>>Yeah. >>Hello. Welcome to the Special Cube conversation here in Palo Alto, California. I'm John for a host of the Cube. We have a conversation around a I for the enterprise. What this means I got two great guests. John Finelli, Vice President, virtual GPU at NVIDIA and Maurizio D V D C T o University of Pisa in Italy. Uh, Practitioner, customer partner, um, got VM world coming up. A lot of action happening in the enterprise. John. Great to see you. Nice to meet you. Remotely coming in from Italy for this remote. >>John. Thanks for having us on again. >>Yeah. Nice to meet >>you. I wish we could be in person face to face, but that's coming soon. Hopefully, John, you were talking. We were just talking about before we came on camera about AI for the enterprise. And the last time I saw you in person was in Cuba interview. We were talking about some of the work you guys were doing in AI. It's gotten so much stronger and broader and the execution of an video, the success you're having set the table for us. What is the ai for the enterprise conversation frame? >>Sure. So, um, we, uh we've been working with enterprises today on how they can deliver a I or explore AI or get involved in a I, um uh, in a standard way in the way that they're used to managing and operating their data centre. Um, writing on top of you know, they're Dell servers with B M or V sphere. Um, so that AI feels like a standard workload that night organisation can deliver to their engineers and data scientists. And then the flip side of that, of course, is ensuring that engineers and data scientists get the workloads position to them or have access to them in the way that they need them. So it's no longer a trouble ticket that you have to submit to, I t and you know, count the hours or days or weeks until you you can get new hardware, right By being able to pull it into the mainstream data centre. I can enable self service provisioning for those folks. So we actually we make a I more consumable or easier to manage for I t administrators and then for the engineers and the data scientists, etcetera. We make it easy for them to get access to those resources so they can get to their work right away. >>Quite progress in the past two years. Congratulations on that and looking. It's only the beginning is Day one Mercy. I want to ask you about what's going on as the CTO University piece of what's happening down there. Tell us a little bit about what's going on. You have the centre of excellence there. What does that mean? What does that include? >>Uh, you know, uh, University of Peace. Are you one of one of the biggest and oldest in Italy? Uh, if you have to give you some numbers is around 50 K students and 3000 staff between, uh, professors resurgence and that cabinet receive staff. So I we are looking into data operation of the centres and especially supports for scientific computing. And, uh, this is our our daily work. Let's say this, uh, taking us a lot of times, but, you know, we are able to, uh, reserve a merchant percentage of our time, Uh, for r and D, And this is where the centre of excellence is, Uh, is coming out. Uh, so we are always looking into new kinds of technologies that we can put together to build new solutions to do next generation computing gas. We always say we are looking for the right partners to do things together. And at the end of the day is the work that is good for us is good for our partners and typically, uh, ends in a production system for our university. So is the evolution of the scientific computing environment that we have. >>Yeah. And you guys have a great track record and reputation of, you know, R and D, testing software, hardware combinations and sharing those best practises, you know, with covid impact in the world. Certainly we see it on the supply chain side. Uh, and John, we heard Jensen, your CEO and video talk multiple keynotes. Now about software, uh, and video being a software company. Dell, you mentioned Dale and VM Ware. You know, Covid has brought this virtualisation world back. And now hybrid. Those are words that we used basically in the text industry. Now it's you're hearing hybrid and virtualisation kicked around in real world. So it's ironic that vm ware and El, uh, and the Cube eventually all of us together doing more virtual stuff. So with covid impacting the world, how does that change you guys? Because software is more important. You gotta leverage the hardware you got, Whether it's Dell or in the cloud, this is a huge change. >>Yeah. So, uh, as you mentioned organisations and enterprises, you know, they're looking at things differently now, Um, you know, the idea of hybrid. You know, when you talk to tech folks and we think about hybrid, we always think about you know, how the different technology works. Um, what we're hearing from customers is hybrid, you know, effectively translates into, you know, two days in the office, three days remote, you know, in the future when they actually start going back to the office. So hybrid work is actually driving the need for hybrid I t. Or or the ability to share resources more effectively. Um, And to think about having resources wherever you are, whether you're working from home or you're in the office that day, you need to have access to the same resources. And that's where you know the the ability to virtualize those resources and provide that access makes that hybrid part seamless >>mercy What's your world has really changed. You have students and faculty. You know, Things used to be easy in the old days. Physical in this network. That network now virtual there. You must really be having him having impact. >>Yeah, we have. We have. Of course. As you can imagine, a big impact, Uh, in any kind of the i t offering, uh, from, uh, design new networking technologies, deploying new networking technologies, uh, new