Jump-start Your IoT Implementation

Transcription

1 Jump-start Your IoT Implementation IBM Code Tech Talk Dec 12 th tation/ >> MARC-ARTHUR PIERRE LOUIS: We want to welcome you to our tech talk series. My name is Marc-Arthur Pierre Louis, your moderator. Today we have two great topics to discuss. They all are related to IoT, and because the main topic is Jump-Start Your IoT Implementation. And to take you through that, we have two technologist. One is Jeff Dare, all the way from the Bangalore Lab, to talk to you about how you are start your IoT with a conveyor belt, and then we have Hovig Ohannessian, who is going to take you through Node-RED, a family product for IoT. You are in for a treat. I am going to turn it over to Jeffrey Dare for the first part of this Tech Talk. You have it, Jeffrey. >> JEFF DARE: Thank you. Welcome to this Tech Talk on Watson IoT. This is the first part where we will see how to build and monitor your own IoT-driven conveyor belt. So let's go on. Before I start, I want to brief you on what is an IoT platform and what does the Watson IoT platform provide for users that you can easily implement all your IoT applications and solutions. So any IoT platform is just an interface between your devices and applications with which you can manage your devices, get the data for all your sensors or networks, it could be smart cars, smart watch, smart refrigerators, whatnot. Right? So a platform should be able to consume all the data from your devices. And it's for your applications to consume that. Right? So the platform acts as an interface between those two. And your platform must also be able to augment data from other data sources as well, like data sources could be as varied as like weather, map, or any other IoT platform. So basically, your IoT platform accesses interface between your devices and applications. So quickly, I'll move on to what the Watson IoT platform provides for you to develop your IoT solutions. It can be broadly explained in four different quadrants. The first is the most important one, the connect. So the connect is where this facilitates you to connect your applications and devices to the platform. There are multiple forms, libraries HTTP, et cetera. It also supports a lot of device management capabilities, like you can

2 do a remove firmware upgrade or remotely reach your devices as well. Of course, all of this is on a secure channel. And it also provides you various visualization tools you will see as part of this Tech Talk, which will facilitate you to do all your visualization needs on the platform dashboard itself. And the next key part is your information management, so that manages all your data that your devices are sending so that it can be in a more readable format for your applications. It does manage all your storage and how you pass all the data and how you manage your entire data. Right? So there are various ways with which you can maintain the state of a device and know what is the current values of all of your device data. And of course, the main key thing for all of your IoT is what sort of analytics that you can provide. Right? So that's where you get a host of IBM cloud services with which you can augment all this data and get meaningful insights to it. Right? So you can actually do real-time analytics, predictive, augmentative, and contextual as well. You have a host of Watson services that you can do cognition for all the IoT data you get. Of course, there are a lot of analytical tools you can use as well from the IBM Cloud catalog. The fourth quadrant is risk management. A lot of people are worried about the security of your IoT. That's where you have your quadrant four, your risk management, do all your security and analytics, like how do you manage all your keys and provide security to your organization and the data and everything there on your devices. Right? So these are the four main things that the Watson IoT platform gives you. So I think that there provide a foundation for us so you can see what this particular platform does. This is the main thing that you are going to see now. So how you can actually build and can proactively act on all the devices. How can you monitor your IoT-based conveyor belt? So the main use case we are trying to solve is how can you connect all of your devices to the Watson IoT platform and monitor and proactively act on the device data; right? So basically, theoretically, this can actually be extended to play on any other device. Of course, right now we are using a conveyor belt, but this can actually be used for any other device as well. So how can you monitor your devices using the Watson IoT platform? The end goal being how can you catch the problems before they actually occur? The main components, of course, for doing this, we use the Cloud Foundry and Watson IoT platform that I just explained. One is we can see 2a and 2b. You can use a device, like your Raspberry Pi and that's sending the data to the Watson IoT platform, or you can use a simulated application with which you can send to the Watson IoT platform. Both of these are sending the data to the Watson IoT platform and you can use the dashboards to visualize the data and proactively act on.

