 Hello, good morning, good evening, good night, good afternoon, anything, you know, wherever you are in the world. Welcome to the mini summit as part of open source summit. And we're going to talk about LF Edge. As I was saying when we were rehearsing, we have the best of the best brain power all here today at your service to answer any questions down from the bits and bytes of latency to the memory requirements to the applications that are being used for Edge. Edge is an extremely important topic for almost everybody these days. And what I'm going to do is I'm just going to quickly introduce what we're going to do and how the scope is today. And for those of you who are not familiar, I'm Arpeggioshpura, I head up LF Edge and networking at the Linux foundation. And we're going to sort of talk about the big picture on Edge. Edge is four times the size of cloud computing. For those of you who have missed the cloud revolution, don't miss the Edge revolution. And I'm so glad you are here today. And you will hear everything that Linux foundation Edge is doing to enable this forward and not just in LF Edge, but across the communities and standards throughout the globe. Let's start off with the definition of Edge. This is a million dollar question, which has now become probably a hundred dollar question because we have figured it out as a community. This is a very important slide. You will see this referred and issued as part of our state of the Edge, which is a top level project under LF Edge. Let me spend two minutes on this and it will be very clear. So at the bottom of the screen, you have two big purple and teal things. There's a user edge and a service provider edge, right? And they bleed into each other. It's not a hard cut, but the general assumption from the community is the last mile separates the two. So then if you double click into the user or what is under the control of a user, you may have constraints, you may have devices that are smart or you will have on-prem type gear. Again, depends on the physical aspects, depends on location, depends on a whole bunch of factors. And then as you cross the first mile, you have things like base stations, things like smart central offices, colos, etc. What we have concluded is anything that is kind of centralized, regional, all the way hundreds and hundreds of kilometers away. And what we say about 20 milliseconds of latency, that's not Edge. So let's be very clear. Not everything is Edge. And I always start off by saying, people say, not all IoT is IOT, not all Edge is IoT, and then not all Edge is the same, right? And so we have a series of projects here that span the various types of Edge locations and Edge terminology here. We will go in detail on each of these projects, specifically starting with our Stage 3 project, which are impact projects, Edge X, Foundry and Acreno, as well as our Stage 2 projects, which we'll get into in the later part of the mini-summit. But before we get into that, I want to spend a minute on showing you how Edge fits in with the rest of the open source and standards community. And this picture is putting it all out, right? You have on the left-hand side, you have all the traditional ways of entering a network, whether it's a telco network through ORAN or an enterprise network. And then that comes in through a core sort of a network, whether it's a public cloud or a hosted cloud or any of the interconnects. These logos are essentially projects that both Linux Foundation, as well as our sister foundations, right, whether it's OpenStack or Kubernetes or anybody, are hosting to enable end-to-end use cases using open source as the building blocks. And the thing that I want to emphasize here as part of LF Edge is we work extremely close with our consortiums and standards organizations. So whether it's at CMAC, ACC for the Automobile Edge Consortium, or IIC, it brings the total picture in perspective. And so the use cases we'll talk about today, these are all based on the projects that we discuss, right? Whether it's Home Edge for anomaly detection or whether it's DevOps at scale for EVE or whether it's telco blueprints, ARVR or whether it's building automation or whether it's predictive maintenance. All these are being discussed today. So stay tuned and be very attentive. Ask as many questions as you want. We will be answering this towards the end of the mini-summit. And with that introduction, I'm going to hand it over to our first presenter and the TSC Chair for the Accraino Edge Stack, Kandan. Take it away. Thank you, Arvath. Thanks for a great introduction into the LF Edge. And I welcome everyone to this summit. And this is Kandan Katherbill, TSC Chair for Accraino Edge Stack. So I'm going to talk about what is Accraino Edge Stack and how it is actually delivering a solution to support different Edge Stacks. So you can see this picture. All Edge deployments are not same and they have different requirements from a power cooling, different sizes of requirements and latency, the security needs, and other aspects like automations and even the connectivity what goes into those deployments. So when we talk about these different type of deployments, you do need a fully integrated solution that can actually again be deployed into this multiple type of Edge location. So for example, it could be a customer location where a different varieties of Edge compute goes in. Then you have access location, primarily to support RAM and connectivity. Then you have a telco network edge where the telco deployments are including core as well as supporting workloads for other customers as well. So we have a different requirements with respect to all this Edge computing. The fact is that there was no fully integrated solution before Accraino Edge Stack and we will talk a little bit about what is exactly Accraino Edge Stack is delivering that supporting this different type of Edge computing. You can see the mission of Accraino Edge Stack. There are four key items that this community is focused on. One is to provide this end to end stack by taking different open sources and doing a development with respect to the Accraino Edge Stack and bringing a cohesive Edge solution for the deployment. So we call collectively as a blueprint. Then we also have projects where specifically enhances features in of the components that goes into the Edge Stack and that including security scanning, that including the items like automated tools to deploy this particular location, operational tools, for example, those sort of items, we call them as a feature project. Then as Arpit pointed out, it is very important to collaborate with all the communities that are developing different parts of the Edge solution. And we work with a lot of upstream community that including ORAN, a lot of open source like OpenStack, a lot of project with an LFH and CNCF and many other open source community. We work very closely with them to bring an end to end solution that can satisfy different Edge Cloud needs. And we also work with a lot of vendors and they are heavily engaged in Accraino to actually take this stacks and support them for the end user who is actually doing this deployment. So if you see this particular picture, it explains how and how many people are actually involved in the Accraino Edge Stack. So we have 40 plus companies has been engaged with respect to Accraino and the 70 percentage of the LFH members are contributing to this project. So the Accraino community with respect to the releases, they have delivered a solution, Edge solution like a fully integrated tested and in actual labs in Accraino and they have been validated to specific use cases and that including wide varieties of use cases. Either it is a telco or it's an enterprise applications. They are all validated by the community and they are readily available for download and a full flush to documentation is given for the user in explaining how this particular stack will work with respect to a specific hardware and what has been tested and what are the results and stuff like that. The solution includes all those edges that I talked about. Also it includes the enterprise applications like connected cards and stuff like that which we will see in the next slide. So before we go into the specific blueprint I do want to talk about a little bit about what does the blueprint really mean because we use this terminology quite a bit and I just talked about the fully integrated stack. The blueprints are not just a paperwork. This is a full integrated solution, the full stack that comes in the form of like a community integrated, tested and fully deployable and end-to-edge stack which has application integration and this is all use case driven. Yes, specific use case of edge computing is being taken into this blueprint as a requirement and then the community integrates them to support for that particular use case and it is fully tested in the community. This is a very key aspect of it because majority of the open source, you know, they develop them but they really don't put them into a testing of all kinds of testing. It could be a security testing, CICD testing and all sort of testing that is needed to make sure that this particular solution can really a production deployable one and that's what this community does it. It verifies the blueprint in the multiple real hardware implementation then the community provides a life cycle support for this blueprint and we do ensure the production quality of this blueprint and in terms of the community, the community has established a wonderful procedure that has been taken and applied to this blueprint whenever they make into the release. The intention here is to provide a low-cost large-scale deployment, zero-touch provisioning and a full industry adoption and based on fully open source and that including white boxes that can be used for this deployment. So that's what we collectively call as a blueprint that's a key deliverable of Aquino and you can see this picture. So