kind of operation we find. We found it at them. We were not able anymore to do burr metal operations directly, but, uh, from the i t point of view, uh, we were how can I say prepared in the sense that, uh, we ran from three or four years parallel, uh, environment. We have bare metal and virtual. So as you can imagine, traditional bare metal HPC cluster D g d g X machines, uh, multi GPU s and so on. But in parallel, we have developed, uh, visual environment that at the beginning was, as you can imagine, used, uh, for traditional enterprise application, or VD. I, uh, we have a significant significant arise on a farm with the grid for remote desktop remote pull station that we are using for, for example, uh, developing a virtual classroom or visual go stations. And so this is was typical the typical operation that we did the individual world. But in the same infrastructure, we were able to develop first HPC individual borders of utilisation of the HPC resources for our researchers and, uh, at the end, ai ai offering and ai, uh, software for our for our researchers, you can imagine our vehicle infrastructure as a sort of white board where we are able to design new solution, uh, in a fast way without losing too much performance. And in the case of the AI, we will see that we the performance are almost the same at the bare metal. But with all the flexibility that we needed in the covid 19 world and in the future world, too. >>So a couple things that I want to get John's thoughts as well performance you mentioned you mentioned hybrid virtual. How does VM Ware and NVIDIA fit into all this as you put this together, okay, because you bring up performance. That's now table stakes. He's leading scale and performance are really on the table. everyone's looking at it. How does VM ware an NVIDIA John fit in with the university's work? >>Sure. So, um, I think you're right when it comes to, uh, you know, enterprises or mainstream enterprises beginning their initial foray into into a I, um there are, of course, as performance in scale and also kind of ease of use and familiarity are all kind of things that come into play in terms of when an enterprise starts to think about it. And, um, we have a history with VM Ware working on this technology. So in 2019, we introduced our virtual compute server with VM Ware, which allowed us to effectively virtual is the Cuda Compute driver at last year's VM World in 2020 the CEOs of both companies got together and made an announcement that we were going to bring a I R entire video AI platform to the Enterprise on top of the sphere. And we did that, Um, starting in March this year, we we we finalise that with the introduction of GM wears V, Sphere seven, update two and the early access at the time of NVIDIA ai Enterprise. And, um, we have now gone to production with both of those products. And so customers, Um, like the University of Pisa are now using our production capabilities. And, um, whenever you virtualize in particular and in something like a I where performances is really important. Um, the first question that comes up is, uh doesn't work and And how quickly does it work Or or, you know, from an I t audience? A lot of times you get the How much did it slow down? And and and so we We've worked really closely from an NVIDIA software perspective and a bm wear perspective. And we really talk about in media enterprise with these fair seven as optimist, certified and supported. And the net of that is, we've been able to run the standard industry benchmarks for single node as well as multi note performance, with about maybe potentially a 2% degradation in performance, depending on the workload. Of course, it's very different, but but effectively being able to trade that performance for the accessibility, the ease of use, um, and even using things like we realise, automation for self service for the data scientists, Um and so that's kind of how we've been pulling it together for the market. >>Great stuff. Well, I got to ask you. I mean, people have that reaction of about the performance. I think you're being polite. Um, around how you said that shows the expectation. It's kind of sceptical, uh, and so I got to ask you, the impact of this is pretty significant. What is it now that customers can do that? They couldn't or couldn't feel they had before? Because if the expectations as well as it worked well, I mean, there's a fast means. It works, but like performance is always concerned. What's different now? What what's the bottom line impact on what country do now that they couldn't do before. >>So the bottom line impact is that AI is now accessible for the enterprise across there. Called their mainstream data centre, enterprises typically use consistent building blocks like the Dell VX rail products, right where they have to use servers that are common standard across the data centre. And now, with NVIDIA Enterprise and B M R V sphere, they're able to manage their AI in the same way that they're used to managing their data centre today. So there's no retraining. There's no separate clusters. There isn't like a shadow I t. So this really allows an enterprise to efficiently deploy um, and cost effectively Deploy it, uh, it without because there's no performance degradation without compromising what their their their data scientists and researchers are looking for. And then the flip side is for the data science and researcher, um, using some of the self service automation that I spoke about earlier, they're able to get a virtual machine today that maybe as a half a GPU as their models grow, they do more exploring. They might get a full GPU or or to GPS in a virtual machine. And their environment doesn't change because it's all connected to the back end storage. And so for the for the developer and the researcher, um, it makes it seamless. So it's