3 So this is all I had on the data. Now actually let's go ahead and directly go into the demo so you can actually see what we can do. So this is the pattern that you see of how you can actually build and monitor your own IoT-driven conveyor belt. Right? So you can get the code off GitHub. This gives you a brief description of what this actually does. And the technologies that we have used for this particular pattern is we, of course, use NodeJS as runtime, and we are using Watson IoT as a platform service to maintain all service needs. We use platform device to work with Raspberry Pi and Watson IoT as well. This is the flow I just explained on the presentation. This is all you have. This is the various components that we have used. And of course, this also gives you related links which you can use if you want more information, you can go ahead to see how that IoT architecture, if you want to have the documentation, you can go ahead and see this. And other related blogs that you can go ahead and see there. So let's directly go ahead to the code. I am just opening the code. So this is provided in GitHub, so you can actually go ahead and see the code that is written for this. So this is, as I said before, it's written in NodeJS, so you can actually see all of the code, and it provides a detailed description of what this whole exercise is on and also how do you actually go ahead and deploy. So as I said before, there are two parts for this. One is to quickly install a simulated conveyor belt as a Web application in Bluemix. This application automatically registers itself as a device onto the Watson IoT platform with which you can actually go ahead and see. And if you want to do a more challenging work, if you are up for a challenge, you can actually go ahead and use the part B where you actually have a Raspberry Pi that is connected to a motor and is using the Watson code. >> MARC-ARTHUR PIERRE LOUIS: Quick question, Jeffrey. If somebody wanted to go the second path where they get the hardware, do you have instructions as to how to make it work, what component to buy, where to buy them, things like that? >> JEFF DARE: Yes. We actually have a detailed description of all of that. So if you actually go down, so this is the sample simulated conveyor belt, but of course, if you want a more challenging work, you can actually go ahead, so we actually give a diagram of how do you actually go ahead and connect the circuit with the Raspberry Pi and how do you connect the battery and the DC motor to the DC controller board. And we also, of course, give the actual printout of all the execute, and of course, we do give the actual requirements of how you can go ahead and buy this and also can connect this. >> MARC-ARTHUR PIERRE LOUIS: Excellent.

4 >> JEFF DARE: All of this is provided in the documentation. >> MARC-ARTHUR PIERRE LOUIS: Excellent. I am always fond of doing projects like that, so somebody could go in, buy the hardware, then hook it up and see it working and see it controlled by Python and things like that. This should be very fun. I guess this could be a Christmas project; right? >> JEFF DARE: Yeah. I think everyone should do this. I think most of them it's not so costly, and you can actually go ahead and buy this. Everyone can try this out, and a you should get back to your old tinkering days, yeah. >> MARC-ARTHUR PIERRE LOUIS: There you go. Thank you. All right. Cool. >> JEFF DARE: So this is the first part where we can actually get the device data. We can use the simulator or the Raspberry Pi to get the data. And the step two is like how can we see all this data on the Watson IoT platform? So you are getting all the raw data, then with getting the raw data, then how do you actually visualize the data, all the capabilities that the platform gives. Let's go ahead and deploy this particular application. There is Deploy to IBM Cloud button, so all you have to do is this particular button helps you to -- it will actually set up everything for you, so it will set up an application in IBM Cloud. It will also set up a GitHub repository for tracking your code and issues, and it also sets up the pipeline. Let's go ahead and deploy this particular solution onto the IBM Cloud. So if you see here, it's starting to load the application. >> MARC-ARTHUR PIERRE LOUIS: So you are already logged into your Bluemix ID? >> JEFF DARE: Yes. So I already logged into my Bluemix. If you are not, you will just need to log into your Bluemix, and then click on this button. If you are not logged in, it will also prompt you to log in. >> MARC-ARTHUR PIERRE LOUIS: Okay. Of course. >> JEFF DARE: As you can see, it is working on creating a tool chain for this particular project. There we go. So it actually gives you various options. This is the name of the toolchain, and it asks me where do I want to deploy this and what is the organization and what to deploy. So it sets up your Git repositories with issue tracking for your application, it gives you a Web UI with which you can edit code and also sets up your code delivery pipeline for your application as well, so let me just go ahead and deploy this. So I will just select the region where I want to deploy this particular application, and then go ahead and deploy. >> MARC-ARTHUR PIERRE LOUIS: In that case, you have a simulator; right? >> JEFF DARE: Yes. With the simulator, I can actually see the

