 Hey, welcome to this continuing coverage of the HPE Aruba Pensando announcement. I'm Lisa Martin. Hopefully you've seen by now the announcement from John and Antonio. We're gonna get into some technical details now. I've got two guests joining me. Please welcome Krishna, Dota Penini, the VP of Engineering at Pensando, and Frank Reichstein, Senior Director of Platform Engineering from HPE Aruba. Guys, welcome to the program. Hi, Lisa. Hi, Lisa, thanks for having us. Sure, so we're gonna dig in here. You guys are tasked with bringing these two worlds together. Krishna, let's go ahead and start with you. Talk to me about the announcement, why this is so significant, and then we'll dig into the technical details. Yeah, so as you know, right, Pensando has been in the market for a couple of years right now, and we had a lot of success with the cloud providers, and we're also working with VMware and Project Montrade. So what we learned in the last couple of years, we're trying to take all the lessons, and I would a little bit go into what we learned with the cloud providers. So we took a DSC card, which is a PCIe form factor. The customer takes the DSC card and inserts it into the server with various horses and hypervisors. So it's really exciting that the DSC is in production with some of the cloud providers already, and some of them we are taking to production in this calendar quarter. And we also have been in production with our first generation DSC cards for a couple of years, and some of the biggest banks and storage platform providers. So this is kind of a big deal for us because we are starting with what we call a DPU that Pensando is building, which is the latest generation of it is called a code named Elba, which delivers the software in silicon programmability while matching the performance of hardware. So internally the DPU has the tight integration between special purpose processors that Pensando built, we call MPUs, and a general purpose processor like ARM course where we do the management and control software and with tied together with offload engines like encryption and compression. The key takeaway from this platform that Pensando built is it's programmable at all layers, either by Pensando or our customers, whether it's in data plane using P4 or control and management plane, right? So what in we learned while developing this platform and taking this production with the public cloud providers, we realized that the platform and architecture is not only very highly scalable with very high performance with respect to packets per second or stateful connections per second or NVME IOPS, but it's also adaptable at a very rapid pace. And another key lesson that we learned from our cloud partners is that with the new DevOps model, operations is as important as functionality. For example, the importance of upgrading the DPU pipeline the subsequent guarantees or providing HA or stateful connections. So that in some case with the component field where it's hardware or software, customer doesn't have any disruption in his network or storage operations. So we took all this key lessons that we learned over the last few years and then we are building this new platform partnering with the Aruba team, which is very high-scale with very high performance at the same time tied with very good operations that it comes the best of both platforms from DPU side and from the Aruba side. Frank, do you want to add on the Aruba platform side? Sure, yeah. So the Aruba networking team has been building network switches for the past 25 years. And we've been following all of the trends and evolutions over that timeframe. And as we've gone through a few years ago we decided to make an evolution of our operating system to scale it up for the modern needs of the modern world. And this included doing things like designing with a microservices oriented architecture to provide for a high degree of resiliency throughout the product line and then being able to extend that single network operating system from the core to the edge of the network. As we've been partnering with Pinzando it came very clear that the evolution of the network the next step was this form of a DPU integrated into that top of rack switch to provide a deeper and richer feature set than what has traditionally been available in your top of rack switch. And so this partnership has enabled us to leapfrog what has been traditional top of rack functionality and add to it things that previously were not attainable in that layer of the network. And Frank continuing on with you talk to me about some of the technology requirements and challenges of designing and engineering and delivering the industry's first distributed service switch what were some of those? Sure, sure. So a lot of the challenges around integrating this type of solution come down to how to ensure that you have the highest performance possible and maintaining a high speed of performance when you're now introducing an additional hop within the network topology inside of the switch. A lot of that came down to integrating the background and skill setting capabilities that come along with OSCX that were made it quick for us to enable a new piece of functionality within the architecture. And then a lot of credit has to go to the Pinzando team for the richness of the feature setting capability set that they have within that DPU product as it stands. Krishna, let's go ahead and dig through some of those core features and capabilities that are really going to be benefiting customers. Yeah, so basically right, taking a little bit of step back we started with the DSC market from Pinzando perspective where we wanted to put DPU in every server and we obviously have success in enterprise customers and cloud customers that we discussed earlier. But we also learned a few lessons while deploying DSC in enterprise markets. In the sense that enterprise markets do not need the performance of every DSC at 200 gigi full duplex network services for every server. And also, what makes this product key is that there are a lot of brownfield servers in current enterprise data centers where customers do not want to open up a server to put a DSC in. So we wanted to give a product with a form factor that Frank is talking about and technology that's very familiar to every IT department, given the Aruba OS in a deployment in data centers. And also, as I said earlier about the lessons that we learned, we came up with this, taking