really kind of a win for both Nike and for the user. And again, University of Pisa is doing some amazing things in terms of the workloads that they're doing, Um, and, uh and, uh, and are validating that performance. >>Weigh in on this. Share your opinion on or your reaction to that, What you can do now that you couldn't do before. Could you share your experience? >>Our experience is, uh, of course, if you if you go to your, uh, data scientists or researchers, the idea of, uh, sacrificing four months to flexibility at the beginning is not so well accepted. It's okay for, uh, for the Eid management, As John was saying, you have people that is know how to deal with the virtual infrastructure, so nothing changed for them. But at the end of the day, we were able to, uh, uh, test with our data. Scientists are researchers veteran The performance of us almost similar around really 95% of the performance for the internal developer developer to our work clothes. So we are not dealing with benchmarks. We have some, uh, work clothes that are internally developed and apply to healthcare music generator or some other strange project that we have inside and were able to show that the performance on the beautiful and their metal world were almost the same. We, the addition that individual world, you are much more flexible. You are able to reconfigure every finger very fast. You are able to design solution for your researcher, uh, in a more flexible way. An effective way we are. We were able to use the latest technologies from Dell Technologies and Vidia. You can imagine from the latest power edge the latest cuts from NVIDIA. The latest network cards from NVIDIA, like the blue Field to the latest, uh, switches to set up an infrastructure that at the end of the day is our winning platform for our that aside, >>a great collaboration. Congratulations. Exciting. Um, get the latest and greatest and and get the new benchmarks out their new playbooks. New best practises. I do have to ask you marriage, if you don't mind me asking why Look at virtualizing ai workloads. What's the motivation? Why did you look at virtualizing ai work clothes? >>Oh, for the sake of flexibility Because, you know, uh, in the latest couple of years, the ai resources are never enough. So we are. If you go after the bare metal, uh, installation, you are going into, uh, a world that is developing very fastly. But of course, you can afford all the bare metal, uh, infrastructure that your data scientists are asking for. So, uh, we decided to integrate our view. Dual infrastructure with AI, uh, resources in order to be able to, uh, use in different ways in a more flexible way. Of course. Uh, we have a We have a two parallels world. We still have a bare metal infrastructure. We are growing the bare metal infrastructure. But at the same time, we are growing our vehicle infrastructure because it's flexible, because we because our our stuff, people are happy about how the platform behaviour and they know how to deal them so they don't have to learn anything new. So it's a sort of comfort zone for everybody. >>I mean, no one ever got hurt virtualizing things that makes it makes things go better faster building on on that workloads. John, I gotta ask you, you're on the end video side. You You see this real up close than video? Why do people look at virtualizing ai workloads is the unification benefit. I mean, ai implies a lot of things, implies you have access to data. It implies that silos don't exist. I mean, that doesn't mean that's hard. I mean, is this real people actually looking at this? How is it working? >>Yeah. So? So again, um you know for all the benefits and activity today AI brings a I can be pretty complex, right? It's complex software to set up and to manage. And, um, within the day I enterprise, we're really focusing in on ensuring that it's easier for organisations to use. For example Um, you know, I mentioned you know, we we had introduced a virtual compute server bcs, um uh, two years ago and and that that has seen some some really interesting adoption. Some, uh, enterprise use cases. But what we found is that at the driver level, um, it still wasn't accessible for the majority of enterprises. And so what we've done is we've built upon that with NVIDIA Enterprise and we're bringing in pre built containers that remove some of the complexities. You know, AI has a lot of open source components and trying to ensure that all the open source dependencies are resolved so you can get the AI developers and researchers and data scientists. Actually doing their work can be complex. And so what we've done is we've brought these pre built containers that allow you to do everything from your initial data preparation data science, using things like video rapids, um, to do your training, using pytorch and tensorflow to optimise those models using tensor rt and then to deploy them using what we call in video Triton Server Inference in server. Really helping that ai loop become accessible, that ai workflow as something that an enterprise can manage as part of their common core infrastructure >>having the performance and the tools available? It's just a huge godsend people love. That only makes them more productive and again scales of existing stuff. Okay, great stuff. Great insight. I have to ask, What's next one's collaboration? This is one of those better together situations. It's working. Um, Mauricio, what's next for your collaboration with Dell VM Ware and video? >>We will not be for sure. We will not stop here. Uh, we are just starting working on new things, looking for new development, uh, looking for the next beast. Come, uh, you know, the digital world is something that is moving very fast. Uh, and we are We will not We will not stop here because because they, um the outcome of this work has been a very big for for