5 devices talking, so that makes it much easier to see. So as you can see here, right, it, of course, set up your GitHub, so you can actually go ahead and see that. And it also does set up your own delivery pipe Lynne for this application. So if you do any changes, all of that gets published to an application in IBM Cloud. So I just open both of them. As you can see here, it actually just did a Git clone of the source file that is present in GitHub, and it give you the entire thing, so you can go ahead and edit this file. So now if you can see here, so it actually does say, like, how it is actually staging, so basically, what you see here is the actual code being deployed onto the IBM cloud. So this is the code that was being deployed onto that cloud. So let me just -- I already have an application that is deployed on my space in IBM Cloud, so we just go ahead and use that. So this is an application that I just deployed sometime back, so I can just click on that. >> MARC-ARTHUR PIERRE LOUIS: I love this. This is a cooking show; right? >> JEFF DARE: (Laughter). Exactly. So this is the application that I just -- like yeah, so just deployed a few minutes back. So that I just clicked on it, so you can actually see the different services. So as I said, right, the Watson IoT services automatically bond to this application, and it gives it to me. It also has the entire delivery pipeline defined for you. So your whole DevOps is actually done for you. >> MARC-ARTHUR PIERRE LOUIS: Okay. The screen is blank now. >> JEFF DARE: This is the application. Of course, you can see that it's running. There's one instance, and this is the connection, so basically, this is the Watson IoT platform that is bound to this application. So basically, I can listen to what all the events that are coming from this particular Watson IoT instance. So if I go ahead -- this is the continuous -- defined for this. Let me see the Watson IoT platform. So I just click on that. So this is the Watson IoT platform that is bound to this application. So let's go ahead and see -- as soon as I click on this, it will launch the dashboard, so even that I actually have -- okay, I think it's opening up. I actually have another instance that is already opened up. So I just click on that. So as we wait for this to load, so now what I will do is I can -- yeah, so this is the Watson IoT dashboard. You can see your different boards. I will go back to my dashboard, and let's open up the application that was deployed. Right? So we can actually see what that particular application does. Right? So I am just opening up the particular application, so this is the application that was deployed. So I am just opening up an instance of that. So this is the simulator application that I just deployed. So of course, it's asking me to give a unique device ID and a password.

6 So let me just do demobelt2, and a password for it. So it's just any eight-digit password. So I just click on okay. So what you can see here, this is a simulator, so it actually starts simulating this particular data. The data, this is what we are interested in. So what is the current RPM, what is the particular time stamp, what is the ID, and this is running. How are we doing all this? Let's now go back to the code to quickly give you an overview of how this particular application is running. So as I said, so this is a NodeJS application, so if you go back, you can see all the NodeJS dependencies and the version of this particular app. Of course, this is the Git repository for this app. And the main file which orchestrates your entire application is seen here on the simulator of the conveyor belt, where you read the recap, get all the credentials, and you map it. Basically, these are the various credentials of the Watson IoT platform that you map to your application. >> MARC-ARTHUR PIERRE LOUIS: Jeffrey, could you say what is VCAP? Some people may not be familiar with Bluemix. You said you are reading from VCAP. What is that? >> JEFF DARE: In Bluemix, if you want to read any credentials of the services that are bound, so right now, if you remember, I showed you that this application is bound to the Watson IoT platform service. Right? So all the credentials from the Watson IoT services can be read into the application using VCAP. Basically if you can see, this is where I am reading all the credentials from the Watson IoT services and getting the credentials from that. So this is a secure way of passing your credentials from the service to the application. >> MARC-ARTHUR PIERRE LOUIS: And VCAP is kept in the environment; correct? >> JEFF DARE: Yes, correct. Yes, it is in the environment of the application. Yes. >> MARC-ARTHUR PIERRE LOUIS: Okay. >> JEFF DARE: This is the main -- let me quickly show you. It is transmitting data from the browser to the Watson IoT platform. So it actually using the protocol MQTT. So that's de facto standard in the IoT to transmit the data from the browser. So if you go to the public, so that's where most of your JavaScripts -- the client-side JavaScript resides. So if I go to the client-side JavaScript and go to the conveyor belt, you can actually see what we are doing. So of course, we see this is the platform device, what they are doing, how they are reading the data. It has all the code. This is the device ID that I just gave with the password. And is it starts transmitting the data. Of course, if you remember, it also registers on its own, so that's the reason we are giving the password. So it registers the device on the Watson IoT platform then starts sending the data to the Watson IoT platform. >> MARC-ARTHUR PIERRE LOUIS: Can you show the loop where it

