(upbeat music) >> Narrator: From theCUBE studios in Palo Alto and Boston, connecting with thought leaders all around the world, this is a CUBE conversation. >> Hi, I'm Stu Miniman, and welcome to this CUBE conversation. We're digging in with Pensando. Talking about the technologies that they're using. And happy to welcome to the program, two of Pensando's technical leaders. We have Krishna Doddapaneni, he's the Vice President of Software. And we have here Pirabhu Raman, he's a Principal Engineer, both with Pensando. Thank you so much for joining us. >> Thank you Stu. >> All right. >> Thank you for having us here >> Krishna, you run the Software Team. So let's start there and talk about really the mission and shortly obviously, bring us through a little bit of architecturally what Pensando was doing. >> To get started, Pensando we are building a platform, which can automate and manage the network storage and security services. So when we talk about software here, it's like the better software as you start from all the way from bootloader, to all the way it goes to microservices controller. So the fundamentally the company is building a domain specific processor called a DSP, that goes on the card called DSC. And that card goes into a server in a PCIe slot. Since we go into a server and we act as a NIC, we have to do drivers for Windows, all the OS' Windows, Linux, ESX and FreeBSD. And on the card itself, the chip itself, there are two fundamental pieces of the chip. One is the P4 pipelines, where we run all our applications, if you can think like in the firewalls, in the virtualization, all security applications. And then there's Arm SoC, which we have to bring up the platform and where we run the control plane and data and management plane so that's one piece of the software. The other big piece of software is called PSM. Which kind of, if you think about it in data center, you don't want to manage, one DSC at a time or one server at a time. We want to manage all thousands of servers, using a single management and control point. And that's where the test for the PSM comes from. >> Yeah, excellent. You talked about a pretty complex solution there. One of the big discussion points in the networking world and I think in general has been really the role of software. I think we all know, it got a little overblown. The discussion of software, does not mean that hardware goes away. I wrote a piece, many years ago, if you look at how hyperscalars do things, how they hyper optimize. They don't just buy the cheapest, most generic thing. they tend to configure things and they just roll it out in massive scale. So your team is well known for, really from a chip standpoint, I think about the three Cisco spin-ins. If you dug underneath the covers, yes there was software, but there was an Async there. So, when I look at what you're doing in Pensando, you've got software and there is a chip, at the end of the day. It looks, the first form factor of this looks like, a network card, the NIC that fits in there. So give us in there some of the some of the challenges of software and there's so much diversity in hardware these days. Everything getting ready for AI and GPUs. And you listed off a bunch of pieces when you were talking about the architecture. So give us that software/hardware dynamic, if you would. >> I mean, if you look at where the industry has been going towards, right, I mean, the Moore's law has been ending and Dennard scale is a big on Dennard scaling. So if you want to set all the network in certain security services on x86, you will be wasting a bunch of x86 cycles. The customer, why does he buy x86? He buys x86 to run his application. Not to run IO or do security for IO or policies for IO. So where we come in is basically, we do this domain specific processor, which will take away all the IO part of it, and the computer, just the compute of the application is left for x86. The rest is all offloaded to what we call Pensando. So NIC is kind of one part of what we do. NIC is how we connect to the server. But what we do inside the card is, firewalls, all the networking functions: SDNs, load balancing in all the storage functions, NVMe virtualization, and encryption of all the packets, data of data at rest and data of data in motion. All these services is what we do in this part. And you know, yes, it's an Async. But if you look at what we do inside, it's not a fixed Async. We did work on the previous spin-ins as you said, with Async, but there's a fundamental difference between that Async can this Async. In those Asyncs for example, there's a hard coded routing table or there's a hard coded ACL table. This Async is a completely programmable. It's more like it's a programmable software that we have domain specific language called P4. We use that P4 to program the Async. So the way I look at it, it's an Async, but it's mostly software driven completely. And from all the way from controllers, to what programs you run on the chip, is completely software driven. >> Excellent. Pirabhu of course, the big announcement here, HPE. You've now got the product. It's becoming generally available this month. We'd watch from the launch of Pensando, obviously, having