primarily it is talks about the blueprint landscape that Aquino has delivered with respect to Aquino release 2 and Aquino release 2 was actually delivered in January 2020 and that including the enhancement to the R1 blueprints and you can see the map here it starts from the left-hand side with thin and thick gateways which goes to the customer location then you have in the right-hand side that is the telco and cloud edge and you have a solution in between which basically provides like the multiple blueprints focused on specific use cases. Some of those blueprints can support multiple use cases but you know they have a primary use case that they are all tested within the community and that's what the R2 have delivered. We are working on the R3 and it's supposed to come in July 2020 and that's going to have like more blueprints and enhancements and features to the R2 blueprints and we are pretty excited as a community to deliver this R3 and this is exactly what the industry needs either it is a telco or enterprise both sectors can leverage these blueprints and we do already see a lot of production deployment of these blueprints by the users and the Karayano community already published who is using that and we are working further with the user community to do more user stories during the R3 which would articulate who is using all these blueprints but the community is quite excited about R3 because it's bringing up like a lot of innovation in terms of many many many applications that can be supported using these blueprints and how do you can get engaged? This particular page has links there are many ways you can actually get engaged you do not need to be a member of LFH or a Karayano to join the community participation either to contribute a project or actively involve in any project all the discussion in the community is open and anybody can actually join and you can see that there are multiple community calls there is a community calendar in which you can see all the calls that's happening with respect to the community but you can actually start with the TSE calls and the community call where like publicly we share a lot of information about innovations happening in this project that happens on Thursday 6 to 7 a.m. PST and you can also see these links where you can actually find additional details that including the mailing list you know the list of blueprints and the Wiki has all the details and recordings of the previous events as well as the calls that the Karayano community have done and we welcome you guys to the community again this is a rightly needed community to support the edge computing need and the 5G deployment and all the cool applications that's happening in the industry that including like connected cards AR, VR and all the stuff so there's a lot of exciting and a lot of innovation happening in this community you are welcome to join us and participate in this community so with that I'm going to hand over to Jim and he is going to talk about EdgeX Foundry Jim Hey thanks, Kannon really appreciate it hi everyone I hope you've enjoyed a wonderful OSSELC really happy to be here with you today and with my fellow LF Edgers I'm Jim White CTO of IoTeC and the TSC chair of EdgeX Foundry I want to kind of give you a quick fly-by of EdgeX what it is what it does and our current momentum and roadmap and how you can learn a little bit more so specifically EdgeX Foundry is an open source vendor neutral project and ecosystem we think the ecosystem is just as important as the project itself as you see with all the people represented here today from our LF Edge community our application is a microservice loosely coupled framework we'll talk a little bit more about that and why that is important we think to the platform we're both hardware and OS agnostic yep we run on Intel we run on ARM we run on Ubuntu, macOS, Windows you name it we run on it and of course we are a Linux foundation Apache 2 project if you're not familiar with open source licensing Apache 2 is the license that says this is business friendly take it use it do as you need with it to help create IoT solutions and we were started back in 2017 so what does all that mean well I like to boil it down for people essentially EdgeX Foundry is middleware it's middleware for the Edge allowing you to connect your things of your IoT and Edge world into your enterprise into your cloud systems but also provide the actuation the command and control back down and importantly provide intelligence or the ability to intertwine intelligence and analytics at the Edge as well and we'll talk a bit about that so without getting into a lot of details given the short amount of time today again it's a project that was started back in 2015 when I was at Dell we've got members online here in fact Jason Shepherd is online who were instrumental in getting that project launched and then we turned it over at Dell to the open source community to Linux foundation in April of 2017 where it's been ever since today again we've got a lot of companies that participate as part of the project and some of them are represented online with us our cadence is we release twice a year we just got done with releasing what we call the Geneva release in May of this year so that was our sixth community release and we're due and on track now for release coming out in the fall around October which we call Hanoi if you haven't detected by the naming scheme of there we kind of follow a naming scheme is similar to Android, ABCDEF type of thing where our names are named not after suites but after some sort of geolocation with regard to our project then what's it about well this is the architecture this is the microservice collection that makes up edgex each of the boxes you see here is one of the services that's currently part of edgex and the idea is there are a set of services at what we call the south side the bottom row of little boxes you see that help digest the data ingest data from sensors and devices and make that available to cloud and enterprise but just as we have the ability to ingest data we also have analytics inside of our environment we provide a default rules engine but you could put whatever analytics local analytics you'd like there to allow for command and control back down to the devices and sensors so again it's a two-way street where you provide the ingestion of data from the edge but also allow for command and control and allowing you to provide your own analytics as a microservice architecture any one of the boxes you see here on the screen is something you can replace with your own code if you so desire we provide a collection of what we call reference implementation services out of the box but the flexibility to provide your value add where you see fit and given our Apache 2 license allows you to do that in a fashion you deem necessary so if you kind of break it down or boil it down what edgex is again is middleware providing the connectivity tissue between the things the edge and then your enterprise your cloud whatever is in your IT world and we provide essentially a dual transformation engine transformation of the south side protocols where you've got things like Modbus and BACnet and BLE and communicate that northward to applications where yes they speak TCP IP but as we all know not all clouds speak the same not all enterprise applications and database speak the same so providing that dual transformation from the thing world into your enterprise and your cloud world again so we look at it as being a transformation engine and one where we're transferring thing protocols into IT data formats and structures as you see fit we are a mechanism because of our microservice architecture as well we lay on top of whatever types of resources you have available we built edgex coming out of Dell with the idea that hey it's going to fit on things like our IoT gateways but from a community standpoint and because of our microservice based architecture you can deploy edgex across whatever type of compute storage and network you have its microservice architecture allows you to lay down the various services from the things all the way to the cloud of the enterprise wherever it seems to make the most sense we're seeing a trend in the industry of trying to get intelligence closer to that thing that device or sensor out there but there are other elements things like maybe your analytics which are going to be a little bit closer to the cloud or at least somewhere deployed in what used to be called the fog and as we're now trying to deem it a more rich layer of the edge our project has considerable momentum in fact these statistics were put together as we were putting together this collection of talks for today and so some of the data needs to be updated a little bit but we have about 180 contributors into edgex today that's actually about I think close to 200 right now in July of 2020 we actually have six and a half million container downloads today and we think or estimate somewhere around 350 to 400,000 deployments so we've got a growing community we've got a growing adapter and user base and we love to see you be a part of either contributing to help make edgex or as an adapter using edgex in your edge solution our cadence has said as I said before was twice a year and we've got a long history behind us already with the six community releases we do have a rich roadmap out there again I mentioned Hanoi as being on the docket for this fall we're going to be looking to improve things like the APIs we're going to be adding more message bus capability lots of details out there on our wiki site which I'll bring up here in just a second so if you're interested in joining our project know that we've got a rich history but also a bright future some key links that you'll want to be familiar with if you start to look at the project and I do hope you will take a look at our code base in github obviously as all open source efforts got to have that code to start to peek at and