a raspberry pi i've got a number of them on my desk sitting next to me that all have iot edge running on them or it could be you know really high-end um you know i series machine or one you know something you know an intel based machine amd based machine or something with you know a gpu to be able to do video processing so it's kind of you know spans the spectrum on what type of compute devices is supported on and we make the this services capability available to customers um largely free of charge there's some components that are are chargeable but it is dependent on on services we have in the cloud so we allow you to extend that and and run either in a disconnected state or in a state where you need to make decisions really quickly with pieces of equipment that are are physically attached to you without having to rely on you know internet connectivity or um you know some utility contractor that digs up a cable and shuts your plant down so so edge really solves a lot of problems and challenges that kind of come into play when you start thinking iot in in non-traditional senses so so how how does iot edge solve some of these problems well what it what it comprises of is um what we call the iot edge runtime that gets installed or deployed on any of these compute devices you know whether it's arm based um you know or or whether it's a um you know an intel based uh you know intel amd based machine um you you installed this runtime component onto the the machine you know it could be windows or linux based um and what it provides you is the ability to to take data from devices that might be physically attached to it you know in the case of uh you know pieces of equipment i was talking about earlier um you know there's a can bus that's inside of the vehicle that that's provided you know specialize in vehicle network that we can actually pull data from uh there's an ethernet in in the vehicle you can pull data from you know in the case of a of a manufacturing plant it might be um you know robots there are programmable logic controllers who want to pull data from they actually run an assembly line so we can actually pull that telemetry and process it locally on the edge so typically um in that data flow we talked about like with a thermostat we have data sent from the thermostat up to the cloud some decision is made and then it um you know whether you know you it's you on your way home wanting to your you know inc or reduce your temperature in the summertime to make it cool when you get home so so the idea is that you can actually take all of those instructions and actually put them on a compute device inside of a plant or inside of a piece of equipment um without requiring requiring any dependency on the internet connectivity um so so we have a service in azure called iot hub so devices can actually talk directly to iot hub so and in fact actually you know we have customers who are doing this they have intelligent devices like thermostats that can talk directly to iot hub from you know within your home so this is actually decoupling that and allow you to run it you know run it locally so the iot edge runtime actually allows you to manage what actually is running on that edge also remotely which is a really powerful story because um when you think about you know typically when you have some bit of functionality that might be processing data from a robot let's say running in a plan it's it's actually you know now once you had deployed the solution um it becomes a challenge to go and touch maybe potentially a thousand different iot edge devices that are physically attached to robots to update this what the software is doing so the iot edge runtime also allows you to to remotely manage it and what the functionality is that's running locally from the hub and we'll we'll see that a little bit when i show a demo um so this is actually this would be my last slide i'll let this kind of build out because it's a kind of a complicated slide but um this shows what the iot edge really is and and if you you know we've talked about containers in the past um you know on other shows and i think you've had others kind of talking about what that uh you know what that concept is um so you know container a way to encapsulate and easily transport bits of logic whether it be an application you know or machine learning module that maybe you you train a machine learning job in the cloud and you want to run locally where the data is actually flowing in um so iot edge allows you to remotely manage the individual services that are running on the edge and those services are deployed as containers so the edge runtime the iot edge platform actually is in itself a container and we'll see that when i actually connect to my edge device so there's a lot of security capabilities that i kind of glossing over a bit on kind of how that iot edge device authenticates to the cloud um and there's there's some a lot of complexity and options there to be able to say hey i'm really an iot edge device for you know customer x manufacturing widgets um so that attestation is a you know something that that we have some some complexity to be able to to make sure the complexity purposefully in a plethora of options to make sure that devices is you know for the customer that that it's intended um in the you know so from from the um the iot hub you know this is the cloud piece um there is a concept of of a device twin which is uh kind of a digital representation of the of the iot edge like what its current state is the last time it checked in um as well as a module twin and what a module twin is is a digital representation of these containers that are actually running down on the edge so you can actually see you know maybe if you had a container that was connecting to a you know a local database that was running on the edge maybe a local sql server so you can actually control connection strings up in the cloud and it will actually get sucked down and presented to the module that's actually running on the edge so the benefit of all this plumbing is that you can actually um deploy stuff which is up here on the right down to the edge and have it run uh irrespective of what's happening with your internet connectivity