our research group. And what John was saying This the fact that all the software stock for AI are simplified is something that has been, uh, accepted. Very well, of course you can imagine researching is developing new things. But for people that needs, uh, integrated workflow. The work that NVIDIA has done in the development of software package in developing containers, that gives the end user, uh, the capabilities of running their workloads is really something that some years ago it was unbelievable. Now, everything is really is really easy to manage. >>John mentioned open source, obviously a big part of this. What are you going to? Quick, Quick follow if you don't mind. Are you going to share your results so people can can look at this so they can have an easier path to AI? >>Oh, yes, of course. All the all the work, The work that is done at an ideal level from University of Visa is here to be shared. So we we as, uh, as much as we have time to write down we are. We are trying to find a way to share the results of the work that we're doing with our partner, Dell and NVIDIA. So for sure will be shared >>well, except we'll get that link in the comments, John, your thoughts. Final thoughts on the on the on the collaboration, uh, with the University of Pisa and Delvian, where in the video is is all go next? >>Sure. So So with University of Pisa, We're you know, we're absolutely, uh, you know, grateful to Morocco and his team for the work they're doing and the feedback they're sharing with us. Um, we're learning a lot from them in terms of things we can do better and things that we can add to the product. So that's a fantastic collaboration. Um, I believe that Mauricio has a session at the M World. So if you want to actually learn about some of the workloads, um, you know, they're doing, like, music generation. They're doing, you know, covid 19 research. They're doing deep, multi level, uh, deep learning training. So there's some really interesting work there, and so we want to continue that partnership. University of Pisa, um, again, across all four of us, uh, university, NVIDIA, Dell and VM Ware. And then on the tech side, you know, for our enterprise customers, um, you know, one of the things that we actually didn't speak much about was, um I mentioned that the product is optimised certified and supported, and I think that support cannot be understated. Right? So as enterprises start to move into these new areas, they want to know that they can pick up the phone and call in video or VM ware. Adele, and they're going to get support for these new workloads as they're running them. Um, we were also continuing, uh, you know, to to think about we spent a lot of time today on, like, the developer side of things and developing ai. But the flip side of that, of course, is that when those ai apps are available or ai enhanced apps, right, Pretty much every enterprise app today is adding a I capabilities all of our partners in the enterprise software space and so you can think of a beady eye enterprises having a runtime component so that as you deploy your applications into the data centre, they're going to be automatically take advantage of the GPS that you have there. And so we're seeing this, uh, future as you're talking about the collaboration going forward, where the standard data centre building block still maintains and is going to be something like a VX rail two U server. But instead of just being CPU storage and RAM, they're all going to go with CPU, GPU, storage and RAM. And that's going to be the norm. And every enterprise application is going to be infused with AI and be able to take advantage of GPS in that scenario. >>Great stuff, ai for the enterprise. This is a great QB conversation. Just the beginning. We'll be having more of these virtualizing ai workloads is real impacts data scientists impacts that compute the edge, all aspects of the new environment we're all living in. John. Great to see you, Maurizio here to meet you and all the way in Italy looking for the meeting in person and good luck in your session. I just got a note here on the session. It's at VM World. Uh, it's session 22 63 I believe, um And so if anyone's watching, Want to check that out? Um, love to hear more. Thanks for coming on. Appreciate it. >>Thanks for having us. Thanks to >>its acute conversation. I'm John for your host. Thanks for watching. We'll talk to you soon. Yeah,

>>Hi, Lisa Martin here with the cube. You're watching our coverage of Dell technologies world. The digital virtual experience. I've got two guests with me here today. We're going to be talking about the university of Piza and how it is leaning into all flash data lakes powered by Dell technologies. One of our alumni is back MERITO, Debbie, and the CTO of the university of PISA. Maricio welcome back to the cube. Thank you. Very excited to talk to you today. CAUTI Gosha is here as well. The director of product management at Dell technologies. Kaushik. Welcome to the cube. Thank you. So here we are at this virtual event again, Maricio you were last on the cube at VMworld a few months ago, the virtual experience as well, but talk to her audience a little bit before we dig into the technology and some of these demanding workloads that the university is utilizing. Talk to me a little bit about your role as CTO and about the university. >>So my role as CTO at university of PISA is, uh, uh, regarding the, uh, data center operations and, uh, scientific computing support for these, the main, uh, occupation that, uh, that, uh, yeah. Then they support the world, saw the technological choices that university of PISA is, uh, is doing, uh, during the latest, uh, two or three years. >>Talk to me about some, so this is a, in terms of students we're talking about 50,000 or so students 3000 faculty and the campus is distributed around