7 processes the data? It must be a loop. Or is it querying it or polling or just simply wait to go get notified? >> JEFF DARE: Yeah, so the Watson IoT platform follows a publish and subscribe model of getting the data. So basically all your applications subscribe to the data, and this device starts publishing. So it's more of publishing the data to the platform and your application starts to subscribe the data. >> MARC-ARTHUR PIERRE LOUIS: Pubsub. >> JEFF DARE: Yes, it follows the pubsub model, yeah. The code is using this particular format of code, so this is where you can see all this data being published. Now let's go to the Watson platform and see this particular device. Let's go ahead. So this is the platform dashboard. I click on device, and if you remember, this is the name that I just gave. Right? Demobelt 2, and this little icon says that this particular device is connected. Of course, we can simulating it, but for the platform, it is connected and is transmitting the data. Let's quickly see the raw data. So if I just go to the state, it actually gives me the raw data that we can see here. So voila, you have your data that you see here. This RPM that you see from the conveyor belt, quickly go ahead and change the value, you should be able to quickly see the change value. Yeah, 1.2. Let me move on, and let me quickly show you how you can Mr. Your own visualization. So how can you monitor your conveyor belt? So et cetera that the main part of this particular Tech Talk, how we can actually monitor all the data from your conveyor belt. So the demobelt 2 is the one we are interested in. I select on that, and a I can see all the device information, like when was this created, who created it, and what is the device type name. And these are the various events. If you remember, it was 1.2, so I can go ahead and change the values, and you can start to see the live data from the device on the platform. So now if you want to actually monitor this data, like I want to build something more real estate, so let's go back -- >> MARC-ARTHUR PIERRE LOUIS: Real quick, Jeffrey, just for the sake of the audience, RPM, is that your Red Hat RPM? It's revolutions per minute; right? >> JEFF DARE: Yes, it's revolutions per minute. That is a state you want to see from your motors; right? So that's the parameter that I want to read from the motor. >> MARC-ARTHUR PIERRE LOUIS: So you can basically increase the speed of the motor by increasing the speed of the RPM? >> JEFF DARE: Yes. That's what I am doing now. Basically, when I increase the speed of RPM, so you can see the conveyor belt moving really fast. Right? So I see it is moving faster. >> MARC-ARTHUR PIERRE LOUIS: Exactly. Cool. >> JEFF DARE: Yeah, so it's not like one RPM. So it's basically

8 one times, so it could be a thousand RPM, but it's 1.1 times of thousand RPM. >> MARC-ARTHUR PIERRE LOUIS: Okay. >> JEFF DARE: That's what you are seeing here. So let me create a new board. I just want to say it as my Monitoring Board. So let's go next. So I have just created a new visualization board that I want to do. So suppose I can actually start visualizing this by adding a new card. So now let's just say first I want to see a list of all the devices from this particular organization for my Watson IoT, so I will just say next, submit. So this is very similar to one we just already saw. And next I want to build a line chart. This is where I can -- how do I monitor the live data that's coming in from my conveyor belt? So I just select on the device ID. This is the device ID I am interested in. So I just click on Next, and now I say what is the data sources that I am interested in? So for you to see here, right, so the Watson IoT platform is included enough to get the actual data points that this particular device is sending. So it automatically senses that this is the data, and these are the various properties that this particular device is sending. Right now I am in the RPM, so I can name it to whatever I want, so I just say Motor RPM, and of course, this is a number, so as soon as I say number, I can actually give my value. So I can say it's because my motor goes from 0 to 10, I just say next. This is the line chart that you see. Let me just increase it to a bigger one. So I just do next. And submit. So this is the line chart. So this is an actual monitoring dashboard that you can build using the Watson IoT platform. So right now, of course, it is sending 1.5, so I think reduce or increase for you to see the actual change in the graph. So I increased it gradually, so you can see a gradual increase in the motor RPM. So there are various other things that you can, of course, build. So I can also go ahead and build a gauge for the same device, of course. And I will use the same RPM. Of course, it's number. So you can actually see. So you can build various different visualization tools for monitoring all your IoT data. Right now I am just monitoring a single data, but it can be for any -- single data point, so it can be for any number of data points that you can see. And as I said before, it need not be just for conveyor belt. You can actually use this to monitor any devices so that with which you can start to diagnose any problem that happened before it actually occurs. You can start monitoring your devices using the Watson IoT platform. So there are various different visualization tools you can actually start going to build. You can actually build a map. Suppose your devices are sending GPS data. You can start to plot that in an actual map and to see where the different devices are