or links to the cloud so including things like our ai services so we have you know azure machine learning azure stream analytics azure functions these are actual these are capabilities that run um you know in the azure platform that you can actually uh you know run locally on the edge um the and then you can also build your own applications and and run um you know run it run them and deploy them as a container that do other specialized things you know some of my industrial equipment customers might have a module that knows how to talk to you know a particular piece of industrial equipment that they purchased so you can actually control and manage that all from iot hub so let's look at what that actually looks like so would you consider the mars rovers iot devices yeah yeah you know the actually that might actually be more akin to an iot edge device because you send it a bunch of instructions they get queued and it just you know does this thing and then phones back yeah so kind of so that would mean that we have iot devices in the interstellar space there's a voyager you know they communicate that's true yeah yeah that goes way back that's in the 1970s yeah yeah there you go all right so i will make this a little bigger i kind of increase it a little too too fast here okay so i'm in a iot hub this is actually a hub that i've deployed in my um in my subscription and um the hub can actually as i mentioned earlier can actually have devices that connect directly to the hub things like thermostats or uh you know ring doorbells would be examples you know so you can it can actually have devices that connect directly to the hub and you know hub iot hub is a core service in azure that to basically facilitates massive numbers of devices communicating and sending data but it also supports edge devices being connected to it so i have um this opc test device and you can see it's showing me how many container how many modules these are containers that are actually running on that device and shows how many clients are connected to that um so so there's some this is actually because of that twin that i talked about the digital representation so there's updates that are being sent because my edge is actually able to talk to the cloud um and you know that internet link is is up all the time because it's running on a virtual machine that i i have uh also hosted in azure that date is fairly real time so when i drill into this um so these are the modules that are actually running on my on my iot edge device and there's two of them that are actually the iot edge itself and they have a dollar sign in front of them edge agent and edge hub so when i go to um let me just go ahead and pretend i'm going to change one of these and if i go to this runtime settings so in here this is actually telling me that there's there's two core containers or images you know container images that are used to actually represent the iot edge runtime so i've installed iot at one time on on this vm but now we can actually centrally manage the version of the container of the version of iot edge just running on this on this vm simply by going to the iot hub and saying i want to run you know instead of version you know 1.0.10 and i'm running a release candidate you know maybe release candidate 3 comes out so i can go in here and literally update this and this will send an instruction down to my edge that'll actually tell it to go out to this container registry and download a new version so this is where some really power powerful capabilities comes into play to be able to manage not just the edge runtime itself but maybe code that you've perhaps written yourself to be able to do interesting things with a robot or piece of equipment um you can actually manage that at scale now i and i say at scale because i want to emphasize that because right now i'm just doing it individually i've drilled into a very specific device and i'm telling it i want to change a version that's actually running on that device so you can actually do this in bulk saying i want to change all of my edge devices or maybe all of my edge devices that are in europe but to you know to run this particular module um or you know update of iot edge [Music] so so you have some you know some pretty great levels of sophistication here to control that and i've got a lot of parameters that i'm passing into that that i i will um kind of gloss over for the moment but what i really want to show is these um these three containers that i have running and this is actually part of our iot edge connected factory capability that we have that some industrial uh you know some abilities for for you to be able to easily connect up to industrial devices so if i want to flip back to my slide deck that i think i just closed but i had one more important slide to show and it um it talks a little bit about what this opc uh letters mean in it so um opc stands for olay for process control it actually goes down to the um i thought i thought it was other people's containers yeah that would actually be really appropriate if it were called that so this is this is a is actually much like a container though it is an open standard and there are lots of actually i should should even go as far to say all industrial equipment companies have devices that can talk um in in a manner to be able to interface with opc uh software um so because and it kind of goes this is a triangle kind of showing you know where you have you know the the most number of devices you know sensors motors actuators and so forth those usually talk over proprietary industrial networks and this is where it gets really hard to take something that's been running in a plant for maybe 20 years that's producing widgets and be able to you know enable it to send data up to the cloud to be able to take advantage of machine learning and try to you know predictive maintenance and you know predictive failure type capabilities to exist in the cloud um you know because these these protocols are old um you know but they're still heavily used and widely used in the industrial space so the only way to be able to interface with those is they have something like an edge device that's physically sitting in the plant