Enabling and Accelerating the Move to the Network Edge
Sign up for Daily CommTech Brief here.
Session Abstract:
One of the predictions by industry pundits is that enterprises will move key latency sensitive services to edge cloud and will develop applications that will be distributed to multiple clouds – i.e. private, edge and hyperscale public clouds. There is also the belief that 5G will accelerate distributed application development in many industries. On this session we will learn from some of the leaders in the industry on what they are doing to help Enterprises distribute their applications to the edge. Joining us are Neil McRae, Chief Architect at BT, StackPath’s Chief Product Officer, Ashok Ganesan, Rupesh Chokshi, VP of Product Strategy and Innovation at AT&T and Juniper’s CTO, Raj Yavatkar.
Executive Speakers:
Neil McRae - Chief Architect, BT
Ashok Ganesan - Chief Product Officer, StackPath
Rupesh Chokshi - VP, Product Strategy and Innovation, AT&T
Raj Yavatkar - Chief Technology Officer, Juniper Networks
Transcription
ENABLING AND ACCELERATING THE MOVE TO THE NETWORK EDGE
Abe Nejad: One of the predictions by industry pundits, enterprises will move some key latency sensitive services to edge cloud and will develop applications that will be distributed to multiple clouds, that's private edge and hyperscale public clouds. There's also the belief that at 5G will accelerate distributed application development in many industries. On this session, we'll learn from some key leaders in the industry on what they are doing to help enterprises distribute their applications to the edge.
Joining us are Neil McRae, he's chief architect at BT, and next to him is StackPath’s chief product officer Ashok Ganesan. Next, we have Rupesh Chokshi he's vice president of product strategy and innovation at AT&T and last we have Juniper’s CTO, Raj Yavatkar.
Gentlemen welcome.
Everyone: Welcome.
Abe Nejad: Well, it's good to have everybody here. It's the third day of the event here in Barcelona. We know everyone's a little bit maybe tired from all the activity, but we appreciate you giving us your time and I'm sure we'll have a great discussion. Ashok, based on the intro that we just stated, I want to talk about the general trends that are driving distribution to the edge. Can you give us some insight on that?
Ashok Ganesan: Yeah. Thanks very much. So there are a lot of key trends that are happening in the industry. And if you think about the new wave of applications, industrial IOT, or Metaverse or smart health, smart car, autonomous vehicle, edge security, there are many such trends where people need fast response to events. People need fast response to user actions. And the more processing you bring closer to user and data, the faster people can respond. So we see a lot of interest in this industry and there is also this whole trend towards blockchain and what it can enable people in terms of building distributed applications.
So I think it's just starting and we are seeing a lot of interest from various companies to kind of do more things at edge. There is also the need to do more data-driven applications at the edge, where the model development can happen anywhere, but the inference, the data preparation, figuring out insights from data, the closer you are to data, the faster you can get to insight, the faster you can get to taking action. Let's take an example. If you have a video surveillance application, do you want to constantly send video streams all the way to the center?
Probably not. You are better off basically looking at all the video streams, look for anomaly and just take the segment of anomaly and send it to the center for doing whatever they do best. So I think the world is going to be a world of many clouds. And the goal is the new wave of applications is going to take advantage of resources wherever they have it available.
Abe Nejad: We're past the move to the edge.
Rupesh Chokshi: Absolutely. I couldn't agree more with you. First of all, thank you for having us. It feels really good to be live. We're all live in Barcelona after a couple of years. So I agree, when I walk around the show, to me I'm seeing the developers and the creators really thinking hard about the distributed edge, taking advantages of the latency and the networking that we are going to see and continue to see. And then you get into the decision making, the insights, the approach, what do you do with that information?
I walk around and I see a lot of computer vision examples, again and again. I think that is a phenomenal area. I think it's going to be applicable to all industry verticals, whether it is something in the consumer or entertainment or other spaces, gaming, et cetera all the way to business applications and healthcare and manufacturing, even here at the show. I mean just the importance of speed, latency, the network characteristics in that distributed application, the marriages happening, it's a phenomenal opportunity for the next generation of applications.