9 present in the map. Another key thing that, of course, right now -- right now, of course, I am using a browser. If you actually go to the same application using your mobile browser, it also starts giving -- it reads your accelerometer values and you can view those in the platform as well. When you try this out, you can actually view this application in your mobile so that you can start transmitting your accelerometer values as well. So this is a very basic thing of like how can you actually build an actual IoT device and then use the Watson IoT platform, and then how do you start monitoring those IoT data using the Watson IoT platform. So this is a very basic thing that you can start building using this pattern. Of course, I think I will be interested to Mr. Something more as well; right? So of course, we can actually go to the next level. So if you go to the documentation page, so there is multiple things that you can do with this particular sample as well. So of course, this is just a starting step. But you can start using this, you can actually start building something more meaningful as well. So if you go back, so if you go to the documentation page of the Watson IoT platform, it has the entire thing, so how can you connect to different devices? How do you analyze your data? And if you come to Getting Started; right? So if you want to build something more, like if you want to start monitoring your device data, so this is something. So this is another application that you can see with which you can actually monitor multiple different data. Right? So of course, right now I just have two devices. Suppose if you have like a hundred devices and you want to monitor all of them in real-time. So that's where you can actually start seeing. Right now the belt1 is stopped, and the belt2 is running. So I can actually increase this particular instance. You can actually see the increase in the time. If I go ahead and stop, of course, you can see that particular device is stopped. So I think this is one part where, of course, you just sow how you build and monitor your own IoT-driven. The next step is you can try it with an actual device, an actual Raspberry Pi and then how do you increase the applications with which you can start consuming all this data? So I think that's all I had for this particular (Inaudible). For the next interesting session on Node-RED and how you build IoT applications using Node-RED and Watson. >> MARC-ARTHUR PIERRE LOUIS: Thank you, Jeffrey, for that presentation. I think it could be a Christmas gift that somebody gives their kids and get the kids busy for Christmas. You know? It would be an interesting project. I know we thank you, Jeffrey, and then we have Hovig Ohannessian from Austin. You have.

10 >> HOVIG OHANNESSIAN: Thank you, Marc, and thank you, Jeffrey. Let me share my screen. So I will do my PowerPoint. So -- >> MARC-ARTHUR PIERRE LOUIS: did you stop sharing, Jeff? >> JEFF DARE: Yes, I have stopped sharing, yes. >> MARC-ARTHUR PIERRE LOUIS: Okay. We see it now. >> HOVIG OHANNESSIAN: Okay. Let me close this here. So the case with the Node-RED is going to be different than NodeJS, the way that Jeff explained his project, but there's lots of similarities of how you can design the IoT. And in here we will bring the same concept of the conveyor belt, and we are going to have the same one. But what we've done here is it's more like simulator. You can take whatever hardware he had, Jeff, in his report and plug it into the IoT platform, and you can extract that to Node-RED. Or you can use the simulator, or you can connect any device that you'd like and have control from Node-RED. So I am just going to go over the Node-RED, as some people would like to understand, and as you can see, Node-RED is kind of like a wiring tool. It's basically an application composition tool that you can bring in and connect things, nodes, with simple code and see things running. The way NodeJS works, it has -- sorry, Node-RED works, it has -- NodeJS has a runtime. Basically, everything is written with JavaScript, and you can see that the stuff related to JavaScript, that it's event driven, single threaded, and what you can do is you can have -- you can use, let's say, if you want a mix between this project and Jeff's project, you can have the NodeJS running, but you also can have Node-RED running on your Raspberry Pi for example, and attach and do basic job to send data, for example, and collect it on the other side and display it with the NodeJS, or you can install the Node-RED on Raspberry Pi and do the control from there and send it over the platform and receive back on Raspberry Pi to perform other tasks. So there's multiple ways you can handle this. I just suggest if you want to have more ideas, you can go on nodered.org and check it out more. So if we move on, just kind of like not to be short with time, what we are going to do is we are going to go to Bluemix catalog, and we are going to show you like Node-RED, what are these and how you can use them. We are going to name them and create, and this is how it's going to look like. And you open it, and these are the nodes that basically can make use of. There's lots of things displayed here, and you just have whatever your project is about or to build you can always come and make yourself these services, and we have lots of variety of Watson nodes that you can take advantage and build nice projects. So this is the parent that we are going to be focusing on. It basically does create, delete, start, stop, increase, decrease. It's similar to the NodeJS application with the difference that just

