 So we're here at Linaro Connect and hi, so who are you? My name is Isshai, I'm from Shiratek Israel, part of AY Group, and we are very happy to be here. We are very intrigued because it's very interesting for us. We have a lot of new developments, a lot of new plans working with 96 boards, and I think this is a good start because this is the first time we are having here, being here in beautiful Vancouver, doing cooperation with 96 boards. So you're doing a lot of different mezzanine boards, right? A lot of mezzanine boards, some of them are here, some of them are going to be defined and showed in the near future, and Guy, please, Guy is working with me, he's the VPR and D of our company, I'm the CEO and Guy is going to introduce us to several boards. So what is the background of the company? What is the main purpose of the company? Shiratek does songs and SPC boards for our two varieties. We do smart boards and we do, of course, 96 boards and a combination of both and future designs, and we're part of AY electronics, which is about manufacturing in Israel. So actually everything we do, we do inside, we do full turnkey, we do manufacturing, purchasing, assembly and design, everything in-house for our company. It doesn't have a lot to do with a cellular and that kind of technology? We also, because we are vendors and we are manufacturers, we are working closely for many companies in the world, such as Quicktail. So we divided the first LTE boards based on the 96 boards, mezzanines and also NBIOT boards based on Quicktail, and this is some of the boards that you can show here today. All right, thanks, thanks a lot. So hi, so who are you? Okay, hi, my name is Guy Zoar and I'm responsible for the R&D in Shiratek Solutions. So what is the collaboration you have with the NARO? We are participating here as a partner of 96 boards, and so you asked before, shy, what is the focus of Shiratek? So one side is a system on modules and the other is the 96 boards, the mezzanines. These mezzanines that you can see, we have several here. Here we have an FPGA mezzanine. So which one is that? This is the FPGA mezzanine, this one here in my hand, which what it does, it's a bridge between interfaces that, for example, Arduino shields that I can put here, which connect to them, all kind of sensors, and the bridge between them and different processor boards, like for example, the Aero Dragon board that I have here in my hand. But what we are doing here, so we are developing these 96 board mezzanines, which allow developers to add more and more applications easily and doing their projects fast. So what they can do instead of develop their own sensors and hardware, they're using different mezzanines on top of processor boards, and that enables them to have access to a lot of resources like weatherage sensors, different interfaces, dry contacts. So for example, we have here LTE, so this is a fast speed LTE, by Quicktel. It has also different sensors on it. Below it is another LTE mezzanine, which is for low power. So it's a CAT, category M1 NB1 mezzanine. So for example, you can select the high speed for an application like a camera, for example a security camera or face detection, that you want to transmit it over the cellular network. And below it, the CAT M1 is more for an IoT application. It's low bandwidth, but it's low power. So for example, if you're going to have a sensor in the ground that needs to operate on a battery for five years, so this is the right platform for you to choose. So NB IoT is a low, low power consumption? Well today IoT is everything, but when I refer to IoT, mainly it's low power, low bandwidth sensors that try to actually provide information about the physical entities around you. So that's what I mean. So you're basically expanding, like let's say the Dragonboard, you're expanding it to do all kinds of more stuff, is that what you do? That's what mezzanine is about, right? Yes, so this is the 96-bolt concept that you have one power. So how do you connect it? So you have here, as you can see, you have here the connectors, there's a high speed and a low speed connector. And I just connected between them, as you can see, just connected. Okay, so I can have one, and then I can have another one on top of it. Really, three? So I can connect actually this one, it's an Arduino one. So you have an Arduino over the mezzanine, over a Dragonboard. But if I had, for example, this one, this is a Bosch sensor, and I would put it over here, and then I can take over it an LTE mezzanine, as you can see. So I have already a three-bolt sandwich, as you can see, the processor board, the Bosch sensor, and the LTE mezzanine. And this, as you can see, and each mezzanine is a functionality. So I have a functionality on top of a functionality on top of the processor board. And again, this allows you to develop more and more applications. So if you think of it connectivity, sensors, processor, and but with another sensor, so it's endless applications that you can connect and develop very fast. So this is actually the concept of the 90s. It's not our concept, it's the 96 board concept. And we're happy to be a partner and to add our boards to this fast-growing community. So, for example, so there's a Dragonboard under there, but then you're connecting an FPGA to it or not? Yes, that's a separate board. This is