CEO PoV: Take Control of your Cloud Application Experience
Enterprises are undergoing a revolution in their infrastructure as the application world is being transformed by cloud technologies. Today your applications can be located in the data center, through IaaS, or in a SaaS cloud. Similarly, users are transforming with mobility; no longer are they tied to one location. Enterprises want to solve the problem of connecting users to their applications in a consistent and secure manner.
The overarching issue is the connectivity between these two worlds as we’re stuck in two-decades old technology. Enterprises are therefore challenged with the requirement to transform their infrastructure while integrating underlying legacy technologies, and for the user to access their applications while maintaining policy and security framework.
Join our CEO as he discusses how IT infrastructure is being transformed during the era of cloud technologies, and how Enterprises across a number of industries can embrace the network revolution.
Praveen Akkiraju is Chief Executive Officer of Viptela, the leader in Software Defined WAN technology. His team is building a network fabric which helps customers to dramatically simplify the infrastructure connecting the network edge to applications in the cloud. Praveen has built several billion dollar businesses in data center and enterprise networking over the course of his 25 year career.
Praveen: Good morning, good afternoon, good evening everybody. This is Praveen Akkiraju, chief executive officer at Viptela. Very excited to be with all of you today. This is actually my first webinar with Viptela, so I’m even more excited about what we want to share with you today. So I thought I’d just have a conversation with you guys today to introduce you to this notion of our fabric architecture, and some of the exciting new use cases that we’re delivering for our customers.
And I’m going to be joined by a couple of our top engineers as we go through this presentation today. As a matter of introduction I thought I’d just sort of give you a little bit of my background. I spent 19 years at Cisco, primarily on the routing side, essentially ran every routing platform there is from the big boxes in the service provider networks all the way down to the branch routers. And after that I spent about four years in the data center cloud space essentially building out hybrid cloud solutions.
So it’s been quite a journey getting to Viptela, and you know, as I was sort of transitioning out of my last role it was looking at, I was advising Viptela. And it was really exciting to see what the company was embarking on. And I thought what I’d do is to kind of talk you through a little bit of that journey of where IT infrastructure is going, and the role that networking is playing in that transformation, and how Viptela fits into that picture.
So you know, we are living in amazing times today, right. When you think about the transformation that is underway in IT it is absolutely profound. It is changing the way we as individuals interact with technology, but also you know, we as knowledge workers or IT professionals interact with our technology. And if there are two really underlying themes that define this transformation it’s really about the application experience and the user experience.
It’s about ensuring that applications are first class citizens. They are the only optimization point as infrastructure is designed, and users and their access to these applications is the primary sort of focus if you will as the overall infrastructure comes together. So when you think about the overall infrastructure, you can kind of divide it into three parts. There is the edge, where most of the users reside. And there is the data center, or the cloud, where most of the applications reside.
And as we talked about there is just a fundamental transition underway. When you think about what’s happening at the edge from a trending perspective users are accessing applications that are spread out across the cloud, across the data center. They’re accessing more live content, more of the collaboration essentially is now we have video and live. And it’s all part of making the user more productive, and more connected in order for them to be able to deliver the results the businesses expect. Now what that means for the network is that there has been a fundamental transformation in the way users connect at the edge.
With the advent of mobility, both within the LAN from a wireless perspective, from a Wi-Fi perspective we have transitioned rapidly from connected infrastructure to mobile infrastructure in the campus and in the branch. And then more profoundly as we go outside of the office technology, such as 4G and LTE, which today have a tremendous amount of bandwidth available, whether it’s to a smart device or to a laptop.
So, the connectivity has essentially very closely mirrored the user’s requirements at the edge. So, there has been a transformation that has been underway for over a decade now. If you shift to the data center a similar revolution is underway in the data center, you know, with the advent of virtualization, the initial focus was to essentially get more efficiency out of your IT infrastructure for applications.