understand what we do we also have technical documentation online as well as having things like our videos and tutorials online we've got a blog site we've got email and Slack channels as most of the projects in LF edge do please reach out to me through Slack if you do have additional questions or would like a little bit of a deeper dive with that I'm going to turn it over to Jason Shepard from the EVE project and I want to thank all of you and wish you the best and have you have a great conference all right thanks Jim always good to see my partner I'm crying starting edgex back in the day and it just I mean what they're talking about what Jim talked about is really just the power of open source and collaboration and of course we're talking about a variety of projects here and this is really about as Arpit said how do we kind of harmonize over time but also recognize where these projects need to be different and a key part of open source as we all know is this notion of a shared technology investment that helps people avoid what I call undifferentiated heavy lifting we need to be focusing on value but then the other key thing here is driving enabling innovation and while we're also facilitating interoperability so of course that's a key effort of the edgex project edgex is an application layer Acrano of course is kind of talked about is about blueprints and helped to kind of bring things together both upstream and downstream project EVE is actually a project that's below the edgex application layer this is all about providing a very consistent foundation for deploying compute at what we call the IoT edge and I'll talk about that a little bit more detail kind of get a little more granular on the landscape here so next slide okay so at a high level what is what is project EVE you think of EVE as a kind of bare metal computing engine it is not about the applications itself it's really about that lower level device an application orchestration the security of those devices and this could be running on a you know an edge gateway a server outside of a secure data center just any kind of distributed compute it supports VMs containers so if you've got a legacy workload you want to run in a virtual machine maybe it's a SCADA application maybe a point of sale system that runs on windows you know you can still run those on this but you can also run modern containerized applications so we really want to kind of be that universal foundation that provides those extra networking and security benefits that we'll talk a bit more with your choice of hardware applications and clouds of course the whole point of LF edge is to create these projects that are valuable independently but better together you choose what you want to use with it and of course you know it's all about creating these open APIs that prevents you lock in helps you kind of maximize your value add you of course add your specific commercial interest on top this is also of course about commercial stuff but again that shared technology investment not only helps you focus your efforts but also drive interoperability and scale through an open ecosystem okay so this is actually a teaser for a white paper that we're releasing next week through the LF edge we worked as a community and came up with this new taxonomy some people are like oh I've got another another edge taxonomy like you know it's the old standards joke how do you fix the standards problem we'll come up with one more standard but this is a new new taxonomy and it's a very different approach we've had really good feedback on this you know as a community from you know the analysts out there and whatnot because this is different instead of saying things like near and far and thin and thick and whatever you know kind of terms that mean different things to different people this talks about absolute technical trade-offs are you on a WAN or a LAN relative to the process you will never deploy your airbag from from the WAN regardless how fast and reliable those networks are and we know that they've been quite reliable is it in a physically secure data center or not you know when you're outside the data center whether it's a smartphone or an IoT gateway or a sensor device you have different considerations you need a zero trust model you have different skill sets et cetera you know is it is it capable of running application abstraction in the form of VMs or containers you know apps like in the iOS or Android sense working it not these are the hard inherent technical trade-offs across this continuum and let's face it it's a continuum and industrial enterprise consumer they all sit on top of this with their specific needs but these absolute trade-offs never ever change you're always right if you talk about absolute trade-offs so just you know it's important to think about this you know read the white paper when it comes out um there'll be some press on it next week but you know E Project E is focused on these smart device edge meaning two things I am outside of a physically secure data center but I can still run apps so on the left side of the smart device edge and it's called smart device edge because it includes PCs client mobile devices you know mobile devices of course have Android and iOS on the left side of this is is a memory constraint of 256 megs to 512 megs of memory below that you must go embedded everything gets really really constrained and and you know custom on the right side think of this as small server clusters you know so kubernetes is coming down from the cloud it's kind of just taking the world by storm over here we're seeing great work with k3s but you need some unique things because you're getting outside of a data center you're getting to constrain devices so so we're kind of bridging that gap in a in a in a community and of course it's about open collaboration to get there so again think of Eve as doing for the IOT ads these headless IOT gateways and servers and hubs and routers in the wild outside of physically secure data centers creating that abstraction layer think of it doing for the IOT edge what android did for mobile so these are just some quick parallels and these slides are actually versions of these slides are in the project you know I know we've only got a short period of time today so so you know there's a very specific approach that we took here as a community you know Eve is a bare metal foundation now you can run Eve in a in a VM you know just as well but you get the benefits of security functions you know tied to the root of trust in that hardware if it's available distributive firewall I mean literally with Eve you can use that open API with your controller of choice in the back end and govern you know this app can only talk to that app on the box this app can only talk to that cloud I can lock down individual ports maybe you don't want USB ports accessible because they're out in the field but you want to open them for a technician that wants to go install something so very specific needs lots of detail online about it but this bare metal foundation gives you a lot of flexibility and that open API that links to Eve gives you complete control over what you do and no lock in to the proprietary method so if you look at the top right there's a lot of stuff out there that's that that you know has extraction you know available you know supports VMs or some new new efforts that are doing VMs and containers kind of this hybrid cloud native approach you know good stuff but you know if you're an end user and you use that those solutions they have a proprietary API meaning you get locked into that solution you have no choice but to use that controller and so Eve of course with the open API you choose you know which controller you use you know my company we offer and offer but we welcome you know other folks to go build their own in fact we're collaborating you know with folks like open horizon with it that just came with telephage and you know like hey you know we want to go enable other options because this is about floating all boats and and rising tides float all boats the other thing you know that to look at the other approach is this notion of an agent-based solution an agent-based solution means you know I've got an agent running on top of the OS and and that agent does the coordination to down the hardware through the OS and of course to the apps alongside it the problem with this approach I mean there's some good solutions out there but the problem with this approach is that you don't get the benefits of security elements if you're not bare metal if you need to integrate that agent very tightly with every OS who use and of course with with Eve you can use whatever guest OS as you want and out there only support containers you know and the proprietary variants of course there's both open source and proprietary here proprietary variants create lock-in if you do all the integration between the agent and the OS to make sure you don't break that device out in the field and have to drive a truck to spend $1,000 and spend $1,000 to fix it if you if you don't do integration you're going to have these gaps and security risk of breaking etc if you integrate the agent with the OS well guess what you just built Eve and so that's what we're doing as a community and you know we welcome anybody to kind of get involved and make this sort of like the again the Android of the OT Edge real quick I won't go through all this because I want to make sure you know we're on time here but you know various different deployment patterns you can deploy in many different use cases your workload consolidation you've got legacy apps like Skate as I mentioned maybe you're deploying you know a edge component of one of the big clouds or your own kind of containerized TensorFlow models or whatever some sort of controller on the back end and that control like beyond-prem or in the back end SAS or otherwise Edge router we see people deploying it with some some network services and it becomes