and has the capability to talk you know some of those protocols um and then you know kind of going up the stack usually you have something called the programmable logic controller and this is um a kind of a small industrial computer of sorts that has uh physical wires getting you know connected to it to be able to control uh you know what happens when an operator hits a start button on a piece of equipment at a station so kind of have this this stack of stuff you know sensors the plc's that can actually talk to those sensors over those industrial networks and then um you know you there there's this this long-standing uh us you know set of interfaces that customers have been using for decades opc classic and opc ua which are are ways to be able to connect to these programmable logic controllers with software and by software i mean you know these programmable logic controllers are pieces of hardware very specialized pieces of hardware they don't have screens on them they're you know they just run the same task over and over again to be able to control um manufacturing lines typically or you know in industrial applications so to be able to get data from them you need to have some piece of software running on a on a computer a pc and the way that that computer can talk to those programmable logic controllers that companies like siemens rockwell allen bradley on general electric all these companies that make programmable logic controllers allow their plcs to talk to this opc language for lack of a better description and this is a slide from the opc foundation kind of showing where these these different you know opc communication mechanisms exist um so why this is all interesting and important is i have a fake factory that i've i put together and this is a windows server it's you know not producing widgets or anything exciting but i i have a an opc simulator running on it so this is actually simulating data that might be getting pulled from one of those programmable logic controllers that might be running a robot or they might be running a manufacturing cell in a plant um so what what i've got a couple different um different simulated signals that are being sent from this uh such as a sawtooth signal see that this is actually just creating uh a value that ramps up and goes back down so so i have a bunch of random data that's actually being sent uh from this factory and these would be indicative of there may be a pressure sensor from one of those programmable logic controllers or um you know in the case of like a counter you know like that is zero over and over again yeah exactly yeah yeah and and you know the counter is just kind of is counting up to 100 and going back down down to zero so these are just um simulated signals that would be um you know the uh you know on a programmable logic controller there'd be a card in a rack that would have a bunch of wire terminals for outputs to be able to fire a solenoid or something like that so these these could be representative of values coming from those cards you know analog and digital signals so i have this this running on on this vm and i have the opc or the the iot edge device i just showed you and i showed that that iot edge device is actually running um a couple of um of modules and these modules are our containers if i go like this opc publisher this is actually coming from the microsoft container registry it's one that we produce and you can see it's in the iot edge repository and it's called opc publisher and i'm i'm getting the latest version of it so this is actually the the vehicle that i can actually use to be able to talk to my uh my fake factory so what i have here um and actually one other little little piece of information um so the um when you create a container in docker or in kubernetes there's a whole bunch of of options you have that you can feed to that container such as um you know details about how it interfaces with a network that and the machine that's running on or details about how it interfaces with storage on the machine that's running on and storage is actually a pretty particular important one because containers they are ephemeral and by that i mean like if this opc publisher container crashes if there's a uh you know segmentation fault in the in the software in this con this container um the iot edge runtime will detect that and they'll just create a new version or new instance of it so it'll keep it running for me um automatically so it's all managed by the iot edge runtime in the kubernetes world kubernetes does the same sort of thing so the drawback with that is obviously if you're storing data things like config files or log files or um you know things of that nature or data files if it's like a database that has has that stores this data into files it needs to be able to persist that data somewhere so when it gets started back up again or power cycle that data is still there when the container fires up iot edge modules work no differently than docker or kubernetes or mesosphere any of those environments so what i have here is um i'm actually telling it to to mount this industrial iot edge folder to a folder called app data inside of my container so this is how i can cut so so i'm basically taking something from my local file system running on my edge and i'm exposing it into the container and particularly what i'm actually exposing it to or the reason i'm doing this is there's a config file in here that i actually have details of what data i want to capture from my fake factory so and is getting it from this config file so this is kind of the glue to slowly stitch things all together so if i go go here this is actually this is my edge device i have it running on linux and if i do uh docker ps this will show me all of the containers that are running on this machine and if you remember when i showed you the iot edge runtime there's a 1.010 rc2 that's that's the edge runtime that's running on here and that's also the agent run time that's on here so these are the two pieces that are iot edge itself and the other three are the three that i show you see here discovery opct win and opc publisher these are ones that we make available for customers to use to easily talk and