the town of PISA, is that correct? Maricio >>Uh, the university of PISA is sort of a, uh, town campus in the sense that we have 20 departments that are, uh, located inside the immediate eval town, uh, but due to the choices, but university of peace, I S uh, the, uh, last, uh, uh, nineties, uh, we are, uh, owner of, uh, of a private fiber network connecting all our, uh, departments and allow the templates. And so we can use the town as a sort of white board to design, uh, uh, new services, a new kind of support for teaching. Uh, and, uh, and so, >>So you've really modernized the data infrastructure for the university that was founded in the middle ages. Talk to me now about some of the workloads and that are generating massive amounts of data, and then we'll get into what you're doing with Dell technologies. >>Oh, so the university of PISA as a, uh, quite old on HPC, traditional HPC. So we S we are supporting, uh, uh, the traditional workloads from, uh, um, CAE or engineering or chemistry or oil and gas simulations. Uh, of course it during, uh, uh, the pandemic year, last year, especially, uh, we have new, uh, kind of work you'll scan, uh, summer related, uh, to the, uh, fast movement of the HPC workload from let's say, traditional HPC to AI and machine learning. And those are the, um, request that you support a lot of remote activities coming from, uh, uh, uh, distance learning, uh, to remote ties, uh, uh, laboratories or stations or whatever, most elder in presence in the past. And so the impact either on the infrastructure or, and the specialty and the storage part was a significant. >>So you talked about utilizing the high performance computing environments for awhile and for scientific computing and things. I saw a case study that you guys have done with Dell, but then during the pandemic, the challenge and the use case of remote learning brought additional challenges to your environment from that perspective, how, how were you able to transfer your curriculum to online and enable the scientists, the physicists that oil and gas folks doing research to still access that data at the speed that they needed to, >>Uh, you know, for what you got, uh, uh, uh, distance learning? Of course. So we were, uh, based on the cloud services were not provided internally by Yas. So we lie, we based on Microsoft services, so Google services and so on, but what regards, uh, internal support, uh, scientific computing was completely, uh, remote dies either on support or experience, uh, because, uh, I can, uh, I, can I, uh, bring some, uh, some examples, uh, for example, um, laboratory activities, uh, we are, the access to the laboratories, uh, was the of them, uh, as much as possible. Uh, we design a special networker to connect all the and to give the researcher the possibility of accessing the data on visit special network. So as sort of a collector of data, uh, inside our, our university network, uh, you can imagine that the, uh, for example, was, was a key factor for us because utilization was, uh, uh, for us, uh, and flexible way to deliver new services, uh, in an easy way, uh, especially if you have to, uh, have systems for remote. So, as, as I told you before about the, uh, network, as well as a white board, but also the computer infrastructure, it was VM-ware visualization and treated as a, as a sort of what we were designing with services either, either for interactive services or especially for, uh, scientific computing. For example, we have an experience with it and a good polarization of HPC workload. We start agents >>Talk to me about the storage impact, because as we know, we talk about, you know, these very demanding, unstructured workloads, AI machine learning, and that can be, those are difficult for most storage systems to handle the radio. Talk to us about why you leaned into all flash with Dell technologies and talk to us a little bit about the technologies that you've implemented. >>So, uh, if I, if I have to think about our, our storage infrastructure before the pandemic, I have to think about Iceland because our HPC workloads Moss, uh, mainly based off, uh, Isilon, uh, as a storage infrastructure, uh, together with some, uh, final defense system, as you can imagine, we were deploying in-house, uh, duty independently, especially with the explosion of the AI, with them, uh, blueprint of the storage requests change the law because of what we have, uh, uh, deal dens. And in our case, it was an, I breathed the Isilon solution didn't fit so well for HB for AI. And this is why we, uh, start with the data migration. That was, it was not really migration, but the sort of integration of the power scaler or flash machine inside our, uh, environment, because then the power scale, all flesh and especially, uh, IO in the future, uh, the MVME support, uh, is a key factor for the storage. It just support, uh, we already have experience as some of the, uh, NBME, uh, possibilities, uh, on the power PowerMax so that we have here, uh, that we use part for VDI support, uh, but off, um, or fleshly is the minimum land and EME, uh, is what we need to. >>Gotcha. Talk to me about what Dell technologies has seen the uptick in the demand for this, uh, as Maricio said, they were using Isilon before adding in power scale. What are some of the changing demands that, that Dell technologies has seen and how does technologies like how our scale and the F 900 facilitate these organizations being able to rapidly change their environment so that they can utilize and extract the value from data? >>Yeah, no, absolutely. What occupational intelligence is an area that, uh, continues to amaze me. And, uh, personally I think the, the potential here is immense. Um, uh, as Maurizio said, right, um, the, the data sets, uh, with artificial intelligence, I have, uh, grown significantly and, and not only the