the FPGA. So how does the FPGA work together with the Dragonboard? So actually, the FPGA, as an FPGA, provides the flexibility of programmable hardware. So you can say it's agnostic to the processor because you can add any interface or you can boost your speed by using the FPGA, but it still works together because in the end, you need to connect it. And the processor, the Dragonboard in the end, will control whatever you do on the FPGA. So of course, in the end, it is one application. So that is how... It's like you're doing a lot of different things. And also you have a Bosch. What is that? So the Bosch sensor, you can see it here in a minute. So on the Bosch sensors, let's put your hat here. I just have quite a lot of interfaces here. Sensors, sorry. Don't remember all the names of it. So you can see that we have a minute. So you can see here the different sensors that you have here. You have a pressure and temperature sensor. You have a drag cell accelerator. You have a gas and humidity pressure sensor. You can see here the block diagram, usually that will interest the engineers. So you have a collection of different sensors. All made by Bosch? Yes, all made by Bosch. So of course, we work together with Bosch for that. But in the end, it's what really is important that there are different sensors for different applications that the developer can use. That's the bottom line for it. Nice. And can we talk some more with these guys? They're also in your team, right? So these guys are... Here's a voice actor. Guy or voice actor? Yeah, can you introduce more about what kind of engineering is going on? What are you doing with your colleagues over here? Okay, so these guys are from Aero Electronics, which in this show we are partnering together with them. And this is Vojtek from Aero Electronics. Hi, so who are you? So I am Vojtek. I am a software developer. And basically my task was to prepare the software running on this hardware. So now we have, let's say, the Dragonboard with the standard Linux from the Linaro. It's a Debian. And so we have the LTE Shield, which was enabled and it's streaming the data. I'm just going to run the software so you will see the result shortly. Basically, we are testing the speed, but now you catch me on setting up the modem so that is why it's not connected. And the second demo is running. So first of all, we are just checking the speed of the modem. The second demo, on top of that, we are sending the telemetry data from the Shield itself, because it has also some sensors on it. But also, from the second Shield we have here, from this is from the SD, and we have here that are doing the format. And we have the 96 boards, so we are connecting these two boards by the FPGA. And the software here is also, here I built the Linux by the Yocto, it's a Yocto project, and to enable SPI, etc. So here we are taking the, I think, the temperature and humidity from the sensor from the SD Shield and sending to the cloud. Here we are doing this by the Wi-Fi, this by the LTE. At the end, this is going to the Microsoft Azure, where all data are processed. And we run the machine learning algorithm, which is predicting the weather, let's say. So it's taking the temperature, humidity, and I think the lights, I think that the light, because you can put it outside, so this is not the best place to test it. And we can predict what kind of weather we have. So did you have some kind of cloud system that does all this? You do everything through the cloud or what are you doing? So basically the data are sent from the board over the Wi-Fi or over the LTE, and the data are, let's say, sent to the IoT Hub from Microsoft Azure, and then we are sending the data over the function, which is defined on the stream analysis, which is defined on Microsoft Azure, to the machine learning, machine learning, again, from the Microsoft. And the data are being processed, and the model was trained also on the Microsoft Azure. I used the pre-trained model. And according to that, at the end, we have some zero and one. One is the rain, zero is, there is no rain. And is this what's called an open source, available for everybody who buys this? Yeah, so the site from the Microsoft site, so this is available, you just needed the subscription to get it, but basically I think the project I did is open source, so you don't even have the money for that. And so around here is because, so you with Arrow, right? Yeah. So Arrow is doing a lot of providing all this kind of technology to people? Yeah. This is a partnership going on right now with Syrathic? Yes. To get all their meson into work on the 96 boards? Yes. Yeah, so basically Arrow is buying from the Syrathic, the Shields, and the Arrow is specific of the Arrow. We are giving away a lot of Shields and the boards during a lot of events because Arrow, every week, we have some events going on. Where? So around the Europe, I think maybe in the US, I don't know, because I'm part of the European part of the Arrow. You're going to Poland, right? Yeah, so I'm living in Poland, but so basically I used to go to the Europe events, so now the biggest one is Electronica in Germany, we'll be there and we'll have a lot of boards to give, so if you want to have the boards, you can come to our bus. You've been to every minute road? Yeah, I've been there and we also give away like, I