this production ready DPU software and hardware which is deployed in public clouds and combined with AOS features that have been rapidly evolving through multiple Aruba releases into enterprise data centers in a switch form factor. So what we think is by doing this, taking the best of both worlds, we are creating a new product category that is for the features and capabilities that are not available in the market from any vendor, specifically providing stateful services at every tour without the complexity of a service redirection because today's data centers, if you want to install services, it's a lot of effort from operator to bring in those services. This obviously also has a great operational model, great TCO and the functionality that customers never see in a tour before. For example, in the first release, we are providing stateful firewall with a visibility at every flow level that goes through the tour which never existed in the market before. New product category, that's a big deal, Krishna. Talk to me a little bit about how long you guys have been at this, you were in stealth mode, crack that open for us. I mean, it has been a less than year but of development that both teams have been doing and we work very closely together and we meet, I mean, for sure, at least more than a week, more than once a week between a pranks team and then send it to them. And there's a coordination between the sales team and the marketing team and the go-to marketing and then how we sell it and the manufacturing team. There's a lot goes on in building this product. I mean, we believe this is the fastest tour, new generation of product that we built because we could do that because the experience of both the teams. Frank, do you want to add anything more to this one? Yeah, I think that really goes to the point here. The capabilities and maturity of the DPU solution that Pensanda was bringing into the solution really allowed for a very fast and seamless integration. On top of that, Aruba OS CX and the platform that we built there with automated API generation and integration with our Aruba fabric composer orchestration layer really created the capability to make things go as fast as possible for this development effort. And so to really take a new product and define a new product space within a 12 month timeframe has been a really exciting and impressive feat by both teams. Very impressive considering the challenges and the dynamics in the market and the global market that we've had. Frank, how big of a lead do you think you have on incumbents here? I think we have a substantial lead on the incumbents here. I think what we're doing is a fundamentally different take on how you do a top of rack switch and the capabilities that we're bringing to bear at the top of rack are fundamentally new and differentiated from what the competition has been thinking about. So I believe we have a substantial lead on the competition. Excellent. Krishna, talk to me about what's next. What's the future? I have some secret sources that tell me that John and Antonia are meeting regularly, pushing you guys. What does the future hold? Yeah, so I mean, obviously this is the start of an exciting journey. This is the first platform you're bringing to the market jointly. And obviously we'll add a bunch of form factors with our upcoming roadmap. So additionally, I mean, the software and silicon performance that with all the services that we deliver as software means the scope and scale of the stateful services that we can deliver and evolve over time. Whether you talk about security or encryption or stateful NAT or load balancing or DDoS, all of this services and then hybrid cloud connectivity. So obviously, there's a lot that we can do with this platform that will be driven by with the partnership with our customers. We also see that the market evolve where all the customers will have, some customers will have DPUs in the servers and some customers will use the new platform that we are bringing together. So we want to evolve the management stack to make sure all of them can be managed uniformly. And anytime, you know, this is a major step for a new category of platform and architecture. We're developing jointly with Aruba and I believe this will be a huge opportunity for both the companies and our customers. And this is exciting times ahead for us. And talk to me, both of your opinions here. Where can customers go to find more information? How can they get started? Frank, we'll go ahead and start with you. Yeah, you can jump to right straight to ArubaNetworks.com and dig in to the feature sets and packages that we have available with the Aruba 10K product line direct from there. Fantastic, Krishna, anything to add? That is correct, actually. So we are treating it as one product coming from both the companies. All the documentation is where, you know, Frank pointed out in Aruba website, we put all the documentation at the same place and we are supporting it as one unified product from both the companies. Are you seeing any, we've seen so much change in the last year and a half, last question. I'm just wondering if either of the HPE Aruba site or the Pensanda site is seeing any industries that might be really prime to take advantage just knowing how many industries all have been affected by the events of the last year and a half. Krishna, any thoughts there? Yeah, I mean, if you look at it right, you know, obviously all of us are working from home and now everything happens, you know, mostly at the edge, right? You know, and we are in that, this platform will help us get there where we get security to the edge and we get more visibility and more services to the edge, right? So I mean, that's what, you know, Pensanda is all about and hoping that, you know, this is this journey that we started with the DPUs, we go with this platform and it'll evolve and it'll help customers, our customers and our partners leverage all the functionality that, you know, Pensanda and Aruba can bring together. Well guys, congratulations on an enormous speed accomplished in not just a 12 month time period but a very challenging 12 month time period. We appreciate you guys breaking down the HPE Aruba-Pensando announcement in more technical detail. Folks can go to learn more information and again, congratulations. Thank you, Lisa. Yes, thank you very much, Lisa. For my guests, I'm Lisa Martin. You're watching this HPE Aruba-Pensando announcement. Thanks for watching.