sort of like the this appliance for secure proxy on the network this is similar the last one is similar to the first one but it's about security analytics you know out of band using a span port I'm kind of sniffing traffic on the network you know like maybe you have an intrusion detection system sort of AI based threat analytics any number of things and EVA is just that kind of one framework you need that's built for the IoT Edge you know 256 megs a memory single note up to a small cluster as we evolve the project and that's all you need and then and then you take your choice of apps so as a road map you know we're always looking to shrink the footprint you know I think we're we're comfortable in a gig today with with apps of course it varies what apps you're running you can run it and even less and what we always want to get kind of that lower point of the smart device edge the IoT component we've been increasing modularity initially we were kind of opinionated as a community we had Zen as the hypervisor well we had people wanting to go implement KVM but then also acorn which is an open source hypervisor very lightweight that can support mixed criticality workloads we basically made a modular so you take your pick and if you only want to run containers and you just take it out we're not trying to reinvent the wheel we've adopted container D is the most popular runtime you know we're bridging the Kubernetes as I mentioned if you're into Kubernetes come join us and and we'll talk about that that bridge but also make sure that is built on a secure foundation that when you're not in a data center mesh networking you know kind of edge to edge to edge capabilities so this is the evolution again not in the data path we are complementary to these other projects go to the web Raspberry Pi support you're not really legit if you don't you know run on a Raspberry Pi for developers so to think what there's an image there go to town community update we're at about 50 unique contributors we've been growing quite regularly we've got deployants out in a number of industry contexts and you know we're always working to collaborate with other projects Ajax Fledge Open Horizon which has a you know a controller that even competes with our own commercial offer but at the same time you know we're all better off if we figure out these sort of standard bases that that we can that we can you know or accelerate our innovations as I mentioned up front some key key links here you know just go go to LF edge and you know even you'll find all these things to open TSC meetings like all these projects it's about you know meritocracy technically the best way to vote is with your your keyboard in right code so with that I'll hand it off to the the panel discussion and get it kicked off and and I'll let me advance it for you there we go all right yeah the team will lead you guys through a discussion on innovation kind of spurred up by the 5G transition all right thank you Hi Jason thank you very much Jason hello everyone first of all big thanks for joining this LF Edge panel my name is Balaji Etirajaloo and I work at Ericsson I'll be moderating this panel so just a quick to talk about myself I spend most of my time in 5G authorization across slicing and edge areas I also saw the LF Edge as a marketing chair and before I start this panel I would like to by the way we put together a great panelist across the industries right starting from chip makers operators and many other areas different vertical segments and now I'm going to introduce my panel panelist just raise your hand so as soon as you hear your name first is Kandan Kandan is a director of AT&T labs and he also leads innovations of 5G and edge computing in addition to working with AT&T he also serves as a TSE chair for a acreno project and also is a member of the technical advisory committee council at Linux foundation my next speaker is Renu Navali Renu Hai so Renu is the Vice President in the data platforms group and she's also general manager of the edge computing and ecosystem division at Intel my next speaker is Sandeep Hi Sandeep yes Sandeep is a global product manager at HP he's leading the HP product solution strategy within the retail global business unit my next speaker is Sajid Balraj Hi Sajid we don't see your video but we see your face Sajid is the director of XR product management for Qualcomm technologies now first of all I really thanks all my panelists taking time and joining us for this interesting conversation around edge I want to just talk a little bit about our topic then I'm going to start asking questions through my panel as we all know 5G deployment has already started and obviously it's going to bring a tremendous monetization opportunities across all industries right not just telco industry just to remain everyone 5G is not just about the radio technologies it also includes edge computing orchestration automation network slicing AI ML and many different technologies that are part of the 5G 5G will leverage all of those technologies with all these technologies coming together with 5G we can enable the whole entire 5G networks as a platform and expose that to various industries and applications right so that can actually create so many new business models and use cases that can actually serve the you know almost all industries so in this panel we talk about business models use cases monetization potential and also touch upon open source okay with that I would like to ask you know direct my first question to Kandan actually Kandan if you're ready you know as you know you are from AT&T and there are a lot of expectations created by the 5G and Edge including you know we hear a lot about Acreno now as the TSHR of Acreno HStack can you share your key insights how you plan to support this technical evolution and also the business model business opportunity that Acreno opens it up Thank you Balaji it's a great question so what is Acreno really doing with respect to you know supporting the Edge computing as you have here then the introduction presentation that I just did a few minutes before and there are a lot of open source in the industry and what is really required is the fully integrated stack for a specific solution and taking each individual open source and putting together is a big effort that's what Acreno is primarily doing it right now so we had two releases already and the release 3 is actually expected of this month in release 2 Acreno delivered multiple blueprints what we call as the fully integrated stack for a specific solutions that including several use cases of telcos either it is access or core or a mixed sort of a deployment or customer home sort of a deployments like SDVAN or universal CPE those sort of deployments Acreno had a solution in release 2 the release 2 came in in January 2020 and these blueprints are fully integrated and again community tested on a specific configuration profile that the blueprint claimed to support for that particular energy use case and these there is a documentation available for user to use it the key aspect is that having a fully integrated solution supported by your open source community that is a very key for the implementation speed and automation that is required by edge computing and 5G and any industry who needs it they need the fully integrated solution and that is what Acreno really provides it so what does it mean for users this is actually provides low cost solution because the community fully integrates it so the users can really see what does it mean to have a fully integrated solution the second thing is that it creates a full ecosystem of vendors and companies you know who consume those blueprints and make a product out of it and also provide support as a open source platform which is actually a needed thing for production deployment for many industries that including telcos and enterprise and also it's very important to you know promote the innovations that is needed in the 5G and as well as the as well as the enterprise application and that means that there need to be a collaboration between a lot of open source communities together and standard bodies something like HCMAC, the 3GPP and those are all needed to make innovation happen faster that's what Acreno community does and it provides like solutions like AR, VR, connected vehicle AI, ML at the edge and instead of having things in the paperwork this community really delivers in a fully working code I think that's the powerful thing this community brings into the industry thank you Kanthan that's good that you brought up standards as well yes there are a lot of collaboration between open source and standards bodies as well that's key and now I'll move to my you know the next question I would like to you know direct this to Mr. Sandeep here but before that audience please also you can ask the panelists questions please type them in the chat and we'll take it one by one those questions please feel free to do that so Sandeep I do understand you know looking at your background you're spending a tremendous amount of time putting together the business model and use cases for the retail industry and can you share some details how you are leveraging in terms of the business model and use cases that you are you know driving toward the retail industry how you're planning to leverage 5G age and AI to solve some of the problems or improve the normal life the experience that we get from the retail industry yep thanks Balaji I think you're right we are definitely hard at work at HPE focused on retail and some other verticals so let me specifically describe the state of retail right and so when it comes to retail it's you know we can arguably say that it's one of the industry that is most transformed by technology at least more than many others and today it starts with a very savvy shopper who really has any info they need at their fingertips and with e-commerce we have seen they are used to that personalization as well as efficiency of