HPE as not only an investor, but they're an OEM of the product. They've got a huge customer base. Maybe help explain, from the enterprise standpoint, if I'm buying ProLion, where now does, am I going to be thinking about Pensando? What specific use cases? How does this translate to the general and enterprise IP buyer? >> We cover of whole breadth of use cases, at the very basic level, if your use cases or if your company is not ready for all the different features, you could buy it as a basic NIC and start provisioning it, and you will get all the basic network functions. But at the same time in addition to the standard network functions, you will get always on telemetry. Like you will get rich set of metrics, you will get packet capture capabilities, which will help you very much in troubleshooting issues, when they happen, or you can leave them always on as well. So, you can do some of these tap kind of functionalities, which financial services do. And all these things you will get without any impact on the workload performance. Like the customers' application don't see any performance impact when any of these capabilities are turned on. So once this is as a standard network function, but beyond this when you are ready for enforcing policies at the edge or you're ready for enforcing stateful firewalls, distributed firewalling capabilities, connection tracking, some of the other things, like Krishna touched upon NVMe virtualization, there are all sorts of other features you can add on top of. >> Okay, so it sounds like what we're really democratizing some of those cloud services or cloud like services for the network, down to the end device, if I have this right. >> Exactly. >> Maybe if you could, networking, we know, our friends in network. We tend to get very acronym driven, to overlays and underlays and various layers of the stack there. When we talk about innovation, I'd love to hear from both of you, what are some of those kind of key innovations, if you were to highlight just one or two? Pirabhu, maybe you can go first and then Krishna would would love your follow up from that. >> Sure, there are many innovations, but just to highlight a few of them, right. Krishna touched upon P4, but even on the P4, P4 is very much focused on manipulating the packets, packets in and packets out, but we enhanced it so that we can address it in such a way that from memory in-packet out, packet in-memory out. Those kind of capabilities so that we can interface it with the host memory. So those innovations we are taking it to the standard and they are in the process of getting standardized as well. In addition to this, our software stack, we touched upon the always on telemetry capabilities. You could do flow based packet captures, NetFlow, you could get a lot of visibility and troubleshooting information. The management plane in itself, has some of the state of the art capabilities. Like it's distributed, highly available, and it makes it very easy for you to manage thousands of these servers. Krishna, do you want to add something more? >> Yes, the biggest thing of the platform is that when we did underlays and overlays, as you said there, everything was like fixed. So tomorrow, you wake up and come with a new protocol, or you may come up with a new way to do storage, right? Normally, in the hardware world, what happens is, Oh, you have to I have to sell you this new chip. That is not what we are doing. I mean, here, whatever we ship on this Async, you can continue to evolve and continue to innovate, irrespective of changing standards. If NVMe goes from one dot two to one dot three, or you come up with a new encapsulation of VXLAN, you do whatever encapsulations, whatever TLVs you would want to, you don't need to change the hardware. It's more about downloading new firmware, and upgrading the new firmware and you get the new feature. That is that's one of the key innovation. That's why most of the cloud providers like us, that we are not tied to hardware. It's more of software programmable processor that we can keep on adding features in the future. >> So one way to look at it, is like, you get the best of both worlds kind of a thing. You get power and performance of Async, but at the same time you get the flexibility of closer to that of a general purpose processor. >> Yeah, so Krishna, since you own the software piece of thing, help us understand architecturally, how you can deploy something today but be ready for whatever comes in the future. That's always been the challenge is, Gee, maybe if I wait another six months, there'll be another generation something, where I don't want to make sure that I miss some window of opportunity. >> Yeah, so it's a very good question. I mean, basically you can keep enhancing your features with the same performance and power and latency and throughput. But the other important thing is how you upgrade the software. I mean today whenever you have Async. When you have changed the Async, obviously, you have to pull the card out and you put the new card in. Here, when you're talking upgrading software, we can upgrade software while traffic is going through. With very minimal disruption, in the order of