But it rapidly evolved to making the infrastructure more friendly for applications. So the evolution of virtualizations and virtual machines towards containers and micro services is a profound transition. Because is fundamentally redefines how applications are built. Now in addition to that the applications today are hosted not just in your data center, but striped across an infrastructure, the service platform could be AWS, could be Azure, or a software of the service platform like SBC, or work day, and such.
So the future application architecture is going to be one where you have applications that are resident in the data center, core enterprise applications, right are spread across an infrastructure, the service platform, or [SAS] platform. So that is the reality that we live in, a fundamental revolution at the edge in terms of connectivity, and a fundamental revolution in the data center.
Now in the data center the first wave of software defined networking has been able to map the network requirement to the applications. So what I mean by that is the requirement was for the applications to be able to access capacity on a dynamic basis. The first stable SDNs back in 2012 delivered by Cisco as well as VMware make the network through segmentation a lot more responsive to applications. But however, that revolution stopped in the data center.
Now if you think about the connectivity between these two worlds, today 90 plus percent of the customers are still connected using MPLS VPNs. Now MPLS VPNs I remember back in 1996 timeframe, you know, along with sort of one of the luminaries in the internet era, [Yacov Rector], had the idea of standardizing the MPLS as a technology for emulating ATM over IP.
That was the original use case, so that technology has served us really well over the last 20 years. At the same time, we’re kind of frozen in time. While bot the edge of the data center cloud have evolved, the connectivity still remains frozen into this technology paradigm. Which is you known, inflexible, expensive, and hard to, and essentially you have no control from an application perspective.
So, what’s also been happening is you’ve had internet connectivity at much higher bandwidth show up, now 100 MB, 200 MB, most of us watch you know, 4K video streams at home with absolutely no problem over our broadband connections. And similarly, most of us access high quality video on our mobile devices using 4G and LTE. So, you’ve had cheaper faster connectivity options that have been rapidly showing up in the enterprise.
So, there is a requirement now to be able to sort of take advantage of this, both from a evolution of the infrastructure perspective, but also from an efficiency and a cost optimization perspective. So, the key requirement is how do you take advantage of these new technologies and sort of evolve if you will to the next generation overlay? So, the core aspect of what Viptela does is really this notion of being able to connect the users at the edge to the applications in the data center with what we call the Viptela fabric.
So, you can think of the Viptela fabric essentially as a next generation overlay that subsumes all the underlying connectivity technology, whether it’s MPLS, whether it’s internet, whether it is LTE, 4G. And be able to provide a saleable, secure, and a programmable fabric to be able to define paths to optimize your applications and your user experience. So effectively think of it as the single platform that is able to deliver multiple use cases, whether it is solving your wide area-networking problem, making the wide area network more flexible.
Whether it is helping you connect to your cloud, and helping your users connect more efficiently to their applications, and infrastructure of the service, or platforms. Or enabling our service provider customers to fundamentally transform their business model with offerings like network as a service. So, we see this fabric as a foundational element of the next wave of innovation in the IT infrastructure. And I think that’s really what is truly unique and differentiated about Viptela.
You know, I’ve always said engineering is about design choices, and once you make those design choices you’re fundamentally wedded to those design choices warts and all. Right, so you accept the trade off associated with those design choices. So, I think it’s really important when you think about all the solutions in the market today to ask the fundamental question where does the technology come from? What are the roots of this technology?
Right, so I’m going to take a minute here to kind of talk a little bit about what [unintelligible 00:09:52] the Viptela fabric, and how we’re approaching this, what are some of the decisions we made as we built this next generation overlay. So, we start off obviously with this notion of a data plane, data plane is essentially where you touch the packets. You’re forwarding the packets, you know, you’re making decisions on sight about the package. So, it’s our V-edge platform [unintelligible 00:10:16] at the edge of the network.
So what we’ve done is we’ve taken essentially a conventional router if you will, and disaggregated it into four different planes. The data plane is the actual platform that’s on site. You can have a software version of it running in the cloud, which is where the packet forwarding happens. The intelligence that essentially defines the paths or the control plane, we’ve abstracted from the platform, and we host it in the cloud. So, we’re able to scale the control plane independent of the data plane.