don't know, about tens or hundreds of boards. And so basically we are buying from the vendors a lot of boards, so we are distributing and also we are giving away. So when this works out, potentially this could become the new Raspberry Pi, right? This could be big. The Dragon Board, yeah? All this stuff that we're showing here. Yeah, I think this is the idea. Even now, I don't know if the Shields are the short information, they're working on the Shield with the Narrowment IoT network, so it's really cool. And where are you based in Poland? In Gdańsk. We have the software house in Gdańsk. Is it good for technology over there? Yeah, we have like, for now we have I think 12 guys, and we are expanding this to have 22 or something like that next year and we have from the embedded world to the cloud, so. So how is it to work together with Shira Tech? How is the collaboration? So let's say it's working. It's working. So what's more planned for the future? Okay, so first of all about the collaboration. It works very nice. It's a lot of fun to work with people from all over the world. We are visiting Gdańsk every once in a while and see all the progress and all the software work that is being done there. About the future. So we started with 96 board mezzanine. We have about five mezzanines by now. We plan to develop more mezzanines in the new future and release them and also carrier boards, 96 board carrier boards. This is the next challenge that we are going to face in the near future. We have a lot of work, a lot of 96 boards work to be done in the next year or so and we are very excited about it. And does the whole 96 boards community with the linear arrow and everything provide a good platform for you to do all this? It provides a great platform in the meaning that we get the Linux distribution fitted to the board that we are using. We get all the software support. It's relatively, for me as a hardware designer, it's relatively very easy to do the bring up process of a new board because all the software support, all the interfaces that you need to bring up there and working. So it makes life easier. Yes, open source is great in this minute. All right, but it's a long road to go still in the open source. Many, let's say, beautiful challenges in the future that could happen. To make all this, you know, changing the world and everything, you know what I mean? A lot is going on. A lot of beautiful things are going on in technology in general. All right, cool. And you're having fun too, right? Yes, yes, sure, because now there are always, let's say, the entering, the word of the, as a solution provider, not only distribution. So we are really going to be, that is not only the boarder, but also solution provider. So we'll have like the, we'll make the products for our vendors or like the Shira Tech, we are working there, creating the hardware, they're creating the software. So it's a very good relation that we are, we are close to the hardware. So we're creating the software which is really working good on this hardware. Is the coolest kind of software development? Yes, it's close to the hardware, right? It's very cool. Optimized for the hardware, it's the most interesting. Because we see that something's not working so we can, you know, we are just sending the one email without all this process and ask them, okay, could you send this? And then, for example, the Shira Tech is supplying, okay, you have to, for them to solder something to the board and test it then. So we are soldering, testing, I said, okay, it is working. So thanks to that, let's say you can make the better software or better boards and then you can provide the customers and they can ask to make the software or they can create by themselves. How hard is it to develop a new app or new application for this? So to be honest with you, creating on this kind of project is very amazing and I think that everybody from our team is working after the working hours because this is so interesting. So we have the new hardware. You can do whatever you want with this hardware. You can even change the hardware layout, whatever, by contacting the guys. And so I would say that it is easy. For example, one of these, right? Yeah. There's a lot of potential. There's a lot of new applications that could be developed for this, right? It could be like, I don't know, for a cow or something to check where they are. I don't know, but there's so many potential applications. So that means you have to limit yourself and say, I only want to do this idea right now. I don't want to have too many ideas, you know, or something. Yeah, so the best idea that you are, let's say, network independence so you don't have to use the Wi-Fi. Like now, we had on the Wi-Fi DNS problem that the hotel was providing DNS with the private IP and we need to change and modify the board. So move the LTE. We don't have this kind of problems. We just plug it and play. We have the SIM cards. And it's really easy to have the demo working over network. And I think that the future is like most of the sensors, the different, we will, I think, we'll use the LTE or the Narrowband IoT to communicate not only the Bluetooth like for the gateways. So I think we will get rid of this kind of gateways. Cool. So looking forward to this kind of 96 boards, huge success in the future. I mean, thanks to your