sub second. Right, so you can change your protocol, for example, tomorrow, we change from VXLAN to your own innovative protocol, you can upgrade that without disrupting any existing network or storage IO. I mean, that's where the power of the platform is very useful. And if you look at it today, where cloud providers are going right, and the cloud providers, you don't want to, because there are customers who are using that server, and they're deploying their application, they don't want to disturb that application, just because you decided to do some new innovative feature. The platform capability is that you could upgrade it, and you can change your mind sometime in the future. But whatever existing traffic is there, the traffic will continue to flow and not disrupt your app. >> All right, great. Well, you're talking about clouds one of the things we look at is multi cloud and multi vendor. Pirabhu, we've got the announcement with HPE now, ProLion and some of their other platforms. Tell us how much work will it be for you to support things like Dell servers or I think your team's quite familiar with the Cisco UCS platform. Two pieces on that number one: how easy or hard is it to do that integration? And from an architectural design? Does a customer need to be homogeneous from their environment or is whatever cloud or server platform they're on independent, and we should be able to work across those? >> Yeah, first off, I should start with thanking HPE. They have been a great partner and they have been quick to recognize the synergy and the potential of the synergy. And they have been very helpful towards this integration journey. And the way we see it, a lot of the work has already been done in terms of finding out the integration issues with HPE. And we will build upon this integration work that has been done so that we can quickly integrate with other manufacturers like Dell and Cisco. We definitely want to integrate with other server manufacturers as well, because that is in the interest of our customers, who want to consume Pensando in a heterogenous fashion, not just from one server manufacturer. >> Just want to add one thing to what Pirabhu's saying. Basically, the way we think about it is that, there's x86 and then the all the IO, the infrastructure services, right. So for us, as long as you get power from the server, and you can get packets and IO across the PCIe bus, we are kind of, we want to make it a uniform layer. So the Pensando, if you think about it, is a layer that can work across servers, and could work inside the public cloud and when we have, one of our customers using this in hybrid cloud. So we want to be the base where we can do all the storage network and security services, irrespective of the server and where the server is placed. Whether it's placed in the call log, it's placed in the enterprise data center, or it's placed in the public cloud. >> All right, so I guess Krishna, you said first x86. Down the road, is there opportunity to go beyond Intel processors? >> Yes. I mean, we already support AMD, which is another form of x86. But other architecture doesn't prevent us from any servers. As long as you follow the PCIe standard, we should, it's more of a testing matrix issue. It's not about support of any other OS, we should be able to support it. And initially, we also tested once on PowerPC. So any kind of CPU architecture, we should be able to support. >> Okay, so walk me up the application stack a little bit though. Things like virtualization, containerization. There's the question of does it work but does it optimize? Any of us live through those waves of, Oh, okay, well it kind of worked, but then there was a lot of time to make things like the origin networking work well in virtualization and then in containerization. So how about your solution? >> I mean you should look at, a good example is AWS, like what AWS does with Nitro. So on Nitro, you do EBS, you do security, and you do VPC. In all the services is effectively, we think about it, all of those can be encapsulated in one DSC card. And obviously, when it comes to this kind of implementation on one card, right, the first question you would ask what happens to the noisy neighbor? So we have the right QOS mechanisms to make sure all the services go through the same card, at the same time giving guarantees to the customer that (mumbles) especially in the multi-tenant environment, whatever you're doing on one VPC will not affect the other VPC. And the advantage of the platform that what we have is very highly scalable and highly performing. Scale will not be the issue. I mean, if you look at existing platforms, even if you look at the cloud, because when you're doing this product, obviously, we'll do benchmarking with the cloud and enterprises. With respect to scale, performance and latency, we did the measurements and we are order of magnitude compared to (sneezes) given the existing clouds and currently whatever enterprise customers have. >> Excellent, so Pirabhu, I'm curious, from the enterprise standpoint, are there certain applications, I think about like, from an analytic standpoint, Splunk is so heavily involved in data that might be a natural