So that’s a big aspect of what you know, software defined as in most modern architectures are built on. The other aspect here, the management and the orchestration plane are also extremely important, because you think of the management plane as the eye in the sky as a way to be able to manage and administer your network, have the visibility, be able to upgrade your devices, be able to push policies. Be able to configure your overall network from a single point of control.
And be able to push for example a policy that the entire infrastructure inherits instantaneously from a single pane of glass, right? So that’s the true power of the management plane and our re-managed platform, which is also abstracted and is part of our Viptela cloud controller platform. And finally, the orchestration plane, which is an extremely important component. Because this is kind of how you spin things up and down as you scale. You’re spinning up more controllers or spinning down more controllers on an on-demand basis.
As well as bringing up the platforms from a zero-touch provisioning aspect, so most customers don’t get to interact with the orchestration plane. It is what our cloud operations teams use to help run the infrastructure. So, the core of the fabric architecture, the fact that we have disaggregated the platform into these four planes, each of these planes is able to scale independently. For example, we can spin up more containers or more virtual machines with these smart controllers as you add more sites, right. Without disturbing your existing infrastructure.
Similarly, the V-Manage platform is multi-tentative, so it allows our service provider customers to be able to provision and manage multiple customers in a single view. So, we have disaggregated the traditional routing stack in order for us, and using the principles of the internet, and that’s really a fundamental aspect of this. The entire infrastructure runs VGP, including the V-edge devices. And as we know over the past two to three decades we have learned a bunch of lessons in the scaling of the internet. And those are the lessons we’ve incorporated into the architecture as a foundational design principle to allow us to get the kind of scale that we need.
Now the other aspect that’s also important here is the day one security, which comes from the fact that our boxes build keyless IP [unintelligible 00:13:18] Essentially, we don’t have a centralized key controller, that the keys are pre-programmed in order for us to be able to connect and stitch up secure paths. So this is the fabric architecture. There is many uses cases now that we have of underlying architecture that’s scalable that can be supported here. So, what I wanted to do was to be able to speak briefly on a couple of these use cases.
And I’ll invite some of the smartest guys in the company here to come join me to tell the story. Now SDWAN was the first use case we deployed the fabric for. We have tremendous traction in terms of the number of customers adopting our technology. I wanted to highlight a couple of our customers here today, Agilient technologies as well as First American Title. Now when you think about Agilent I think their fundamental requirement was to be able to deliver a much more cost effective a much higher performance platform as far as their applications were concerned.
So, they were primarily on MPLS circuits. They wanted to be able to add internet to the mix, but they wanted to make sure that it was secure enough for their enterprise applications. And more importantly they wanted to make sure that they were able to get more performance, MOS scores for their voice over internet, relative to their MPLS deployment. So that was the key requirement, and so they deployed the fabric. They were able to incorporate the internet and be able to sort of bring all of this up.
And dramatically reduce their costs, but also dramatically reduce the operational complexity associated with the operation of this infrastructure. So, let me give you an example for you know, when you think about sort of acquisitions, companies making acquisitions. So, you bring in a new company, they have their own sort of IP addressing infrastructure, their own sort of connectivity. Normally it takes you months to be able to integrate that into your corporate IP network. So, in this case you just drop a couple of V-edges. The policies are inherited because you’re able to deliver that from the cloud.
And you’re able to bring on this company quite rapidly. So, one of the data points as I continue down this path, you know, we will be sending the recording out to everybody via email. So, if you want to kind of follow this at your own pace you’re welcome to kind of do that. And you know, I think there is a lot of richness here in terms of what this fabric can deliver. So, let me take this opportunity here to sort of invite to join me Arthur Kahn, who is our VP of solutions engineering. One of the founding engineers of this company, who has been a key part of actually building this architecture.
He has a tremendous amount of experience across all the large networking companies, and you know, has helped, is sort of our man in front of customers. So, Arthur, welcome to the show.