data has become, um, uh, become larger the models, the AI models that, that we, that are used have become more complex. Uh, for example, uh, one of the studies suggests that, uh, the, uh, that for a modeling of, uh, natural language processing, um, uh, one of the fields in AI, uh, the number of parameters used, could exceed like about a trillion in, uh, in a few years, right? So almost a size of a human brain. So, so not only that means that there's a lot of fear mounted to be, uh, data, to be processed, but, uh, by, uh, the process stored in yesterday, uh, but probably has to be done in the same amount of Dinah's before, perhaps even a smaller amount of time, right? So a larger data theme time, or perhaps even a smaller amount of time. So, absolutely. I agree. I mean, those type of, for these types of workloads, you need a storage that gives you that high-performance access, but also being able to store the store, that data is economically. >>And how does Dell technologies deliver that? The ability to scale the economics what's unique and differentiated about power skill? >>Uh, so power scale is, is, is our all flash, uh, system it's, uh, it's, uh, it's bad users, dark techno does some of the same capabilities that, uh, Isilon, um, products use used to offer, uh, one of his fault system capabilities, some of the capabilities that Maurizio has used and loved in the past, some of those, some of those same capabilities are brought forward. Now on this spar scale platform, um, there are some changes, like for example, on new Parscale's platform supports Nvidia GPU direct, right? So for, uh, artificial intelligence, uh, workloads, you do need these GPU capable machines. And, uh, and, uh, Parscale supports that those, uh, high high-performance Jupiter rec machines, uh, through, through the different technologies that we offer. And, um, the Parscale F 900, which should, which we are going to launch very soon, um, um, is, is, is our best hype, highest performance all-flash and the most economic allowed slash uh, to date. So, um, so it is, um, it not only is our fastest, but also offers, uh, the most economic, uh, most economical way of storing the data. Um, so, so ideal far for these type of high-performance workloads, like AIML, deep learning and so on. Excellent. >>So talk to me about some of the results that the university is achieving so far. I did read a three X improvement in IO performance. You were able to get nearly a hundred percent of the curriculum online pretty quickly, but talk to me about some of the other impacts that Dell technologies has helping the university to achieve. >>Oh, we had, uh, we had an old, uh, in all the Dell customer, and if you, uh, give a Luca walk, we have that inside the insomnia, our data centers. Uh, we typically joking, we define them as a sort of, uh, Dell technologies supermarket in the sense that, uh, uh, degreed part of our, our servers storage environment comes from, uh, from that technology said several generations of, uh, uh, PowerEdge servers, uh, uh, power, my ex, uh, Isaac along, uh, powers, Gale power store. So we, uh, we are, uh, um, using a lot of, uh, uh, Dell technologies here, here, and of course, uh, um, in the past, uh, our traditional, uh, workloads were well supported by that technologies. And, uh, Dell technologies is, uh, uh, driving ourselves versus, uh, the, what we call the next generation workloads, uh, because we are, uh, uh, combining gas, uh, in, um, in the transition of, uh, um, uh, the next generation of computing there, but to be OPA who, uh, to ask here, and he was walked through our research of looking for, cause if I, if I have to, to, to, to give a look to what we are, uh, doing, uh, mostly here, healthcare workloads, uh, deep learning, uh, uh, data analysis, uh, uh, image analysis in C major extraction that everything have be supported, especially from, uh, the next next generation servers typically keep the, uh, with, with GPU's. >>This is why GPU activities is, is so important for answer, but also, uh, supported on the, on the, on the networking side. But because of that, the, the, the speed, the, and the, of the storage, and must be tired to the next generation networking. Uh, low-latency high-performance because at the end of the day, you have to, uh, to bring the data in storage and DP. Can you do it? Uh, so, uh, they're, uh, one of the low latency, uh, uh, I performance, if they're connected zones is also a side effect of these new work. And of course that the college is, is, is. >>I love how you described your data centers as a Dell technologies supermarket, maybe a different way of talking about a center of excellence question. I want to ask you about, I know that the university of PISA is SCOE for Dell. Talk to me about in the last couple of minutes we have here, what that entails and how Dell helps customers become a center of excellence. >>Yeah, so Dell, um, like talked about has a lot of the Dell Dell products, uh, today, and, and, and in fact, he mentioned about the pirate servers, the power scale F 900 is, is actually based on a forehead server. So, so you can see, so a lot of these technologies are sort of in the linked with each other, they talk to each other, they will work together. Um, and, and, and that sort of helps, helps customers manage the entire, uh, ecosystem lifecycle data, life cycle together, versus as piece parts, because we have solutions that solve all aspects of, of, of the, uh, of, of, uh, of our customer like Mauricio's needs. Right. So, um, so yeah, I'm glad Maurizio is, is leveraging Dell and, um, and I'm happy we are able to help help more issue or solve solve, because, uh, all his use cases, uh, and UN >>Excellent. Maricio last question. Are you going to be using AI machine learning, powered by Dell to determine if the tower of PISA is going to continue to lean, or if it's going to stay where it is? >>Uh, the, the, the leaning tower is, uh, an engineering miracle. Uh, some years ago, uh, an engineering, uh, incredible worker, uh, was able, uh, uh, to fix them. They leaning for a while and let's open up the tower visa, stay there because he will be one of our, uh, beauty that you can come to to visit. >>And that's one part of Italy I haven't been to. So as pandemic, I gotta add that to my travel plans, MERITO and Kaushik. It's been a pleasure talking to you about how Dell is partnering with the university of PISA to really help you power AI machine learning workloads, to facilitate many use cases. We are looking forward to hearing what's next. Thanks for joining me this morning. Thank you for my guests. I'm Lisa Martin. You're watching the cubes coverage of Dell technologies world. The digital event experience.