11 the image is missing, and basically, this one is only a simulator for now. It creates and deletes the device for you, basically. So that's what's happening. The connection of these nodes, we are going to cover all this. And just one important piece that if you try to use this, let's say, the Node-RED project, just make sure you can see the device manager, device type manager here on the right corner. Sometimes it's like red dashes, like it's not available, which means that you need to go -- and I will show you later on. It's three lines. You open that and go under the palette, and you just basically switch to the Node-RED contrib IBM wiotp device, and I will show you more about it. It does similar to what Jeff was showing. Going to be displaying actually the speed on the graph here, and we are going to see how it's performing with every single change we do about it. So I will take questions while I am going to try to move on also to the demo. Kind of like if you have -- if you get this chart, let's say check the pattern, check the GitHub, and also get feedback. Let me go real quick on the pattern. It's basically -- it's covered kind of like in here. We make sure that we put the links and whatever you need to start with the project here. Especially there will be device, whether it's simulator or real device, and Node-RED is talking to that. It will store this in Cloudant, and it also will be talking to the IoT platform, and from there we are going to be doing some display. So these are the steps. And let's say -- let me show you the repo. You can come in here. The case is a little bit different than NodeJS in the sense kind of like Jeff was showing, that with one click, you can deploy. It's simple with NodeJS. But in here I found out that it's not that simple. I am checking the pipeline, like how we can improve it. It's sticking to the tool chain. I am going to figure out how to make it easier for people. But for now just you have everything covered here in README, and basically, let me scroll a little bit and show you. It shows the details, like what you need to see to make sure you have -- what you need to configure on this project, the Cloudant and the IoT platform services. We are going to cover later on like hands on, but let's go over this. And it shows every single detail that you need to do to make things easy. Also like display, like if you want to do it manually, push your own, let's say, deploy teams on your own. This is the display you are going to see. And also that's how it looks like when you get it and this is the raw data that we are going to parse the speed and show it on the graph. And this is how it's going to look like. We have videos. We have -- you can try live. Then let's say if you click on this, you can see the sample. Slate-iot.mybluemix.net/motor. Some people would like to see how it will be displayed. If you