any kind of goods or service delivery that they are demanding and for a while now brick-and-mortar retailers have been investing to respond to this both in customer personalization as well as operational efficiencies and they are digitizing their stores with several different types of sensors analyzing that data to create some insights that the store managers or associates can track so in that setting we are talking about problem statements in retail like inventory management customer engagement or let's say theft detection or promotion effect and just to name a few and there are many others and there are there have been some challenges here mostly with cloud-based solutions many retailers may not want to get logged into a cloud service for a variety of reasons it may be the cost or control over the data privacy or connectivity or even latency right so one of the major challenges were just the proliferation of many different types of sensors proprietary protocols and cloud interfaces which created several siloed implementations and we are happy that LF edge actually has provided that open framework now that we just saw Jim go over that is both sensor agnostic and cloud agnostic and HP wanted to be part of this to help push this ecosystem forward so ISPs don't need to build custom solutions that either cannot scale or end up being costly and they cannot take one implementation to the next one so with now 5G and edge compute solutions retailers can solve several of these problems on their premises they can get a you know ready to go edge compute that keeps their data under their control without worrying about privacy or even connectivity and with the right sensor and in size engine recipe that is for their use case of choice they can process all that data locally and create real-time actionable insights either for the store manager or associates or really all the way to the back end of the corporate and we think this is going to bring a you know big opportunity to move this industry even more forward and it will have several business models for the ISPs for the resellers the value chain partners across the board and really excited you know to be working on this at this point in time yeah thank you thank you very much Sandeep Hi Sajith Do you hear me Yeah Balaji I can hear you Thank you Sajith Welcome thanks for joining us you know I know you are highly experienced based on your profile in the AR, VR areas and a lot of use cases that you are involved and you know the virtual reality and the augmented reality they are pretty interesting areas but you know at the you know that use case and new business model coming out of that specific to you know cluster of use cases you know that leverages 5G the edge extreme latency is very critical and also private networks right they all play a major role to deliver this AR VR experience across wide variety of industries and different use cases right now we see a tremendous opportunity and potential in this area from your experience how do you see this whole ecosystem AR VR developing Yeah thanks Balaji and first of all I want to apologize just for the lack of video some last-minute technical glitches with the camera but yeah just going back to your question Balaji yeah so one of the things that we are seeing is we're seeing a lot of interest in AR and VR in the enterprise you know I think the main use cases we are seeing in as it relates to enterprise around training remote assistance and data visualization and collaboration and then what's happening is today basically enterprises are using you know two types of headsets for these for realizing these use cases one is a PC tethered headset and then the other one is a standalone headset and the PC tethered headset is essentially you know a device where you have a headset and all of the processing is done on a high-end PC and you know the user is connected to the PC via a cable you know these devices actually deliver extremely high quality close to photo realistic experiences however you need a dedicated setup right and you also have these cables that actually limit the utility the other the other type of device is a standalone headset now standalone headsets are devices where all of the processing is done on the headset you know these devices are flexible you can use them anywhere and you know they're not constrained by a cable or you know the need to have a dedicated setup so really if you think about it you know the ideal scenario is to have the best of both right so you have the flexibility and ease of use of the standalone headset combined with you know the processing power of the PC right and this is where you know 5G and edge compute comes in you know so now with 5G and edge compute you can think about you know the 5G you know high throughput low latency link acting as this bus right that connects you know processing on the standalone headset and augments it with processing available on the edge right so in terms of what we see so now we see really the next step in terms of evolution of AR and VR in the enterprise is to move to a 5G enabled headset and you you're going to see you know initially bespoke on-prem deployment where you'll have specific use cases for a specific set of users that are going to be realized by an on-prem 5G-enabled headset with edge compute right and then eventually what's going to happen is you're going to have numerous use cases right with all sorts of use cases and you know multiple users on the same system and that's where you're going to see features like network slicing become very important because you have to maintain the quality of users as you basically scale these deployments right so so in summary that's sort of the way we see it right so you have existing standalone PC VR systems that and that are being used to realize those use cases transitioning to you know on-prem 5G 5G connectivity and edge compute and then followed by you know the scaling you know to multiple use cases and multiple users Yeah, thank you Sajid Yeah, absolutely I mean the network slicing can create you know a specific slice for with an enterprise maybe 5-6 slices one could be for the VR use case one could be for different other use cases and also you can expose the capability to the third party maybe to other drone providers to the gaming industry right the many options now I would like to move to Reynu Reynu you are the last person I'm asking the question so you probably need to you know you heard what others are saying but here it's interesting thing what I see you spend a lot of time in building the edge ecosystem and also software and platforms and also worked in the open source industry as well how do you see the open source and many other technologies help to create you know different business models in solving the enterprise problems and use cases now you can feel free to talk you know open now any business model in use case that we have not covered until now right sure thank you Balaji and I think I know the panelists before already reiterated you know they spoke about a number of the benefits of open source than edge computing so I'll reiterate a few things the first thing is when we think about edge computing and what it is it's really about harnessing the data at the edge close to the origin of data and it is to drive you know efficiencies in the enterprises new business models new monetization models and also to deliver new types of services or revenue opportunities for the industry at the base of it all the under the the most important thing is harnessing this data for monetization capabilities or business models and in order to do that you know the technology underpinning is so critical to harness this data and the technology underpinning in addition to hardware in my opinion one of the most important pieces is open source software and open source as well as open source communities that you have in LFS when we look at all the different use cases each of these use cases requires you know a very unique you know supply chain you know ecosystem value chain you know retail industry versus so I think looks like my audio is not that great so maybe I now it's good through the form no not now it's good oh it's now it's good okay little um so so let me continue so when we look at all the you know various use cases they're actually a convergence of you know IT, OT as well as CT or communications technology workflow and this extends not only from the on premises or IoT edge with use cases like enterprise private wireless or private mobility but also on to the service provider or the telco and and for all of this convergence again you know our open source communities or projects like are are so important because one of the most important things like condensate was it is the integration of the various workloads in order to let me try again yes thank you you will now be placed into conference we are some echo I think there is some extra feed is on the line I think is this is this better yeah now it's fantastic okay so so in addition to integrating all of these workloads to deliver on the use cases some of the work that the Kareno community is doing is tremendous right we one of one one example is the the integrated cloud native blueprint that the number of ecosystem partners you know including Intel we submitted where it's not only enables various network functions but also various edge and 5g functions on various Kubernetes clusters the same thing we're seeing extending even to the on-premises edge where there is use of you know Kubernetes for UCPE and other types of workloads where again you want to integrate edge applications and services in addition to the to the network functions so so I I mean my opinion I think the open source community the open source innovation all of that serves as a tremendous catalyst for the ecosystem community and if you know it really helps accelerate our time to market and time to deployment for the various business models and monetization models yeah thank