Atif: Thank you, Praveen. So, I’m going to walk you through one of the use cases recently. We started on this journey of deploying SDWAN solution at a key healthcare Acadia. And I’ll just walk you through some of the challenges, which they were having, and how we addressed those challenges, and what has that experience been so far. So, I’ll just start with giving you some background, for those of you who are not familiar with Acadia. Acadia is a Fortune 100, one of the largest behavioral health organizations with approximately 575 facilities across US, UK, and Puerto Rico.
So up till now all of their sites have been connected via MPLS, and these are primarily low speed, 1.5 MB per second, MPLS circuits. And one of the major drivers for them to transition to this next generation network has been their need to move to cloud based health applications. As well as staff applications, so they quickly realized that with their current infrastructure network was prohibiting the adoption of cloud-based applications as well as staff applications, such as Office 365.
If you look at–
Praveen: [Talks over] They were not actually able to support any cloud-based applications in their current infrastructure.
Atif: They were not able to primarily because the bandwidth and the current infrastructure. Because if you look at these traditional networks the way MPLS networks are deployed from all the facilities you back haul all the traffic back to your data center, and then you exit to the cloud. And first of all you have low bandwidth, secondly, you’re back hauling your traffic all the way to data centers, which might not be geographically feasible.
So, the primary way or the requirement here is to be able to onramp to the cloud as quickly as possible. And you have to onramp to these clouds in a very secure fashion from all these facilities. And now whether you securely onramp to the cloud from the local facilities using the internet, local breakouts, or you provide regional access in order to onramp to the cloud. So your solution has to provide those optimal onramps where it provides you the best possible way to get to the applications, wherever those are residing, whether those are in the cloud, public cloud, or whether you have to get to the applications in the data center.
Praveen: And one of the other important things you know, obviously with the electronic medical records and sensitivity around privacy, right, there probably was some pretty stringent security requirements as well that needed to be met.
Atif: Yes, absolutely, so security is the key here. So, whenever you exit to the cloud, whether you’re providing onramps from the local facility, you have to make sure that the onramps are provided in a very secure fashion. So, security absolutely is the key, and Acadia quickly realized with the SDWAN solution they were not going to be able to meet those requirements. And at the end of the day they were not going to be able to deploy cloud-based applications.
Praveen: Yeah, and you know, one of the interesting things here obviously, you know, we talk about EMR and voice, but Office 365 is now rapidly becoming you know, the preferred way most enterprises are using the Microsoft Office Suite. And it’s an interesting problem isn’t it, because it’s sort of a [SAS] platform. It’s a kind of one sided, where you can’t actually, we don’t have a footprint in the Microsoft Cloud. Yet you know, customers absolutely want us to be able to provide a deterministic experience to the [SAS] platform. So, they talked a little bit about sort of this notion of how do we solve that problem?
Atif: So absolutely, so when we talk about cloud onramps you have to make sure that you take an onramp which provides you the best experience to the application. Now if it’s Office 365 and you have onramps, which are regional, you can also get to let’s say the cloud locally. So your solution has to be smart enough to be able to figure out which onramp or which path provides you with the best user experience to the cloud. Now you don’t have let’s say the SDWAN device sitting in Office 365.
But then the same time without that there is a need to be able to look at the experience, look at the application experience, and then come up with the best possible route to get to that application in the cloud. So that’s absolutely something, which Viptela has been focused on for the last year or so. And we have come up with a solution here, which provides you with the most optimal route or the path into the cloud applications.
Praveen: Right, and you know, and just coming back to the earlier point right, on you know, the foundational design choices make a huge difference. This is kind of where the fact that our technology is grounded in the principles our routing really allows us to sort of solve that problem.
Atif: Absolutely, so it’s a marriage between routing, and you have to look at the quality of the experience for the application. So now with the next generation networks it’s not just looking at the packets, destination addresses, and just routing the package to the destination. So, there is more to it. So now it’s all about the application performance. You have to provide the best application performance possible for these solutions. Otherwise it’s just a no go, because previously all the applications used to reside in one place, which used to be the data center.