fit or other things where it might not be fully tested out with anything kind of that ISV world that we need to think about. >> So if we're talking in terms of partner ecosystems, our enterprise customers do use many of the other products as well. And we are trying to integrate with other products so that we can get the maximum value. So if you look at it, you could get rich metrics and visualization capabilities from our product, which can be very helpful for the partner products because they don't have to install an agent and they can get the same capability across bare metal virtual stack as well as containers. So we are integrating with various partners including some CMDB configuration management database products, as well as data analytics or network traffic analytics products. Krishna, do you want to add anything? >> Yeah, so I think it's just not the the analytics products. We're also integrating with VMware. Because right now VMware is a computer orchestrated and we want to be the network policy orchestrator. In the future, we want to integrate with Kubernetes and OpenShift. So we want to add integration so that our platform capability can be easily consumable irrespective of what kind of workload you use or what kind of traffic analytics tool you use or what kind of data link that you use in your enterprise data center. >> Excellent, I think that's a good view forward as to where some of the work is going on the future integration. Krishna and Pirabhu, thank you so much for joining us. Great to catch up. >> Thank you Stu. >> Thanks for having us. >> All right. I'm Stu Miniman. Thank you for watching theCUBE. (gentle music)

>> Narrator: Live, from Madrid, Spain, it's theCUBE, covering HPEE Discover Madrid 2017, brought to you by Hewlett Packard Enterprise. >> Welcome back to Madrid, everybody. This is theCUBE, the leader in live tech coverage. My name is Dave Vellante, and we're here with my co-host Peter Burris, and this is day one of HPEE Discover Madrid. Susan Blocher is here, she's the vice-president of portfolio marketing at Hewlett Packard Enterprise, and Bruce Trevarthen joins her. He's the CEO and founder, I believe, of Layer X. Welcome back, both of you, to theCUBE. >> Thank you David. >> All right so Susan, big show for you guys, and we have these six months cadence of big messages >> Susan: Yes. >> And customer shows, so what are we going to hear this afternoon at the keynotes? >> Wow. I'll tell you we've got a lot of exciting news to talk about. First of all, the way customers are consuming IT is really changing, cloud is changing the game. We got some amazing announcements to talk about around how we're going to help customers in the hybrid IT space consume IT differently. We're going to talk about how we're helping them manage across multi-cloud environments. We're going to talk about bringing artificial intelligence and machine learning to the data center which is really transformational. So, lot's of exciting news here. >> Good, okay! So we'll be covering the keynotes here just actually in about a half hour or so, we kick off. Meg, Antonio >> Yes. you've got a new leader so we're going to hear from him, we've been hearing from him for some time now. >> Very exciting. >> Looking forward to hearing from him. Okay, Bruce. It's been awhile since we talked about layerX. Tell us what's transpired in the last couple years. Set up layerX, what you guys are all about and what's new. >> Sure, so it's a cloud service provider based out of New Zealand. Multiple platforms giving us that resilience. You know, that sort of general cloud people all know what cloud is these days. But really for us the journey it just continues. We keep, from a strategy point of view we keep looking at where is cloud adoption at, where is cloud going, are these hyperscale providers going to enter every country and every market? And really, sort of, make us, sort of in country boutique operators less relevant. So you're always asking that question and then you're sort of hit with this new wave of expectations down from the clients. Hybrid IT has been the big push in the last 12 months and what's really encouraging for us when we get hit with this new sort of level of interest and a slight tangent on this manage services delivery is that HPE already thinking the same way. They've already come up with a product line that's going to plug that gap. So we work very closely with HPE with their edge line and the OEM team globally, to deliver HPE hardware on customer site or on premise. And then we put our own software on that, we link it back into the core V-grid environment, and that really, for a customer they keep those workloads on site where they need to be. And then you've got that public cloud environment for the disaster recovery and the workloads that don't need to be on site. >> So let's unpack that a little bit. Tagline, Hewlett Packard Enterprises uses make hybrid IT simple, that's the objective. >> [Susan]- That's right. >> You know, IT is complicated, hybrid IT is complicated. What's the starting point to make it simple Bruce, from your perspective? Is it to make the infrastructure as invisible as possible, is it bringing the cloud upward