Abe Nejad: Raj, from Juniper’s stand point, the push to the edge?
Raj Yavatkar: I think the push to the edge is happening. There is no doubt about that. I think from infrastructure perspective, it's an important shift because unlike the traditional networks, now the edge cloud requires a different type of automation and sort of a self-managed networks. Because there are going to be thousands of such edges and you need to provide the kind of service that cannot be provided through a centralized approach, more cloud-managed, more self-managed kind of thing, that's happening. And like Rupesh said, I see that our customers, the Telco see an opportunity now to offer monetizable services using edge cloud. So that's another thing we are looking at how to work with our customers.
Abe Nejad: And from BT’s standpoint?
Neil McRae: Yeah. I think we're seeing lots of edge applications and actually it's not just about latency. It's just about how you build a platform that really gives customers the immersion that they want. So we've got a partnership with the BBC right now, back in the UK called the Green Planet. We have edge compute, we're partnering with Nvidia, they've got all this all these GPUs. And you walk around with an augmented reality. And we have David Attenborough appears as an augmented reality item.
You can't do that from the center. You've got to do it from the edge. So we've put in a private network with edge capabilities, and we're streaming this live AR/VR to people on their own handsets as they walk around using private networks. So that's one example of an entertainment thing that's, showing people's impact on the planet, which is more and more important. But overall, I see the need for compute being in just about any location that you can think of and all of these are edge.
And what Raj said, I think he's underestimated the importance of it, in the way he said it is, automation is crucial in this space. To try and run all this compute, all over the planet, all over somewhere like Barcelona. If you don't have automation, you just won't be able to provide the experience. You won't be able to do it. So automation is crucially important. And also learning from the data about where do you want to put the capability on the edge?
So we are actually looking at some of our legacy mobile thinking in terms of how do we move user plane across the country, as someone's on a train or in a car, and keeping that edge requirement close to the user, even though it may not be a latency driven thing, it's just an experience driven thing? And we see huge benefits from it, both in reliability, but also in scalability.
Raj Yavatkar: Yeah, I think I want to add one more thing. It's a great point. So take things like watch party, like the cricket season is just going on. Ashes Trophy, or IPL or whatever. All my friends are watching the game together. They have slop channels going on, WhatsApp channels going on to share information. One person having an advantage of seeing like a few balls ahead of others, it kind of ruins the experience. So how do you synchronize that? How do you make sure that when people are watching an event together, they don't spoil the experience and they get this augmented experience of watching things together in a geographically diverse and dispersed environment?
It's a huge challenge, but that's, again, to your point, walking around. People are working on things like this. It's exciting to see that they are looking at ways to solve the problems so that the end users get the best experience they can get out of any sporting event. So it's an exciting time
Abe Nejad: Rupesh you mentioned the application developers. I want to sort of put a focus on that for this session, if you don't mind. Let's start with you and talk about how you are allowing application developers to take advantage of the edge. Give us an example. So people out there can understand.
Rupesh Chokshi: Absolutely. And as you said, right in the metaverse, I want to be batting and bowling. So I can get my chance in the field, but back to business. So I think there's what has been more and more important through this learning and journey, is the importance of how you design and deploy the distributed applications. And a lot of it is more modernized. So it's not just take what you had and putting it somewhere. This is rethinking and the network plays a very important component. The edge cloud plays a very important component. So the ecosystem is coming together and we're seeing a lot more sort of modernized platforms for the applications, because then you can deliver to the experience as Neil was talking about.
So let's do some examples. So we have, we'll pick a couple of areas. Hospitals are looking at lots of ways to improve the patient experiences, improve sort of the productivity of the workers that they have. And through the pandemic, we've learned the importance of split second decisions, how do you go about certain things? And we are seeing more and more computer vision being applied that is running at this edge with the network, with the latency, with the jitter, with all of these characteristics, that really make a difference, that if that patient is going to stand up and fall, can you do fall prevention?