But now Praveen as you also mentioned it’s about connecting the users to applications. The applications can be anywhere. So that’s where the SDWAN fabric comes into the picture, extensible SDWAN fabric with optimal cloud onramps. Plus, the ability to extend the fabric all the way into let’s say public clouds such as infrastructures of service. So, example, we have some of the customers who are moving into AWS, Azure, and they can extend our fabric seamlessly into the public clouds as well.
So, the fabric has to be flexible enough to be extensible regardless of where the endpoints are, and at the same time providing the most optimal onramp to the cloud.
Praveen: Great, thanks Atif, that’s pretty exciting, and I think you know, we can sort of talk about this architecturally as engineers all day long. But at the end of the day when you’re solving real customer problems, as you can see some very tangible business outcomes for the customer, it actually becomes real. So, you know, we’ve, you know, we’re pretty proud of the fact that we’re able to delivery a lot of this actual tangible business benefits to our customers.
So let’s talk about another use case here. So, I’m going to invite one of our really bright engineers, joined us out of school, and he’s a key member of our solutions engineering team, Ali Shaikh. Ali, why don’t you come on and join the show, no drumroll here.
Ali: Good morning, good afternoon, and wherever else in the world you are, welcome.
Praveen: Yeah, so what I wanted to talk to you about is a slightly different take on a use case, right. You know, obviously, you see a lot of partnerships with service providers, and you know, in a conventional sense when I was running the managed services business right, in a large company, you know, for us service providers tend to be sort of a channel in order for us to be able to get to the end customer. I think what’s really exciting here we’re on tract to talk about with you is how we’re actually partnering with Verizon, one of the largest service providers in the world, to allow them to innovate, right, on their offerings to their customers with this notion of a network as a service.
So, I know we have a pretty interesting example here, so just talk to me a little bit about sort of you know, Fifth Third Bank, which was a customer that we’ve jointly been able to deploy a solution at. So, what were some of the challenges, and what’s some of the framework of how we partnered with Verizon here?
Ali: Absolutely, so the way the story goes is Fifth Third Bank being a regional bank in the United States, they were looking for a holistic solution. They wanted to gain the advantages of increased bandwidth, but not just that. They wanted to look at an architecture that would deliver them the agility to expand their business and be able to delivery new services within their own infrastructure.
So that’s the big driver for them. They’re looking for bandwidth increases. They’re looking for the agility to expand their network, and of course having flexibility within the selection of transport is a big one always. Those were the requirements they brought to Verizon and to us that they needed solved. From Verizon’s perspective Viptela was an excellent fit, because Verizon isn’t just some tiny box. Viptela is providing a fabric, the means to deliver services. For Verizon, it was perfect because they were able to delivery a solution that was able to augment the bandwidth that they had at the branches.
They were able to delivery an additional circuit to provide LTE backups from the sites. And so, from an immediate gain perspective the customer immediately sees improved performance for applications, and is also now false tolerant, so they’re seeing the benefit immediately.
Praveen: Right, and this is kind of interesting here because you know, we’ve seen you know, Verizon portfolio expand significantly. I mean they have always been a leading managed service provider globally. But they have added the ability now to also incorporate their internet broadband services and their LTE services as part of their managed service, right? I mean that’s kind of the real power of the solution that they are offering.
Ali: Exactly, and for Verizon, it’s sufficiently a win because they have so many business units they can all become parts of the holistic solution that can provide the customer end to end excellent experience.
Praveen: Right, and you know, what’s also unique about this is Verizon is actually hosting our controllers. And they are providing this network as a service using Viptela as a fabric, but also incorporating some of their own intellectual property.
Ali: Exactly, so what this allows them to do is by virtue of hosting the entire environment themselves, managing it from start to finish, they’re able to have this platform at their disposal to introduce new services for the customer on demand.
Praveen: Right, so Viptela fabric essentially is the foundation of their network service offering, and you know, I think that’s what makes this thing really interesting. Because you know, as I’ve said in the past, you know, service providers are no longer just in the business of selling connectivity, circuits or broadband. It’s really about adding that value-added service, and I think you know, our approach to our service provider partners is to kind of enable them to innovate, right. It’s not just to use them as a channel, but to enable them to innovate.