model? Maybe talk about those steps. >> Sure. Well, I mean, one of the first things we try to do to make it simple is we don't mention cloud. We talk ultimately about what workload is the customer consuming and where do they belong? And so, we're invariably seeing more and more workloads that really shouldn't go centralized in a data center they should be on site. So, GPU accelerated desktops for oil and gas research, or some of our clients doing 3D engineering, you know, CAD design work. You can put that in a data center, and we have, but then you're at the mercy of the fiber connections. Speed of the fiber connection, the resilience of the fiber connection, and the cost absolutely. And so keeping some of those workloads on site just makes sense. But how can you then leverage the benefit of that centralized IT in the event of a disaster if all of your workloads are actually on site? And that's where it's got to be hybrid. You can have those workloads on site but all your files and all that capability is sort of mirrored in the cloud environment. So if you have a fiber cut, then you can use a cellular network to get there. Or if you have an on site disaster, then you can spend the equivalent resources in the data center, but on demand, rather than dedicated to you. >> We like to say that customers want or the way that we summarize it at Wikibon is, customers want the cloud experience where the data demands. >> Dave: 'cause we do talk about cloud >> 'cause we do talk about cloud periodically. Well, but you have to, because at the end of the day it's driving a new way of thinking. Not just about the technology, but how you solve business problems. And it comes back to how do you think about the business problem differently. I love New Zealand, I've been there a couple times. I've worked with a lot of customers and the minute that you said New Zealand I was like, right! How do, how does the cloud experience, how are you solving problems differently than you did a few years ago because of not only the HPEE partnership, but thinking differently about these problems? >> Thinking differently is definitely something you have to do to stay relevant, right, to keep up with the market. Almost ten years ago we thought what we felt was a little differently, when we adopted the HPEE 3PAR, and that really was a technology that gave us the ability to change our mind regarding storage. Spin forward now to 2017. In April this year we put in our first HPE Synergy platform. This month we're just putting in our second HPE Synergy platform. And Synergy gives us for compute what HPE 3PAR gave us for storage. The ability to change our mind, to be programmatic or autonomous with the deployment of resources for a customer need. And so for a public cloud environment, that's basically spinning up compute nodes as required for the demand within the clusters. But it also introduces by way of the technology capability, a new channel, or a new revenue opportunity. Because now we can actually programmatically spin up compute nodes of any flavor, for a customer in a private cloud environment. So this is physical tend to the customer opposed to virtual, you know, cloud. We can do that just as easily as we can a VN because of Synergy. >> And that's really exciting. I think what Bruce is really representing here is that he can focus on business outcomes for his customers. And you, Dave, you said it makes the infrastructure transparent. Transparent but underneath that is really differentiated capability and value like the ability to spin up and spin down composable infrastructure on demand. Like the ability to bring world class security to that infrastructure. So all of those things are underpinning the services that layerX is able to deliver. >> So I would think part of making Hybrid IT simple is not just throwing a bunch of products at your customers. >> Right. >> We heard on the last financial call that HPE is changing the way... >> Exactly. >> ...it reports. It's going to report hybrid IT, which is essentially your portfolio. >> Susan: Exactly. >> So it's server, storage, networking and relevant services around that >> That' right. >> Susan: And software. >> And software that powers all that, so talk about how you're going to market and how that aligns with how you guys want to buy. >> Yeah, well think about it from, let's talk about it from the layerX perspective. When you look at Synergy, that is not a piece of hardware, that is truly software defined intelligence built into innovative hardware. Based on our Gen 10 server platform, which in and of itself is the world's most secure industry standard server platform because we have built in silicon route of trust, and things like that, so what you get is all of that put together. All of that integrated. That software defined intelligence, the technology innovation, the infrastructure innovation. And wrappered with the services that both support the layerX company and their customers. >> Maybe talk about your customers a bit more. What are they really pushing you hard to do? What are the big challenges they face, and how are you addressing those? >> One of the most common conversations with cloud is obviously cost. Everyone's