12 play with that, it's going to display at the bottom this raw data, but I want to bring in the graph, and whenever I play with this sample, you will see things changing with every change you make. So that's next. Just remember something, that this Node-RED project, it took about less than a week to finish it. So with Node-RED, you have, like, not only the speed of the device, but your own speed of accomplishing projects and getting things done. If you have time and you want to focus on writing your own -- doling with your own, let's say, coding where you want to, let's say, the whole NodeJS, think about how you want to connect, how it's going to happen, think more details, Node-RED will come in and help you and make things easy. So this is for the simulator. Let's see how am I going to cover this. And I want to show you. So if we come to the Bluemix catalog, and you type in Node-RED, basically, you are going to see this. We are interested in the boilerplates. What are the boilerplates? Think about packages. Node-RED will give you Cloudant service along with this package. But the IoT platform here, it will give you basically -- it will be a Node-RED with the IoT platform service with it, and also Cloudant. So you can just click on this and name it. Let's say we name it demo-iot-control. And just create this. >> MARC-ARTHUR PIERRE LOUIS: This is full control; right? This gives you IoT platform, Node-RED, and Cloudant; that's the whole package; right? >> HOVIG OHANNESSIAN: Yes. Yeah, sometimes it depends on the connection also. Let's go back and show you -- okay. I have done previously, just in case things get too long, so I have an instance running. All right. So let's come to connections. As you can see, we have the Cloudant, and we have the IoT service instance here. So let's wait until it finishes running. Well, probably I can use this one. Same thing. As you can see, it has both Cloudant and IoT platform. And you just say visited, and you come here, and it opens up. So the first thing you are going to see, there will be like by default some nodes coming at you. You can keep it or delete it. It's up to you. Let's leave it for now. What we are going to do is go grab the JSON flow for this code, and we are going to paste it in here. So in order to do is that, we come to the repo, basically choose flow, click on it, open it up, and choose the role, just Control-A, Control-C, and you can come in here. These are the dashes I was telling you about. Just before copying and pasting, I can cover the palette I was mentioning about. If you don't see it, this is the one I was mentioning about. All the details are available. If you don't know how to do it, reach out to us, we can help with you this. Now, in my case, it's installed, it's in use. So In your case, you have to install it if you had a

13 problem not seeing it. But anyway, let's come and paste the code in here, and we are saying import, and right here we move it up, and that's it. Very easy. So this is what I have, and a few things. You may want to make sure that it's happening that you want to check the credentials that are available. You see that all this are checked in, when you copy/paste it, it will kind of come with it, but just get a look on it. Okay. So as we can see, I am using Cloudant. And I am using a database, the name of events. I am going to have to create that database inside of Cloudant. We are going to do that, but before doing that, I want to make sure -- okay. One important thing that people face is that let's say they copy and bring the code in, and they give a name different than what I had. Let's say you are copying it, but my default one, as I mentioned in here, so as I mentioned in here, it's simulate IoT. It's called simulate IoT. When you grab that code and bring it in, make sure you name it what you have. My code in here is -- you make sure you come in and you change. Look for simulate-iot. Right here. And make sure you replace it with your app name. Beltomotor. I wish it was a little more efficient, but for now this is what you can do. Let's rename this so it will work with your editor or with your application. That's one important thing not to forget. So right there, we are done. Click on Done, and it's deploy. So what we need to do is come in here and let's say launch Cloudant. I think it's already opened. We can refresh it. And we are going to create database with the name of events, as we mentioned. All right. It's created. That's one thing we've taken care of. That's our IoT platform. So I am going to show a few things in here. So as you can see, we have a display. I am going to show the display. The display basically -- this UI that we were talking about, it handles by clicking just buttons, and it will create it, delete, increase/decrease, start/stop, let's say, control the device. But also you can control that manually. For example, like if you see it's empty, you can come, and let's say manually, like you can say create a device, and you just refresh this. Now, you may have a problem that one thing to -- that I forgot to mention to you is you need to create an API key and token for this. Oh, if you kind of like face the problem that you see that it's not happening, like you can connect to the platform, make sure you use the API key and token. So we come to the platform, and you can see that manually, we created this device. Let's say if I click delete and I come here and I refresh it, it's going to delete it for me, and it's all simulated devices. It's creating and deleting for you. Basically, you -- if you are new to Node-RED or new to IoT and you want to work