interface these industrial protocols so and then if i go into this folder if you remember in the configuration i told it i want to take the the i iot edge run folder and surface it to my container and if i look at this file can see that i'm actually connecting to um to my iot junk box which is my fake factory so this is actually uh this is a the port that's used by opc ua and i'm specifically telling it i want to take data from the counter and the sawtooth so these are our um you can see we can see right here the counter and sawtooth that and i'm sampling them at a rate of of um in every two seconds it's actually getting uh data from from that particular tag it's called an opc world so the way it works is you have a plc that might have thousands of wires going to you know different different racks that are deployed in electrical panels around the plant um and you attach an opc server to that to that programmable logic controller and then you create tags that are actually watching data in that plc that surfaces it to things like um you know displays in the maintenance crib to you know maintenance supervisors about the health of their of the pieces of equipment or in this case i'm actually vacuuming up that data you know with this with this edge device that i'm actually uh you know taking the these tags that are exposed for my fake factory and sending it up to the cloud so where this this um kind of comes comes full circle and comes back together um because i have a if i could just type correctly um so so i have data coming in actually just to prove that data is coming in if i go go back to my my hub i don't have any other devices that are connected to this iot hub and you can see um the message count because i fired it up earlier today around lunchtime has been slowly climbing so i've been getting you know every two seconds i get um i get a message actually i get two messages so i've been getting a you know a lot of messages coming from my my fake factory that's actually been sampling those messages and and um and you can see that i only have only have the the one connected device to the sub so now you know dave i have my edge configured is pulling data from my fake factory now i want to do something with it and to kind of pull it all together the final piece of the puzzle i there's a service in azure called time series insights this is a a um a easy to use service service that allows you to see time series data so uh what is time series data that would be you know data that's that's you know trending or changing over time to kind of visualize in it you know in a in a visual manner or be able to uh trigger alerts or alarms or some you know state or threshold is actually um exceeded so it's got you know some capabilities to kind of establish some trend lines and determine what's normal and alert you if things deviate past normal so there's some some capabilities in there that that allow you to set that all up but what i'm going to use it for is to show you the the data coming in um and while that's loading you i can see some other interesting stats of um you can see it's very close to real time that you know i have a couple spikes where it jumped up to this is the ingress lag the time it takes for data to come out of my hub into my time series database is uh a second or less so it's pretty darn close to real time as long as that edge device is connected so what would happen if my edge device were disconnected well that's another really cool feature of iot edge is that with all these services running on that iot device iot edge will actually store that data locally in that machine until service is restored or the connectivity is restored and it will upload all of that data to my hub and let me do things with it later so um so i you know right now i'm doing stuff with a fairly real time um but i i you know can also it can work in a disconnected state so i'm just going to zoom in uh to a particular period of time here and to be clear you could do anything you wanted with that data right you could you could populate a dashboard with it you could correct send it to excel you could do whatever you wanted yeah exactly so you can see here's my sawtooth data that i'm actually you know graphing out here um you know so this is actually um you know showing the the data coming in i mean this isn't super interesting and compelling but it what it what it demonstrates is something that um you know i worked in the controls automation space for for probably a better part of a decade doing a lot of you know industrial automation related um projects for for customers primarily in automotive and kind of vehicle space and this is actually a really challenging you know we charge you know six-figure contracts uh for building applications that would interface with pieces of equipment and surface data like this to customers i mean this is going back you know 20 20 or so years ago but you know this is showing very quickly that i can take in an edge device uh configure an iot hub and modify the config file to point it to an opc server then any industrial customer is going to have that sitting in their plant usually in some way shape or form if they're getting data from it for maintenance purposes or statistical purposes and we can ex we can tap into that data stream and do interesting things with it whether in this case interesting things are showing it on a graph but it could actually be much more sophisticated like feeding this to a machine learning model um you know that that might actually give me some you know you know some indication that that the machine is um uh you know having some abnormalities at certain points of the day or maybe you know create some relationship between that and temperature or the operator using a machine so so it allows you to make some very powerful correlations with that data that might be hard to do but just looking at that data raw so and you can do so very quickly and this is also not a very expensive you know thing for a customer to to do and you know i think an iot hub is you know less than a hundred dollars a month um you know time series insights and i'm running in a consumption mode which means i only pay for what i'm using so you know