So, when we announce a partnership it is really about sort of doing some co-engineering, right? Co-developing with them this solution, and Verizon is a great example. They have tremendous capabilities in their cloud in terms of the suit of capabilities that they’re bringing together here.
Ali: Absolutely, this project actually highlights exactly that vision for service providers and us integrating together, in that they will be able to develop whole new ways of being able to provide services, the ones that customers already expect, you know, monitoring, operations, configuration. But also, the next generation of services in terms of being able to augment the networks on demand to provide the kind of business agility businesses expect now.
Praveen: Right, you know, one of the other things I wanted to talk to you about, you know, you spent a lot of time in the field with our customers. You know, talk about sort of how the fabric and the fact that we can incorporate LTE allows us to rapidly bring up a new site, right, even before the MPLS circuit maybe is provisioned.
Ali: That’s absolutely a must win for customers these days, the time to delivery of bringing a site online. They cannot wait for a circuit to be delivered. It could take any number of weeks to months to get a circuit. Having the flexibility to be transport agnostic, and particularly be able to leverage LTE, means that a site can come online when a customer desires it, regardless of when the physical connectivity may get there.
Praveen: So, they start off, for example the site comes up, an LTE connection easily provisions. So, they could come up on LTE, and then switch to MPLS when that circuit is available, right, and it just happens magically. There is no reconfiguration required. The traffic distributes itself based on the policy configured.
Ali: That’s exactly right, the fabric Viptela provides allows you to think of your network from a policy and business logic standpoint rather than having to worry about individual configurations, and circuits, and physical ports. You have to reimagine your network from a policy standpoint. What does the business want? What is the policy? And the network will handle that.
Praveen: And I think this is where the programmer really comes, right. So again, you know, you’re a really hands on guy. So how do you like define a policy and push it out? And how much time does it take if you want to change a policy in say a Fifth Third Bank?
Ali: So from a policy standpoint most of the work is actually just imagining. It essentially is what is it that we need to deliver? Writing the policy and pushing out the changes is a matter of minutes at the most. You make the change. It instantly propagates across the network. The controller based architecture is fundamental to that, by virtue of a centralized control plane.
Praveen: So, you do this in the V-Manage, right, that’s where you make the policy configuration.
Ali: We make the centralized changed in V-Manage, it propagates, whether it be ten sites, a thousand sites, 5,000 sites, the change is instantaneous. The policy goes into effect wherever I want it to.
Praveen: And we have this new policy builder that’s coming out, which is going to make it even more simple for us to be able to define these policies and push them out. So, we’re very excited about that. So, thanks a lot Ali, that’s great to spend the time with you.
Ali: Thank you so much for having me.
Praveen: And hopefully I’m not pulling you away from some really important customer stuff. So, you know, I think as you can see there is really this notion of a fabric is fundamental to what Viptela does differently. You know, it incorporates the best principles of routing, the best principles of cloud, the best principle of simplicity in the architecture, and it has multiple use cases. So, when you think about what makes us different, right, I’d really summarize this into four key buckets. We’re enterprise class, you know, we are the largest footprint in terms of large scale enterprise deployment anywhere from you know, mission critical deployments in banks, in insurance companies, in the manufacturing sector, in retail, in healthcare, pretty much across the segments.
We’ve got great traction, great feedback from our customers. And one of the really important aspects of being enterprise class is the ability to handle [brown fare]. So, what do I mean by that? Right, I mean every customer is kind of somewhere along this journey into the cloud if you will, right, in terms of where their applications are migrating to and such like. So similarly, I think you can go to a customer and say hey, slash cut everything over to this brand spanking new infrastructure. It just doesn’t work that way. So, most customers want to continue to have an MPLS circuit because it provides them the kind of SLAs and guarantees that they are used to for decades.