14 on this, this is a great place to start. Like, you can create and delete and see how it's functioning and see how UI is working with it: But also, don't forget that you can easily control a real device with the Node-RED here. So if we come and open -- let's take this, and use -- you can see the UI is in this template node. This template node is basically the -- they call it a mustache template. It's a simple Web template that you can just put in codes, and in here I am using HTML and JavaScript mix, but sometimes what you can do is like if you grab, let's say, any, and as you can see, they call it mustache because of the curly braces, and you just put the name that is holding the data, and it will be displayed for you. But anyway, to gain time and for me to be able to show it to you, I am going to open the slash belt of this. I open it, it's right here. You see belt motor and slash belt. So what I do is right now we don't have guys created. Let's say create. Now I come over here and refresh. Just so I can show you. There is a lot to cover, and if you have a lot more questions, we will take it at the end, and you can always get in touch with me. Okay. So to create it. Let's say at the start, it's basically zero. It starts. But I need to show you. So as previously in the previous demo, with NodeJS, we kind of like created a graph. We are going to do the same thing, so you can say add new, and line chart, and automatically brings the device in. You say next. Connect data set. It's the eventdata we named it. I am interested in the speed. That's what I want to display. Number. Unit you can name. Just let's see if it's -- here we go. Okay. So we say next. I am going to choose the extra large for better view, and submit. And it's right here at the bottom. And I am going to bring it on top so you can see it, and as you can see, there are details SVT device itself, so it's better to keep them next to each other. As you can see, it's zero, and let's say I'll increase it, and I'll submit the speed. So here, as you can see, it went up. So you can play with this and see it running. So unlike the NodeJS and JavaScript, so the only thing I can show you as far as code is basically these nodes. This is where the credentials of creating the device in the IoT platform. And everything -- any request that you are doing is happening behind the scenes here. You don't have to deal with it like NodeJS. Node-RED is taking care of that for you. So this is how I am creating the device. This is how also I have the credentials to contact the lead. I am making sure I give it the name, the device guide, and all this information ahead of time. And this is where you can grab this code, change it, play with it, and basically I am doing these calls and pass it on. So this way you can see the device is taking action. And this is where it's inside the platform, and I am grabbing

15 it from the IoT platform and throwing in the events, the Cloudant database. From there also I am collecting it for the display on the UI. So I think with this, I have almost covered. Let me also mention if you need -- if you don't know how to create IoT token or key, there's a how-to I made. You can use that and come and see the steps. It's I made a simple step-by-step for you to create that if you need it. >> MARC-ARTHUR PIERRE LOUIS: Okay. So your time is almost over, so if you want to take questions, if there are some questions worth answering live, can you call them out, to either Jeff or Hovig? Okay. Let me switch to the Chat to see if there are any. >> JEFF DARE: Hovig, I think there's a question on the chat that's asking what is the error message with the device manager, so I think that's -- because I think that device is already registered, but maybe (Inaudible). >> HOVIG OHANNESSIAN: So sometimes you think it is a problem, but if you come to the debug, yeah, let's see. It's talking about credentials. This is where you want to plug in the API and the token. So it's looking for the credentials that you have that enables you to talk with IoT platform. But for me, it's doing the job, so I don't need to go further more because as you can see on the IoT platform, it's doing its job, like it's giving the device type, and it's giving the name, and I don't have any problem with that. You can see I didn't go further to add IoT token key to make it. So if that bothers you, you can add it, but it's doing the job. >> MARC-ARTHUR PIERRE LOUIS: Any other questions? Okay. Well, it doesn't look like we have any other. I want to thank Hovig for the presentation, IoT with Node-RED. Node-RED is a powerful tool. When Node-RED was just being created, I held a Hackathon where Node-RED was the winner, so I am very familiar with Node-RED, very meaningful to me. So please, we will have this presentation, the recording available for you, and you can review it, and you also can play with it for Christmas, get your IoT skills up and get your Node-RED skills up as you want to get into the IoT world. Tomorrow's Tech Talk is going to be about Bridging the Gap between Data Science and Business Users. I posted the link to register for this Tech Talk tomorrow. If you want to go there, you can go into the chat and get that link. But this is a very important topic as we switch to data science in 2018, where data scientists is going to stop being a very difficult area, where we are providing tools that make it easy for to you do data science, and this talk is going to talk a little bit about that and how to bridge the gap between the data scientist and your business users. Join at 11 tomorrow so you can hear about how you can bridge that gap. Very interesting topic. That's going to be the last one for the year,

16 so you want to close the year strong and get ready for 2018, come tomorrow at 11:00. We are waiting for you. Thank you for your time. Thank you, Hovig and Jeffrey, and thank you for your time. Good-bye. (End of session, 11:00 a.m. CT.) *** This text, document, or file is based on live transcription. Communication Access Realtime Translation (CART), captioning, and/or live transcription are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings. This text, document, or file is not to be distributed or used in any way that may violate copyright law. ***