and just say a couple hundred dollars a month and you'd be able to have a solution like this tapping into data coming in from a factory wow that's really cool and uh there are so many uses for this right i mean it's incredibly powerful right yeah i kind of you know glossed over the surface and i know some of it you know for somebody that hasn't had you know some experience in opc world that might be kind of confusing to understand what that what that picture is and um you know kind of how it fits fits into into the the puzzle um and you know i did have there was there's one slide here that kind of does talk about this is this is a i didn't want to show this because this has a lot of uh boxes on it and it looks fairly complicated but what this is illustrating is that in an industrial application you typically have a lot of complexity you know you have the the the plant network that actually has all these devices and stuff running on it you know whether it be supervisory and control stations you know like machines around or workstations around the plant that are used to be able to control pieces of equipment and then you have you know the programmable logic controllers i talked about the plcs um that are actually running the machines in most companies keep the the networks that actually are running this this really important and safety critical pieces of equipment separate from the i.t network that is used for email and browsing the internet and you know running line of business applications and that sort of thing so putting an edge down in this space in the red space allows us to actually connect and actually there's some you know some some pictures of that on here it allows us to easily connect to um you know to to that data stream and be able to make decisions locally so i showed the data going up to the cloud but instead of going all the way up to the cloud i can actually have you know like a machine learning model that i've trained in the cloud running on the edge and actually generating an alert maybe to a plant operator when something happens so you know you don't have to go all the way up to the cloud and back to be able to make a decision yeah so yeah some some really uh you know to me i think this is a a cool product um you know that is ubiquitous it can run any type of general purpose compute device um instead of windows languages space ships yeah yeah our spaceships yeah i've got uh i've got uh about 300 raspberry pi's running my weapon system and the falcon the funny thing is that raspberry pi is probably more powerful than the things that they would have had available to them that is so my uh my my son's really into older computer hardware and looks at all the subreddits and you know some of the older computer hardware and there was a compact server like a dl-180m you know circa you know 90s era server one of the comments was a raspberry pi my desk is more powerful than that thing that's making your your electric meter uh yeah spin like crazy yeah i had a luggable you remember the compact suitcases i remember them my yeah my dad had one of those too and a little i don't even know what size screen that was was it probably like a nine inch monochrome screen green screen and then digital remember digital computers oh yes digital actually had a sniffer and they would they would uh essentially take a a compact luggable you know throw a net card in it throw some dedicated software in it that that replaced the the dos based os okay and uh and uh it was it was a network sniffer and you could lug it around and plug it into the network and it would it would capture network traffic and it was a dedicated device just just for that it was crazy because they they charged like two or three times i think the price of a standard compact luggable it was crazy and that's all it was was just their special sauce put on top yeah it was just a net card and no os except for their their application their application acted as the os and that was the deal but uh man i can see so many you know manufacturing huge inventory control huge yeah and like i said i've got industrial customers running these on vehicles you know that so yeah i mean you can do something really yeah i mean think about a tesla and i mean it's it's pretty much already an internet of things device right yeah yes yes i mean it's just crazy really neat yeah so hopefully this uh this you know provided kind of some insight you know and i just talked on the on the opc stuff of of um the holy for process control stuff for for being able to do industrial communication but i can build a you know python application and tell the iot hub to deploy that python application on all you know thousands of my edge devices running around the globe and so so it actually is a really powerful devops story too to be able to make it easy to deploy updates to you know equipment you've got around the cloud yeah all right well man listen uh is there any place people can go to get more information yeah um so you know obviously at azure.com but i will uh i'll give i'll pass the link um azure and dust tree lt and i spelled it wrong but uh [Music] yeah so this is actually exactly what i just showed and kind of talks about how to install this and configure this um so i'll we'll make sure we get the the link in the notes okay well i've got to get back to work yeah me too yeah thanks man yeah no problem thank you oops it's not working hang on oh there i am hold on oh where am i my head disappeared that is funny guys if you can still hear me yes again yeah nice that that's awesome i think you might be muted um yeah i think you might your mic's not working [Music] hang on i'm going to redo there you go yep you're you're right now all right well bud thanks that was awesome i really appreciate your time i'm gonna get back to work now yeah me too all right we'll talk to you later man is that a is that like a legit problem where'd you pick that up at all right guys that's your taste of premiere i'm your father
Info
Channel: Taste of Premier
Views: 133
Rating: 5 out of 5
Keywords:
Id: vH1ayWB5iEA
Channel Id: undefined
Length: 44min 23sec (2663 seconds)
Published: Thu Oct 22 2020
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.