you thank you so much we do we do have a few more minutes left for the panel if somebody wants to jump in in those please feel free to from the panelists right please please jump in and add some more comment anyone can take question yeah so thank you very much for for the you know thank you panelists for for sharing your views to shape up the strategy and edge area and now I will hand over to Matt we'll talk about LF edge more in detail hello everybody where's my my video there we are okay so good morning my name is Matt Trafaro and I'm the chief marketing officer of Aprilio but for the context of this call I co-chair a Linux foundation project called State of the Edge the State of the Edge is one of the top level projects at the the Linux foundation's LF edge and it's a it's a pretty unique project when looking at at open source and one of the things that I'm I'm really excited about is how the Linux foundation has has embraced these innovations in open source and so I'll tell you a little bit of the history of State of the Edge and what we're looking to do now so the the mission of the State of the Edge is to it's very simple it's that to provide free information free research that is vendor neutral and can be used by all to accelerate the edge computing industry and we founded this organization back in in 2007 and it was founded by Jacob Smith and myself Jacob is the CMO of Packet which is now owned by Equinix and we had a number of other organizations Arm, Edge, Gravity, Raffae, a bunch of other folks that helped underwrite the work and we hired some some analysts and got a lot of volunteers and put together a almost a 65 page report that was designed to level set the industry on what the trends were what the the background was and all that and one of the the interesting things and this was 2017 and the first report was published in 2018 and one of the interesting things back then is the there was a lot of frothiness in the terminology if you were here earlier when Jason Shepard gave his presentation on the LF Edge Taxonomy every organization that had anything even remotely to do with Edge or even if they thought they had something to do with Edge had released a a diagram a set of terms you know and it was they were all over the place I mean just different terms and so we were trying to write a book that that spanned all of these efforts and we couldn't figure out which terms we should use so we created a glossary that we just used internally and we sent it around to a lot of peers and got feedback and said you know is this the right definition of access edge is this the right definition of this in some cases we made up words because there wasn't anything to accurately describe it or the words that were in place where you know overused to mean too many different things and out of that came this document that was this glossary for edge computing and I thought maybe there was an opportunity to turn this into an open source project so I approached the Linux foundation and I said I have this glossary it's got a lot of peer support it's vendor neutral it's technology neutral what about putting it in a Giphub repo slapping a created commons license on it and treating it as an open source project and the Linux foundation to their credit accepted as a project it became one of the founding projects of LF edge when LF edge launched and it was a powerful force in the industry and then you know we continued to publish our reports but Jacob and I had always had the vision of finding a long term steward for this project so the end of last year we approached the Linux foundation and said you know couldn't we contribute the entirety of state of the edge to the Linux foundation and that was completed earlier this year the open glossary project was renamed the state of the edge and now we have a a number of activities under there so just to to show you what is available so you know we found this in 2017 in 2018 we first published the open glossary of edge computing that became that open source project we then published our first state of the edge report it's available for free at stateoftheedge.com it's still a very relevant and accurate document in my opinion so I think it's if you're looking for a broad overview of edge computing and how to think about it and frame it I think it's a good document we also created an edge computing landscape and the original version was just a hand laid out PDF file but as we migrated that to the Linux foundation it's now a dynamic landscape which you can see the early versions of it it hasn't gotten to v10 yet but you go to landscape.lfedge.org and there's this dynamic landscape it's based on the the system that the CNCF created which is a database driven you can slice and dice the landscape as many ways as you want if you represent a company or an open source project and it's not on the landscape you can have it added there's instructions on the lfedge wiki so if you go to the landscape page you can find those if you can't find them you can send me an email or hit me on twitter and I'm happy to help but we'd like to get every relevant company and project on the landscape in the beginning of 2019 we published a topic specific report we did it in partnership with some research that seagate produced called date at the edge that also is available under creative commons license from thestaytheedge.com and then most recently at the end of December last year we published the Stay the Edge 2020 report and we have plans to publish a 2021 report this year the 2020 report is a pretty ambitious we built a forecast model for IT infrastructure and data center infrastructure to support the growth in edge computing and all the use cases and it's I think it's a pretty interesting opinion but it's based on on on some pretty buttoned up research done by an independent analyst and so it's at least fun to look at and think about and you know it's a model so it's definitely wrong but it's probably directionally right the so now the Stay the Edge has three working groups um which will seem familiar based on what I already said we have a research working group which is responsible for producing and publishing the reports as I mentioned we just kicked off the planning for the 2021 report if you have interest you should join the the the mailing list we are going to be doing some surveying and some brainstorming on what topic we should cover we're open to just about anything we'll try to be responsible to the community and the other LF edge projects we also have the open glossary of edge computing which was recently updated pretty significantly to fully align with the LF edge white paper and the LF edge taxonomy that Jason mentioned so we're still doing some cleaning up on that but the open glossary now represents the spans you know all the projects of LF edge and is the sort of canonical representation of how the LF edge suggests we use certain terms again it's an open source project so if you see something you don't like or you see something that's missing you can create an issue in get hub or you know create a pull request we're happy to look at all of those and evolve it definitely a community driven project and then as I mentioned we have this edge landscape at landscape. LF edge dot org that we're creating and that is the essence of stay the edge it is a if you have an appetite in involving yourself in an open source project that you know was relatively easy to get on the ground floor this is this is it if you have people on your team that maybe you're in their marketing department or you know our technical managers that have some some language skills that might want to participate that's a great way for people who who you know aren't necessarily prepared to commit code to participate in the community and you know I'm super excited to now be a part of the Linux foundation and part of LF edge it's a really great organization so I will now hand this back to Arpit Joshua Perra to take us to the the next the next presentation all right thank you very much and Matt appreciate all the effort so as we bring in the final section of of today's mini summit I'm I'm really happy that you know I would say 90 percent of you have stayed through the the online which is unheard of right so I'm going to talk about a few of the projects that are very exciting but they they are in the early stages or they're in the growth stage so you've heard from the key projects stage three a cranial and ejects also heard from Eve and state of the edge so I'm going to cover the rest of them today in very brief overview but what I do want to emphasize here is you're seeing an ecosystem of premium members who are behind or the driving force behind the LF edge and you can see from the the the premium members that it it it represents a very wide variety and a cross-section of the community you know silicon agnostics you know arm Intel Qualcomm you got you know the cloud players you know you got Baidu Tencent and you've got the telecom players AT&T entity equinex etc and you know you've got the industrials you know GE OSI soft dynamic I mean very very good cross-section and then you have the end-to-end system players as well so effectively this community is shaping up very nicely we saw almost 150% growth in the developer community in in the last quarter and my theory is you know as as technical people work is the best form of distraction given the pandemic so we're seeing a very steady increase in growth in in the developer community so let me hit on some of these key projects and I will only hit the highlights each of these project details are on the wiki and feel free to you know connect with the ptls so let me start off with a home edge home edge is a project and think of it this way you know it's a it's a samson contribution but it think of it inside our homes you got all these connected devices and historically people would say oh yeah it's a tv and it has it's at top and then there's nest and thermostats came in