work. Yeah, thank you. So one more thing about the potential that you mentioned about these boards now, IoT is booming and everything is getting connected. For example, with this model, the LTE Card 4 model, you can connect any device that needs high bandwidth and need to transfer a lot of data. So we have this solution for this kind of projects. On the other hand, we have a very similar mezzanine. This one is a new LTE mezzanine that we are just launching. Actually, this is just from the production facility. We got it a few days ago. The first one out of the SMT. Yeah, yeah, something kind of, yeah. And this is, for example, for the other end of the project. This gives you LTE connection, but in a very low bandwidth. So if you were talking about the cow that you want to know where it is and monitor what is it doing and how is the milk production going on, this solution could be great because it's low power. Like longer range? The range is the same because it's connected to the cellular network. But if you need a solution that can, needs to deliver a very low bandwidth, small quantities of data, it needs less power. It can be running on battery for a few years even. So this could be the solution. So we have a solution for every project. For the high end, for high bandwidth applications and also for low bandwidth applications. But to make a new solution sometimes is expensive, right? So you have to limit yourself how many things you want to do, but it seems it looks like you're doing so many already. We do so many because Shira Tech solution is in a very interesting position. Our mother company, AY Electronics, is a supply chain manager. So they're experts in getting devices. They have huge warehouses with all the devices you want for hardware designer. It's like a kid getting into a candy store. So we have access to a lot of devices. They have their own manufacturing facility. So if I design a new board, the production cycle is very fast. It's very short. And I have the ability to deliver new products very fast. And this is the main reason that in a few months we already delivered five different products and more to come. And what are the prices for these? It depends in our... Of course the price is good. It's the best price, but it depends how much Arrow wants to earn. Approximately... Arrow's business is not just the development board. It's like the components. So people... They start with the development board, but maybe they want to buy after that one million pieces later. Right? So that's a whole different conservation. Right? But some of the boards, like the LTE cut-for board, it's a little expensive. It's not that expensive, but it costs because the LTE model costs a lot of money. So what, 80, 100, something? Around 100 and something. But it's the high-end solution. High-end solution. The rest of the mezzanines are very cheap. It's the development board. The Bosch sensor mezzanines should be very cheap. The LTE and Arrowband model is also going to be very cheap and this is why we think it's going to be a very attractive solution for a lot of IoT applications. It's going to be in a very affordable price. Nice. And when you're talking about the Bosch, the Bosch is good for what? Okay, so other than the modems, we have the Bosch sensor mezzanine. There's a whole bunch of Bosch sensors in there. A whole bunch of five Bosch sensors starting from temperature, magnetic sensor, accelerometer, continuing with barometric pressure and gas and all those kinds of sensors. All on a single board. Gas, awesome. Yeah, yeah. So all of them on a single board and you can develop the amount of application that you can use it is huge. It's actually huge, yeah. Nice. And the last one? The last one is the FPGA board. The FPGA board is very interesting because in the middle here we have Altera or Intel PSG, Max 10 FPGA. It's a programmable device, of course. But this board, it can serve as a development platform for FPGA, especially for people who want to get started developing FPGA. It's a very nice board. But as it is, if you don't want to develop an FPGA and you want to use it as is, it's actually a bridge. A bridge between the 96 boards on one hand to Arduino, Raspberry Pi and Grove. So it takes four ecosystems, four development platforms and combine them into one solution. So you can have 96 boards and use any Arduino shield or any Raspberry Pi hat that you want using this board. And the price is a little bit higher, right, for this kind of board? No, actually it's very affordable. The pricing is very attractive for this one because we made it with the cooperation with Intel and we get good price for the main component. The price will be surprising for this one. Surprising? Yeah. Do you have no idea? I have an idea, but I'm not sure that I'm allowed to tell. Oh, okay. But it has a very attractive pricing. What I'm hoping is that Intel is very happy with the narrow. Intel should be a big member, right, and come and make some ARM chips. This is ARM FPGA. There's ARM in there. Yeah. This is a 28 nanometer, this one? Yeah, I can't remember exactly how many nanometers. But they also have a 14 one. Yeah, so that's great. So that's working out? Yeah. Cool. All right. So thanks a lot. Thank you very much for all this demonstration here at Shibu. Thank you. You hope you have fun at the Linaro Connect. Thank you. Bye.