(upbeat music) >> Hi, Lisa Martin here with theCUBE. You're watching our coverage of Dell Technologies World, the Digital Virtual Experience. I've got two guests with me here today. We're going to be talking about the University of Pisa and how it is leaning into all flash deal that is powered by Dell Technologies. One of our alumni is back, Maurizio Davini, the CTO of the University of Pisa. Maurizio, welcome back to theCUBE. >> Thank you. You're always welcome. >> Very excited to talk to you today. Kaushik Ghosh is here as well, The Director of Product Management at Dell Technologies. Kaushik, welcome to theCUBE. >> Thank you. >> So here we are at this virtual event again. Maurizio, you were last on theCUBE at VM world a few months ago, the virtual experience as well. But talk to our audience a little bit, before we dig into the technology and some of these demanding workloads that the University is utilizing, talk to me a little bit about your role as CTO and about the University. >> So my role as CTO at University of Pisa is regarding the data center operations and scientific computing support. It is the main occupation that I have. Then I support also, the technological choices That the University of Pisa is doing during the latest two or three years. >> Talk to me about something, so this is in terms of students, we're talking about 50,000 or so students, 3000 faculty and the campus is distributed around the town of Pisa. Is that correct, Maurizio? >> The University of Pisa is sort of a town campus in the sense that we have 20 departments that are located inside the medieval town, but due to the choices that University of Pisa has done in the last '90s, we are owner of a private fiber network connecting all our departments and all our (indistinct). And so we can use the town as a sort of white board to design new services, new kind of support for teaching and so on. >> So you've really modernized the data infrastructure for the University that was founded in the middle ages. Talk to me now about some of the workloads, Maurizio, that are generating massive amounts of data and then we'll get into what you're doing with Dell Technologies. >> Oh, so the University of Pisa has a quite old historian HPC, traditional HPC. So we are supporting the traditional workloads from CAE or engineering or chemistry or oil and gas simulations. Of course, during the pandemic year, last year especially, we have new kind of workload scan, some related to the fast movement of the HPC workload from let's say, traditional HPC to AI and machine learning. And also, they request to support a lot of remote activities coming from distance learning to remotize laboratories or stations or whatever, most elder in presence in the past. And so the impact either on the infrastructure or, and especially on the storage part, was significant. >> So you talked about utilizing the high performance computing environments for a while and for scientific computing and things, I saw a case study that you guys have done with Dell, but then during the pandemic, the challenge and the use case of remote learning brought additional challenges to your environment. From that perspective, how were you able to transfer your curriculum to online and enable the scientists, the physicists, the oil and gas folks doing research to still access that data at the speed that they needed to? >> You know, for what you got distance learning, of course, we were based on cloud services that were not provided internally by us. So we based on Microsoft services, on Google services and so on. But what regards internal support, scientific computing was completely remotized, either on support or experience, because how can I bring some examples? For example, laboratory activities were remotized. The access to the laboratories was (indistinct) remote as much as possible. We designed a special network to connect all the laboratories and to give the researcher the possibility of accessing the data on this special network. So a sort of a collector of data inside our university network. You can imagine that... Utilization, for example, was a key factor for us because utilization was, for us, a flexible way to deliver new services in an easy way, especially, if you have to administer systems for remote. So as I told you before about the network as a white board, also, the computer infrastructure was (indistinct) utilization treated as a sort of (indistinct). We were designing new services, either for interactive services, or especially for scientific computing. For example, we have an experience with utilization of HPC workload, storage and so on. >> Talk to me about the storage impact because as we know, we talk about these very demanding unstructured workloads, AI, machine learning, and those are difficult for most storage systems to handle. Maurizio, talk to us about why you leaned into all flash with Dell Technologies and talk to us a little bit about the technologies that you've