and now fridges talk and every and and then you have the personal assistance in the house right all of these create their own frameworks their own apis and a cluster of of of unrelated frameworks if you may inside the house now couple that with the fact that there is no it in manager in the house right number one number two the next gen services inside are now more two ways they take advantage of latency so they could be surveillance services they could be you know public urgent broadcast they could be security they could be a whole variety of things so what this project is doing is it's enabling a very simple framework that allows the house or inside the house to sort of run applications on top utilizing common sort of open source code right things like how do I service offload when a device doesn't have the capabilities right that is inside and I'm looking at some of the use cases here how do you maintain low latency without sacrificing the privacy right there's a lot of data that is inside that you don't want to leave so those are the kind of the things that the home edge team is working on they have the first release that came out last year the next one called coconut is planned a lot of VPN data storage modules etc and feel free to join and participate this one of our budding projects the next project that I want to talk about is beetle so as I said LF edge is you know an entity or a framework that allows cloud edge enterprise edge you know all these various markets if you may that each have a have a stake in the ground for the edge to come together so think of beetle coming from a cloud perspective and it is it is a very general purpose platform that has allowed very light and secure applications that need to scale to be pushed out to the edge and again this is from a cloud centric view of things right and so things like drone processing or things like AIML you know just pushing it down so that AI can do visual inspection of video images right and all that needs to be done in a zero touch remote management case right so so you've got all the architecture diagrams which again you don't need to be an expert on this but the general goal of this project is to you know be independent of the hardware right so x86 arms miss MIPS etc independent of the OS and then you know push in services all the way to AIML video and things like that okay now beetle is coming out with with as a container just you know mid 2020 and a lot of remote management capabilities is included in this so you know feel free to participate on that you know good contribution happening there as well and then we have Fledge which is a very fast upcoming project where think of and if you look at Jason's diagram or the diagram I started with you know you saw the user edge in sort of three buckets rights to constrain the sort of the smart device edge and then the on-prem kind of mini data center type edge you know think of this as as kind of very extreme in terms of industrial edge and any any things that allow or need to be deployed as a framework to to the turbines and to the motors and the transformers and things like that right inside the plants the factories the mines etc that's what that's where Fledge helps and you know anomaly detection machine learning etc is all kind of included here in a recent release sort of the Google cloud tensor flow integration was also supported and and I think the summary of the release is really around you know manufacturing energy weight water oil and gas right so constrain environment pluggable microservices and you can see that this is also one of the growing projects here at LF edge and then finally I want to highlight a project that just got in called open horizon it's a you know a seed from from IBM and this this is an interesting project because you know the scale at which you need to push containers and and the number of locations you know 10,000 upwards whatever is is something that you want to do it very seamlessly right so from a compute device to a management hub and vice versa right so and and when you do that you need to include the ML models in it right and you want to make sure that if the devices are constrained and they appropriately understand that so how do you write the policy for that how do you make it fully autonomous so that the agent can run on every node every device sorry to enable it and then you know the lifecycle management is all all created on the back end right and how do you do it autonomously so these are kind of the things that the project to open horizon is is solving it was launched into Q2 and it's being incubated also as a Stage 1 project so you can see from the feel of you know LF edge and the projects that we talked about that our goal as LF edge is to harmonize and unify the communities whether you have a perspective of an IoT whether you have a perspective of an enterprise whether you have the perspective of a cloud service provider or a telco right and there are technologies like AIML 5G you know GPU, TPU, NPUs right on-demand network slicing and things like that they're all coming together to enable these projects to come up our projects are focused on the boring part okay so don't don't get too excited but the boring part is the lifecycle management it's the framework it's the middle where the plumbing layer many many words for it okay it's non-differentiated software which by the way constitutes for almost 70 to 80 percent of the code and I would say 300 percent of the headaches for deployment in terms of interoperability and things like that so that's where open source shines and that's where we have seen a huge huge interest in our communities our mandate from all the projects is to make sure that these projects are hardware independent silicon independent cloud independent OS independent protocol independent I mean you got 170 180 messaging protocols you know we're not going to pick one but we're going to abstract right so that you don't have to worry about it you know cloud connectivity you can have so many clouds there you don't have to worry about it right and I didn't put language independence but you can write it in any language you know coding software language and so we're going to bring in the best of the telecom cloud and enterprise edge right so location latency mobility right as I said five to 20 milliseconds is the sweet spot and as as like any other umbrellas it's hosted by Linux foundation as an umbrella right so umbrella of many projects eight and counting and you know a huge community to work with so you know if you if you got a good teaser and if you believe you're not you're already contributing which I know several of you are not we'd love to for you to become a member I mean that's an easy thing to do but while you do that make sure to get a Linux foundation ID go on to the wiki page join the workflows now what do you do you can do any of these things you can attend meetings attend events just write test cases participate in blueprint just quite ask questions answer questions right and and that's how you get into the community so for people who are already familiar with open source no issues for people who are not this is a very good area to get started remember I said it's four times the size of cloud in terms of the market potential so lots of money to be made and lots of activities going on in fact if you look at some of the Forbes publications that happened in April and May post pandemic or not post but during the pandemic it was very clear that the role of network and the role of edge so 5g iotn edge has significantly increased as we move to the next era so with that said I am going to pause here and I will remind everybody to ask questions in the chat box we're going to answer them in the arrival we're not going to prioritize any but we'll just keep answering them till we run out of questions and with that said I'm going to start off with a whole bunch of questions that have already been submitted and I'll I'll moderate again there's a chat window please add keep asking Q&A if you want to ask a specific question to a specific presenter today please put the name in there and then I'll direct the question towards that person or that project all right so one of the first questions that came out was can you do and I'm not going to speak the name here just to to keep the identity here but one of the first questions was can you describe the Venn diagram of a Crano Eve Edge X Foundry Fledge how should users decide which projects should use should they use all of them that's an excellent question I'm going to take a crack at it and I'll let you know sort of Jason chime in but the fundamental assumption on how users decide is based on their use case you know start with the use case look at blueprints that exist see what frameworks do you need right what problem you're trying to solve and I think most of the users or end users go from that perspective and then start deciding the clear answer on should you use all of them the answer is no that is we have not seen a use case which will need all of the projects all at once that's why the first diagram I showed where we briefly or roughly position these projects in the different definitions of the edge right whether it's the user edge or the provider edge and and then whether it's constrained smart edge or or or prem it kind of roughly gives you the outline of where it needs to be settled so Jason you want to add anything onto that since you've been doing a lot of sort of yeah cross project work yeah can you guys hear me yeah might be a yeah cool yeah yeah that's probably what you're Albert said I mean the goal here is to have all these projects you know again we want to harmonize over time there's a necessary reasons why there's always going to be some differences you you make tradeoffs hence why the taxonomy that we've developed as a community that I quickly tease come you'll come out next week with a really nice white paper there's necessarily trade tradeoffs that you