trying to commoditize this resource to the Nth degree every day, but the vGrid which is the our brand for our cloud platform, The vGrid position really is around performance and reliability and we back that up through HPE hardware platforms and a software stack that enables that. But our customers are really driving us to make sure that we stay relevant. Not only with that performance and reliability but still on cost. Even though we are giving them enterprise and beyond capabilities as an SMB, cost is still a major defective for an SMB. So for us to keep our overheads low we need automation. You know we're not going to go put in, no disrespect to the product line, but we're not going to go and put in maybe an Apollo or a CloudLine solution, we're going to stick with Synergy and previously the ProLion because of the added value wrapped around that that actually gives us the peace of mind and the operational efficiency through our engineering team to get the work done far more effectively. Now with Synergy takes it up to a whole new level because this is all composable now. My CTO mentioned to me the other day they just put in a new 8450 3PAR. And he said, "All I had to do "was create the CPG's in the 3PAR and OneView did the rest." He's like I don't have to go into all these other steps that he used to have to do. So, it saves time and time is expensive. Not only from a human resource point of view, but go to market speed. >> Well, converged hardware was about having a common set of support technologies. The whole notion of hyperconverge starting to converge the actual administrative tasks. But what I remember, the last time that I was in New Zealand and talked with large users, was a real emphasis on analytics because of New Zealand being an island with great resources in some respects and less resources in others, energy, telecommunications. How is the modern economy of New Zealand with some of the constraints that it faces driving the use of digital technology to lift up industry, services, and the quality of life in New Zealand? >> We're seeing that in a very far reaching kind of industry verticals. And more so now with obviously IOT's become a pretty hot topic, but IOT backed by all the smart and on-demand composable architecture is really making a difference to primary industries, making them more productive more effective, more efficient. But really the customers in New Zealand we're a nation of early adopters. We have 96% of our companies are six or less people. So, we're dealing with SMB's that have to box above their weight. They have to adapt, they have to do more with less. You know all of this cliches that really encumber the average small company, and we have a lot of them. So the demands from an IT perspective are give me what my enterprise counterparts have but at a per user, or resource unit per month kind of model so cloud just makes so much sense for them. >> Susan, big takeaways from Madrid? What do you want the world to walk away with? >> Well I think first of all, when we say we're going to help make hybrid IT simple, what we're talking about and really exemplified with layerX is we're talking about from the edge to the core to the cloud. So, really end to end. The other really exciting thing that we're here talking about is AI, artificial intelligence. Deep learning, machine learning. And you talked about it in the context of edge computing and IOT which is obviously super hot, but we are also bringing AI to the data center. So as we look at-- >> Peter: In other words, making data center operations, IT operations, >> Making the data center autonomous, self healing, self managing. You look at the automobile industry, autonomous cars, right? Well think about how that's going to be applied to autonomous data centers. That's what we're going to be talking about. >> Shoes for the cobbler's children. >> You got it. >> Well, and think about the impact that has on the business where you're allowing people not to spend money on whatever, lung provisioning, >> Right. >> And server management, but really focusing on some other more strategic aspects of their business whether it's digital transformation, AI, other data-oriented activities. >> Exactly. >> Sometimes the data has to be here and you want to make sure that when the data's there it has the same services are available to the business, >> Susan: Yes. >> to take advantage of that asset where it is. >> Real time analytics for the data that matters to our customers at the edge and in the cloud, as well as applying that same AI to the telemetry of the data center and using that to make the data center more efficient, more effective, more autonomous and self-healing. >> Awesome. So, keynotes are coming up very shortly. We'll be running those on our twitch channel twitch.com/siliconangle. You can check those out obviously at HPE as well, HPE.com Susan and Bruce, thanks very much for coming to theCUBE, >> Thank you so much, appreciate it. >> setting up the afternoon. Really appreciate your time. >> No problem. >> Thank you. >> Alright, keep right there buddy. We'll be back after the keynotes. This is theCUBE. We're live from HPE Discover, Madrid. (electronic music)