So, what our fabric does is it subsumes the existing infrastructure as we add new infrastructure, right like an internet broadband connection or LTE, and distributes the traffic across these based on your policies. So that’s really the power of the fabric in our ability to transition our customers from a brown field, right, as they go towards newer technologies it is a key aspect of how we’ve thought through the design. The cloud native aspect of it pretty much as we talked about in the fabric, all of our control plane, our management plane, our orchestration plane, is hosted in the cloud either by ourselves or our partners like Verizon.
So, it enables that sort of central point of control that Ali was talking about. It enables the ability for us to be able to push policies, to be able to upgrade devices, and you know, catch security problems very very quickly. You know, one of the things we’re going to be quite excited about, we’re going to launch pretty soon, is our analytics platform. You know, we get some very clean data from a lot of our edges. And that is extremely useful to our end customers as they make decisions on you know, provisioning of circuits or security and such.
So, the fact that we are able to provide this level of flexibility at different deployment models, different consumption models, APIs to be able to interact with all of our controllers. In fact, some of our customers have essentially taken our APIs and plugged it into their management system, and they’re able to programmatically control the infrastructure through their own portal, right. So, the platform is inherently flexible to be able to adapt itself to the different use cases. So, we believe that we’re at the start of a great journey here. A single fabric, multiple use cases, everything from solving your WAN problem to cloud onramps that Atif talked about, to network as a service that we discussed with Ali.
As well as you know, when you look forward you know, interesting-ish problems like internet of things, where we’re working with some of our large manufacturing partners and customers to be able to enable. So, you know, it’s a great journey, and I look forward to coming back to visit with you guys, hopefully in the next few weeks, so we can share more of this journey. So, what I thought is I’d probably take a couple of questions. And maybe I can also invite my colleagues over here to help me with these. So, let’s start off with the first on here.
Maybe Ali, you could kind of help us with this, can I define policies for different SAS applications to have different internet exits?
Ali: So, for this particular question the answer to short form is yes, you can. The more convoluted answer is depending on your network infrastructure and how far you want to go, how many internet circuits you have, how many sites you have, how many fast applications you have. You can absolutely go as granular as you want, per SAS application you define a specific internet exit to use. Or you can use simply the best available internet exit available at a given site. The flexibility that the fabric provides you is infinitely greater than you will find in your centered infrastructure.
Praveen: So, let me pull in Atif here, right, and talk a little bit about this sort of process of doing an AWS migration, right, and how we address the challenges here. So maybe Atif, you can talk a little bit about sort of how do we function in the AWS environment, right, is the question there.
Atif: Yeah, so as I mentioned earlier AWS, we do have the ability to spin the edge in AWS environments. So, if you have a VPC and AWS we can frontend it with V-edge, and extend our fabric seamlessly into AWS. And now you get all the benefits of the fabric, which now extends all the way to AWS. So, AWS becomes like one node in the fabric.
Praveen: And so, it inherits all the policies, all the security, everything else, and I think eventually we have a transit VPC that we land the fabric on to, and then connect into your application VPCs on the back end. There is another question here, are most of the deployments with service providers mainly in the US? So actually, we have customers that are service provider partners are jointly providing in the US, in Asia, as well as in Australia. And in Europe as well, so here is the interesting part, irrespective of sort of you know, what you see out in the press you know, essentially most service providers work with customers. And you know, we’ve been able to partner with service providers across the globe to be able to deliver the right solutions to our end customers.
So, we actually recently had a pretty big win in Australia with a service provider there, very excited about that one. Because I think at the end of the day as I’ve mentioned our partnership with service providers is not just about viewing them as a channel. It is about enabling their business model and enabling them to deliver their intellectual property on top of our fabric. And I think that’s what makes us unique in terms of the way we interact with them. So yeah, I mean great questions. Appreciate the dialogue here, as I mentioned I’m really excited about this journey.
You can expect more innovation from us as we continue to roll out this fabric. And we expect to come back to you and share some more of this as we go along. So, thank you for your time today, and you guys have a great day, evening, or afternoon, wherever you are.