If that patient is in pain, can you proactively address something? So we are seeing lots of these examples, I think manufacturing is another very, very ripe area because you have a lot of robotics. A lot of automation. Productivity is very important. So the ability to move things fast, the ability to recalibrate based upon the information flows, et cetera, very important. Ties to what Raj said, the network, the automation, the application, the edge cloud, very important, put together a couple of examples.
Neil McRae: And let me build on that. I mean, healthcare is one of the right grounds for this in my view. And we've been working on what we call the connected ambulance. So equipping an ambulance with a lot more of those sensors, those cameras, those scanner, and then giving the developers of those sensors and cameras, an API, so that they can grab the GPS route that the ambulance is taking from the incident to the hospital, then actually signaling the network, say, okay, I'm going down this path, make sure the network is a hundred percent available because I'm going to send some really high definition MRI scans to a surgeon who's waiting. And effectively, by the time the ambulance arrives, the emergency surgeon that might be working the patient has already done a huge amount of diagnosis.
And the developers of those devices are like, this is amazing. With 5g I say we can turn on this fast path through the network. And as the ambulance gets to the hospital, they're able to really make a difference in giving much more granular, high-resolution data to a surgeon in a way that without edge, we're using an edge platform in the hospital. Without that edge capability, without the ability to signal and give the developers these APIs, we just wouldn't be able to have such a cool service. And the hospitals are loving it because they're able to triage a much faster and then move on to the next patient or deal with patients in a much more effective way.
Abe Nejad: I'm going to get to Raj in just a moment about the challenges in moving to the edge. But before I do that Ashok, anything to add as far as application developers and how they can take advantage of edge?
Ashok Ganesan: So again, one of the myth or misconception is when people think of edge, they think of constrained resources. I think with the technology advancement, we are able to give them really high performance processing at the edge. Really high performance storage at the edge. There are companies building storage technologies that can do million IOPs. And again, you talked about Nvidia and moving GPU closer. We are basically bringing really high performance computing to the edges so that the application can freely write their [12:14 inaudible]. Developers can freely write their applications so that they don't have to worry about, okay, am I going to be in a constrained environment?
And we can actually, the automation technology that all of us are talking about, can basically look at availability of resources and orchestrate and place the application in the right place as close as possible to the user. So the whole industry is kind of working together to solve this problem. So what we want to tell developers is just imagine and let your dreams come through. We will take care of the infrastructure. We'll take care of figuring out where to give you the resources. And this industry is kind of stepping up to help them.
Abe Nejad: Raj, certainly some challenges as we move towards the edge and all the sort of tangential technologies that rely on the edge. What are some of those challenges? And do you see the industry overcoming them anytime soon?
Raj Yavatkar: Yeah. So I think all of you talked about the opportunities and what the developers can do, but I think the real challenges are with the, what we mean by the edge and who provides the edge. So there is so-called near edge, far edge. Far edge is suitable for IOT applications, very close to the census and all, and then you cannot have traditional operating system, traditional hypervisor. You need a very small footprint, very energy efficient thing. Those platforms are not still available. That's still a challenge.
When you come to a near edge, of course, StackPath platform does a great job because you need to be cloud agnostic. If you want the developers to focus on developing applications, they should not worry about infrastructure, which cloud, which edge cloud and who provides it. The hyperscale are moving into the edge cloud with each one of them providing a proprietary stack. But somebody who can provide a cloud agnostic infrastructure where infrastructure becomes, it's taken out of the creation. Application, developers can simply use SDKs APIs, like Neil said, and start developing these applications. Those two things have to happen.
Abe Nejad: Rupesh challenges.
Rupesh Chokshi: I agree. It sounds like the platform is ready, some decisions have to be made. Multi-cloud is here to stay. And automation, the speed. And what we are doing from a network perspective is again, focusing heavy on the sort of the latency, the speed, which the examples are all about speed. Speed matters. And why are we packing it to 5g? Because 5g delivers all of those capabilities. So all of it coming together is sort of an opportunity, but also a challenge and then making it simpler. That is the key, making it simpler.