make of course there's some overlap between projects today but we you know we want to be an inclusive community and then kind of work together to go harmonize as it makes sense while also recognizing the necessarily unique parts and and as a community you know it's about being like I said before valuable independently as projects but better together and and if you need just that lower level abstraction layer great you know if you need that application in our probability framework cool you know and you want to use for various degrees of proprietary but if you over time is used more of these projects together you just have this more open foundation that you can build an open ecosystem around which is I think is super important to grow the to the true business potential in any of the markets whether you call it edge IOT AI whatever it's just business so yeah that's why we think it's key that this foundation that we're building of these abstraction layers enable maximum flexibility for your business going forward and minimize undifferentiated heavy lifting very good thank you the next question was what is supported software and hardware platforms targeted initially for LF edge so I believe this was during Cundance and Acrinos presentation so Cundin do you want to take a crack at this question from maybe a Blueprints perspective yeah sure the hardware and software it's really different upon specific use case and each blueprint get validated on a specific hardware that blueprint requirements are and you know it varies quite a bit actually from different sizes of edge deployment to different you know different implementation models and requirements around them so there is no one specific hardware or software again that's why we call us like end to end stack that including the hardware and if a if a user is looking at the documentation of that particular blueprint they can actually see what the hardware and specific stack has been validated and I think this applies to other projects in LF edge as well and they articulate to you know what hardware that specific software has been certified with so you should should follow the document and the community guidance on how it is actually tested that should articulate what the hardware is but again from a community aspect of it we are not restricted to any specific hardware especially anything on the open compute hardware and the open source varieties of hardware and any type of chipset we would be able to support I think that's what community is actually focused on so there is no limitation but the blueprint may be focused on specific type of hardware depending upon the use case Thank you very much I think Yeah, thank you very much I believe Jim might have answered this but does LF edgex support OPC UA for the rest of the audience Jim Yeah, sure and the answer there is yes we do have what we call a device service or a connector essentially for OPC UA one in go one in see they are still just being reviewed right now so I think we'll see them probably hit the next release which is our fall release but you can get the early code and I think to the folks who asked that question I put the links in the chat window but please feel free to reach out and can help you more we also have commercial implementations I know in my own company we have a commercial implementation of edgex that offers OPC UA connectors today Very good, thank you There was a question or there was a question around most of these presentations at the talks about IoT use cases but edgex is more than IoT and we 100% agree with that any comments at the forums that focuses on other use cases so as we said right not all IoT is edge edge has you know a series of other use cases beyond IoT and I think it'll be clear if you go through the blueprints under a kernel it has several non IoT use cases as well so absolutely agree and I mean I the forum yeah if you want to add yes go ahead I mean I'll offer something I obviously want to give others a chance to talk but like yeah absolutely and that's why the taxonomy you know says smart device edge and not IoT edge you know for sure like we as a community believe you know that edge that a lot a lot of different use cases the IoT edge is a sub component of the smart device edge IoT use cases tend to be upload centric whereas you know a non IT use cases like clients for smartphones tend to be download centric and you just need different considerations and that's the type of stuff that Ukraine is looking at and and other projects and you know so there's there's a mix of all of the above IoT is like a workload under edge but there's a lot more than that thank you and while Jason you're at it there was a next question for you as well why is Eve not using Yachto project for integration it seems to have custom packaging well we need is like yeah yeah we use you know you know obviously Linux as a baseline Yachto we're working with some folks in Yachto about how do you run Yachto up above and as some people qualify their stacks on on Yachto OAS or you know any number of Linux distributions and then also if there's more specific questions you'll get involved join the join the TSC and let's talk again we're not trying to be different just to be different we're making technology choices based on what the community is driving and what we think it's the right thing to do for the industry so thank you all right again there's a couple of minutes left if you want to squeeze in any of your final questions so we we have a few more I would like to you know ask Reynu a question since we had some audio issues there but one thing I want to highlight on the call itself is the chair for the governing board on LF edges is Melissa she's from Intel I think the tack chair is from Intel gyms and nature and and you know Reynu is supporting a lot of things so first of all thank you really for for driving this ecosystem forward but one of the things that has come up in terms of you know retail initiatives that intel is working very closely on on Ajax and things like that from a use case perspective how do you see you know that market evolving given where we are today because I think the code is there a lot of these you know businesses have are where we are but you know we are we are ready from a software perspective so any any thoughts on that so I think can you guys hear me okay I know we had a lot of issues can you hear me okay hello can you hear me okay oh okay so so you can hear me okay so I know the retail industry has definitely been you know it's been going through some uncertain times with the with the pandemic but as we think about you know the benefits of what not only open source can do but also edge computing and 5G can do for the retail industry in terms of completely reinvigorating or revitalizing you know the whole customer experience the shopping experience the ability to you know again you know drive efficiencies in the supply chain as well as being able to deliver a variety of different types of virtual or online experiences so when we think about the possibilities of 5G edge computing as well as you know open source and ecosystem innovation I do believe that you know hey the times will change for the retail industry and I do believe that the industry is has an opportunity to you know really go through like an strategic inflection point where it can truly change the entire customer experience as well as the virtual or online experience for for the various for the shopping experiences that the retail industry can offer so that's that's my belief I know the the software is ready the ecosystem is there you know I think you know we we as a population will not you know stop using the retail industry for a variety of different products so I think I do believe that that that industry will go through a strategic inflection point and and use technology in different ways to deliver much better experiences and and the customer engagement term engagement strategy thank you very much and you were coming loud and clear this time so thank you I think with that said I think we are I think the rest of the questions where you know will the session be recorded will the slides be uploaded you know etc etc I think there's one more question we'll probably see if we can take it offline but is it possible to provide required service uptime in container based services so this is probably a longer question on on uptime for container based services so this is more on SLAs and things like that I think several of these blueprints are looking at a hard uptime requirement as well service uptime requirements whether it's containers or VM is is kind of just a technology implementation but kind of if you can answer in like five ten seconds if if that is true or not otherwise we can take that offline with with the person asking nothing you touched on the the exact point it's basically based on specific requirements that the blueprint is trying to address and it's always not exactly the same and also the resiliency come into a lot of a lot of play with respect to the overall availability it could be a site level or it could be a specific hardware level or it could be something you know multi-site level there's a there's a lot of play come into into implementation in assuring the overall SLA of a platform and I think it's not only applies for specifically the the blueprints but also the upstream software right so because a kind of bear is based on most of the projects are based on the upstream software that including Kubernetes and other software that our friends talked on the LF edge and you know things that we get from for example OpenStack so I think in combination of all this thing the blueprints looks at the overall availability needed and it address them okay very good thank you very much so if we have not answered questions we'll take that offline but thank you very much to everybody for answering the Q&A and thank you for sticking around I think we're we're done from from a timing perspective