Abe Nejad: Overcoming challenges in edge Ashok?
Ashok Ganesan: So, yeah, I agree with all the points made. So scaling an application, deploying it in multiple locations and automating it and managing it. It's not an easy problem, but this is where the industry is stepping up, the whole printers, DevSecOps and MLOPs, and the tools and technologies that are coming up to solve the problem is very promising. And again, things that all these service providers are doing with programmable network. Infrastructure, the code. The whole software defined network, software defined everything. It's actually a fresh way of looking at things now. So we basically have programmable metros, programmable 5g infrastructure.
It gives us tremendous opportunity to take compute and processing as close as possible to large point, yes. When you go further, there are constraints, but the thing is not, everything needs to be processed there. And this is where the intelligent orchestration can decide what needs to be processed at the far edge and what needs to be processed at the near edge. So there are challenges, but I think this industry overall I'm seeing a lot of promising technology and this whole blockchain with distributed state management, that's awesome. That solves a key problem for many developers in terms of how do you do distributed transaction? So it's very promising. While there are challenges. I think I see a lot of promise in how people are looking [16:01 inaudible].
Raj Yavatkar: If I may, from a developer point of view. It should not matter which edge cloud and who provides this. And that's where I think the telcos have real opportunity. You can be multicloud cloud agnostic, and the developer focus on developing applications.
Neil Mcrae: Yeah, I mean, I hundred percent agree. I think one of the things for me about edge that's going to be exciting is, as we put more controllability into the 5g network with the Rick, imagine a world where the Rick can detect where those CPU available and suddenly use that CPU to run an edge application. That could be something like a mobile base station on a soldier's back backpack. And he's running in the theater. And all of a sudden they need some sort of capability. There happens to be spare capacity on the cell edge, use that, orchestrate it and then make it feel available.
So I'm really looking forward to how we bring all these compute components together in kind of what I call [17:05 inaudible] that you can grab bits of compute that are near to where you need it to give the experience or the reliability that customers are thinking. And that's one of the things that I think is so exciting about edge is it could be everywhere or in some places. And how we orchestrate overall the compute that's available, if you just think how much compute is available in the show today with everyone's phone, it's huge.
How could we orchestrate that together to solve some of the bigger ones in the world and also give our customers the best experience that we can do because of that compute. And if you think, one of the really simplest UCs. I call this one of the best edge apps, Apple tags, they use everybody's iPhone to show where they are. And that's just one really simple use case that's massive. You can see where your luggage is in an airport, because one of the luggage guys has got an iPhone. That's a fantastic use of edge and it's that sort of application that excites me because who would've thought of doing that.
And how do we then take the great stuff that we've got in that state and blockchain, and really democratize that, so that any bit of compute we can bring into processing those applications and given a really great experience and the choices and the innovation that's here is wild. And actually, if you're here and go down a whole seven, where all the little innovation companies are, they're all thinking about these sort of things. And for me that's amazingly exciting. And our job is to make it easy and make it simple for them to use the network and the edge capabilities
Abe Nejad: Neil, I wanted to get to you. And you already mentioned this a little bit previously about partnerships. I want to get to you on that in a moment, but blockchain as been mentioned twice in this conversation. Again, for the audience out there, that's not as up to speed as you guys are. What's the relationship between blockchain and edge, and provide an example if you can?
Ashok Ganesan: So the whole idea is decentralization. So there is no one authority and it's a community that's kind of validating things. So it's a perfect example of again, how decentralization can actually help various solution providers to solve problems that doesn't require central authority and doesn't require approvals from various organizations. So I think it's a very good example of how distribution is actually going to help various industries.
Raj Yavatkar: But also I think an example could be that if you want to provide a service using distributed edge, the blockchain helped, an example is the project [19:52 inaudible] amount of Stanford. That enable public wifi, shareable wifi across any consumer by using NFTs as a way of paying for them. Similar thing can be enabled in a distributed edge environment where if you have an application that spans multiple edges, you have completely decentralized the service offered by blockchain. You can base your application on top of that.
Abe Nejad: Rupesh anything to add on blockchain?
Rupesh Chokshi: You know. I mean, we're taking some concepts from different industries and putting it all together over here. And the used cases are real, I would say. It's not something that we're trying to find. I think there is an opportunity for the developers to really create some differentiation for their own value proposition that they take to their customers. And we're here to make that happen.
Abe Nejad: Good point. I'd love to talk about that further maybe down the road. Neil, back to you on partnerships, what's the importance of partnerships when we're talking about accelerating and the move to the edge.
Neil McRae: Yeah. I mean, I think it's crucial. And the first partnership is to be open. So let anyone use your platform no matter where they are. And I think we see that the internet is just a story of the more when you make it, the bigger and better it gets. But I think working with the right hardware partners if you're building infrastructure, quite often there's a view that you buy a server, you stick it in a rack and you're good to go. You've got to tune it, you've got to optimize it. You've got to build automation around it. So in my mind, having a really strong partner at the infrastructure level, and then for me and actually I've been working with Raj on this is we want an automation partner that really makes this simple, because if you're not automated, you're just not in this game at all.
So getting that right, and having a flexible automation platform that can scale with the needs that you've got, that hides all the complexity from your customer, so that you're not knifing and forking things through the system, it just happens. And making that platform open so that you can make it available. And one day you wake up and developers use your API to do something amazing. And that's great. You're monetizing that API. You're getting something that probably wouldn't have been there before. So I think that's really important. And then figuring out how you work with the community that wants to leverage your services, how do you help them get the most out of what they're trying to do and give them some support, some leadership, some help. Perhaps some areas where they can experiment on things. I think it's really important.
All of our customers are working with different hyperscalers, different cloud partners. We have to go where our customers are. And the best way to do that, is to work on partnerships. But those that are successful in edge and edge applications will be the companies that are really great at collaboration, that can work with other companies, can work with other organizations and bring them together to create something that is greater than the sum of the parts. And that for me is, you talked about challenges earlier. I think that's one of the challenges we in our industry, we have to go out there and be the cheerleaders for 5g, the cheerleaders for edge and say, Hey business, Hey hospital, Hey, whatever it is you're doing, here's the value that we can bring to this, and then build those agile teams to build these new applications that really make a difference in the world.
Abe Nejad: Yeah. Ashok from StackPath perspective, the importance of partnerships.
Ashok Ganesan: I want to actually amplify some of the points that Neil made. So open standards, that's really key. So I come from networking industry, like Cisco, Juniper, partnered with BT, AT&T, fans, telecom, everybody to kind of build what internet is today. And everything was open standards. NPLS, everything was open standards. There was NPLS world Congress. And there were people who were basically building interoperability solutions. Like interoperability was a key. And I think we need to do the same thing for cloud. The whole federation across clouds needs to become an open standard.
And it's very promising to see things like what's happening in CNCF and Linux foundation and other places. And similarly on [24:24 inaudible], there are a lot of interesting work going on when it comes to segment routing and where things can go in terms of intelligent routing of workloads. So, open standards is really key. In terms of partnership, again, like partnering with service provider is key. Partnering with technology provider is key for someone like us. And of course, partnering with DevOps, SecOps people to kind of simplify automation, that's really key. And of course the main partnership we all want are the developers. We want basically to kind of create an environment where developers can be encouraged to build more distributed applications. And the more the industry comes together to provide them that I think we'll all be successful. And Kubernetes is looking out to be a very promising platform for that.
Abe Nejad: Rupesh, technologies like edge driving partnerships, would you agree?
Rupesh Chokshi: Absolutely. Absolutely. And I think we've already announced some of those partnerships in this space. And I like the point Ashok made that there is opportunity for federation in the clouds. Right now it's kind of viewed as a multicloud, but it's still not as open as it should be. So there's opportunity from that perspective. What I find interesting is that it's one of the areas where I've seen even more interoperability and ecosystem where, whether we're talking about GPU from Nvidia, or whether they're talking about the core networking technologies or the physical infrastructure, or that intelligence software defined network. So lots of different industry expertise has to come together, but goes back to the point about let's make it easy. Let's make it simple, because if you put all the burden on the creator or the developer, they're going to shy away. It's like we just made them the project manager for all of this stuff. And so, we got to do it right. And it's very important.
Abe Nejad: Certainly Raj, certainly Juniper is no stranger to partnerships.
Raj Yavatkar: So I think we need more than partnerships and open standards in my opinion. And let we start with an analogy. When map [26:32 inaudible] was introduced, it was introduced because the stateless distributed applications could be written in sort of a very parallel way. And that provided a library so people can start, then came Spark and Spark led to companies like Splunk and data bricks. They provided an open collaborative, open source platform that anybody could contribute to start using it. We need a similar effort in my opinion, for the edge infrastructure and applications.
Application developers will use open standards, maybe it's decades. But if we can create an source based ecosystem based on common frameworks, then anybody can contribute to that involvement. So I think we can learn from data analytics industry, how it evolved over the years, last 10, 15 years, same approach could be applied here.
Abe Nejad: So I wanted to wrap, and I'm going to go down the line here. I'll start with Neil. Neil, can you give us an example and you've already kind of given us a couple of use cases and one example, give us another example of how BT has helped an enterprise company move to the edge.
Neil McRae: Yeah. We are working with Belfast Port. It's a massive container port. As you know there's a big supply chain shortage. They're trying to figure out, how do we unload containerships quicker. We put edge compute on the site, we've put a whole lot of what we call cameras or servers, and we're able to give them much more visual information and analytics about why it's taking them so long to get containers off a ship. They've taken 20% out of the time that it takes. They're reorganizing all the trucks that come to pick those containers up and given them a much more, if you like, Uber like application that says here's when you want to come and collect your container because it'll be ready to come off. And all of that is only possible because of the edge, because you've got these cameras that are super high definition, the bandwidth that's involved, you just can't take that back to the cloud process and bring it back. And I think that for me is a fantastic example of really using edge in the moment to tackle some real live business problems today.
Abe Nejad: Ashok.
Ashok Ganesan: Yeah. So we are working with a lot of companies to distribute applications. So the ones that I would call out are, we are working with a crypto platform, blockchain platform company. So we are learning a lot in terms of how to give them a high performance connectivity and platform so that they can basically onboard just their platform as well as the ecosystem of applications that are being written on top of that platform. So it's a very learning experience for us. And again, there are a lot of new wave of ideas that are coming up on top of that platform, whether it's next generation gaming, next generation healthcare or next generation retail. There's a ton of things that are being developed on top of that platform. So again, really looking forward to see where this industry is going to go. And I agree with Raj that yeah, open source, we all love open source. And we are all benefiting from that. That community contribution is certainly helping us. And we certainly want to do our fair share.
Abe Nejad: AT&T is supporting enterprise companies move to the edge?
Rupesh Chokshi: Absolutely. You know, lots of examples. We'll start with an interesting one. We are working with a real estate developer, JBG Smith. We've been public. They're in the Virginia, DC area in the US, national lending, which is the second headquarters for Amazon. And the whole idea is to build this sort of smart city, ground up. So we'll put the 5, we'll put the 5g, we'll put the edge cloud right over there. So the experience for the people who are working there and the experience for the people who are residing there is very, very different. And the vision around the smart city is going to be realized in a very different way. So that's one.
Now we've been public about our work with general motors connectivity for all of their locations, cars, et cetera. We just talked about helping ford motor company where they make the electrical vehicles and deploying what we call [30:28 inaudible], which is a multi access edge compute platform, bringing network connectivity and edge together, and I can keep going on. But the idea is that look, every industry, every part of the life is getting re-looked at, and this is our opportunity.
Abe Nejad: Yeah. Raj.
Raj Yavatkar: So we are an infrastructure company and we work with our customers like AT&T and BT. So our approach is to enable edge compute platforms. For example, we work with a company called [30:56 inaudible] systems. They do far edge platforms. So we're enabling applications there so that we can, in an enterprise environment. You can start using IOT sensor and smart devices, and start connecting them to the far edge to start enabling applications. Another thing we are doing, our AIML OPs based enterprise access platform is [31:16 inaudible]. There, we have edge compute at each of the enterprise access locations and we are making the open so people can start writing applications, especially AIML based applications to take advantage of network characteristics, network changes to deliver service.
Neil Mcrae: Yeah. I mean, we are working with Raj and Juniper on using Conrail to signal network paths from edge to core. Our whole 5g cores is based on that. And without that controllability, it's really hard to give the services. So we've got a lot of great stuff going on with Juniper in that space as well.
Abe Nejad: So the title of this session is enabling and accelerating the move to the network edge. And again, I'm going to go down the line. This is the final one I promise. Neil I'll start with you, the word accelerating, clearly there's a lot being done for this move to the edge, but if there is one thing that you feel like, gosh, I wish this part was moving in the open source sector, wish this part was moving a little bit quicker. What would that be?
Neil Mcrae: I think it's how do we do more in automation? It's just such an enabler for everything. Not just in the edge part, but across the whole digital domain. Automation is really what makes things successful or not. So I'd just love to see more things that we have to deal, just be taking care of the automation. There's amazing amount of work going on in that space already. But I think we've just hatched out of the egg and we've got to learn to fly and we've got to soar high and I think there's so much more to come in automation, but that's the key crux for me.
Ashok Ganesan: Yeah. I agree with you, automation is going to be the key. And the more we simplify the automation part, it makes it easy for people to distribute their application at scale and manage it and troubleshoot it. And yeah, automation is going to be key. And certainly again, the new way of technologies is helping us accelerate things further. So the more [33:18 inaudible] labs can certainly help a lot in terms of making sure that we as a community come together to give a lab for developers to build that application. And again, open source communities are certainly valuable in that regards.
Neil Mcrae: And also use that automation to experiment on a live as well. Let's try a few things and see how it goes and iterate quickly.
Ashok Ganesan: Yes. We need to basically get an [33:40 inaudible] lab going on for developers so they can get the best out of the platform that they're working on.
Rupesh Chokshi: What I would say, I agree with the points that are being made. I think simplicity is what needs to be accelerated because there's a lot of information out there. So that creates a lot of analysis paralysis. Should I do it this way? Should I do it that way? Should I ever do it? Should I wait? I'm in a hybrid world? Am I brown field? Am I green field? So that itself like, okay, so the word acceleration, I would back it back to simplify. So we, as the industry need to simplify, so the adoption can be accelerated
Abe Nejad: Well said, Raj.
Raj Yavatkar: For me, that acceleration will happen if I give the analogy to say map [34:24 inaudible] and Spark like frameworks. We need a cloud agnostic, open source, distributed age application framework that is available to anybody. And it starts really evolving very fast so that people can develop applications without having to worry about where to look at the application, where to distribute it, how to distribute the information and so on.
Abe Nejad: Great. Well, it was a good discussion. I think we touched on some points. Certainly we could talk more about all of those points, but we only have so much time today. Everybody has to get back to their meetings and appointments and so forth. I'll start in the end here. Raj, thank you so much for your time. Juniper networks is a big supporter of NMG in many ways, we'll be doing some work coming up very soon, so we appreciate that. Rupesh as always. Well, first of all, it's good to have operators on the sessions. But it's specifically good to have you and your perspective is always appreciated.
Ashok want to thank you and your team for supporting this session, if it wasn't for you guys, we wouldn't be sitting here. So we appreciate that. Again, your perspective as well, and Neil as always, this is your second session with us, I think in a few days or a couple days. So we appreciate your time and again, great to have BT on.
Neil Mcrae: And great to be here in person. Thanks for inviting us.
Abe Nejad: We appreciate it and have it good rest of the day.
Everyone: Thank You.
For any inquiries, please email anejad@thenetworkmediagroup.com