implemented. >> So if I have to think about our storage infrastructure before the pandemic, I have to think about Isilon, because our HPC workloads was mainly based off Isilon as a storage infrastructure. Together, with some final defense system, as you can imagine, we were deploying in our homes. During the pandemic, but especially with the explosion of the AI, the blueprint of the storage requests changed a lot because what we had until then, and in our case, was an hybrid Isilon solution. Didn't fit so well for HB, for AI (indistinct) and this is why we started the migration. It was not really migration, but the sort of integration of the Power Scale or flash machine inside our environment, because then the Power Scale or flash, and especially, I hope in the future, the MVME support is a key factor for the storage, storage support. We already have experienced some of the MVME possibilities on the Power Max that we have here that we use (indistinct) and part for VDI support, but flash is the minimum and MVME is what we need to support in the right way the AI workloads. >> Lisa: Kaushik, talk to me about what Dell Technologies has seen. The optic the demand for this. As Maurizio said, they were using Isilon before, adding in Power Scale. What are some of the changing demands that Dell technologies has seen and how does technologies like Power Scale and the F900 facilitate these organizations being able to rapidly change their environment so that they can utilize and extract the value from data? >> Yeah, no, absolutely. Artificial intelligence is an area that continues to amaze me and personally, I think the potential here is immense. As Maurizio said, right? The data sets with artificial intelligence have grown significantly, and not only the data has become larger, the models, the AI models that are used have become more complex. For example, one of the studies suggests that for a modeling of natural language processing, one of the fields in AI, the number of parameters used could exceed like a trillion in a few years, right? So almost the size of a human brain. So not only that means that there's a lot of data to be processed, but the process stored ingested, but probably has to be done in the same amount of time as before or perhaps even a smaller amount of time, right? So larger data, same time, or perhaps even a smaller amount of time. So, absolutely, I agree. For these types of workloads, you need a storage that gives you that high-performance access, but also being able to store that data economically. >> Lisa: And Kaushik, how does Dell technologies deliver that? The ability to scale the economics. What's unique and differentiated about Power Scale? >> So Power Scale is our all flash system. It uses some of the same capabilities that Isilon products used to offer. The 1 FS file system capabilities. Some of the same capabilities that (indistinct) has used and loved in the past. So some of those same capabilities are brought forward now. on this Power Scale platform. There are some changes, like for example, our new Power Scale platform supports NVDR GPU direct, right? So for artificial intelligence workloads, you do need these GPU capable machines and Power Scale supports those high-performance GPU direct machines through the different technologies that we offer, and the Power Scale F 900, which we are going to launch very soon is our best highest performance all flash and the most economical all flash to date. So it not only is our fastest, but also offers the most economical way of storing the data. So ideal for these type of high-performance workloads, like AIML, deep learning and so on. >> Excellent. Maurizio, talk to me about some of the results that the University is achieving so far. I did read a three X improvement in IO performance. You were able to get nearly a hundred percent of the curriculum online pretty quickly, but talk to me about some of the other impacts that Dell technologies is helping the University to achieve. >> Oh, we are an old Dell customer and if you give a look what we have inside our data centers, we typically joking. We define as a sort of Dell technologies supermarket in the sense that the great part of our servers storage environment comes from Dell technology. Several generations of Power Edge servers, Power Max, Isilon, Power Scale, Power Sore. So we are using a lot of Dell technologies here, and of course, in the past, our traditional workloads were well supported by Dell technologies. And Dell technologies is driving us versus what we call the next generation workloads, because they are accompanying us in the transition versus the next generation computing, but to hope to adhere and (indistinct) to our researchers are looking for, because if I had to give a look to what we are doing mostly here, healthcare workloads, deep learning, data analysis, image analysis, same major extraction. Everything have to be supported, especially from the next generation servers, typically to keep with GPUs. This is why GPU direct is so important for us, but also, supported on the networking side, because the speed of the storage must be tied to the next generation networking. Low latency, high performance, because at the end of the day, you have to bring the data to the storage room, and typically, you do it by importing it. So they're one of the low latency, high performance interconnections. Zones is also a side effect of this new (indistinct). And of course, Dell Technologies is with us in this transition. >> I loved how you described your data centers as a Dell Technologies supermarket. Maybe a different way of talking about a center of excellence. Kaushik, I want to ask you about... I know that the University of Pisa is a SCOE for Dell. Talk to me about, in the last couple of minutes we have here, what that entails and how Dell helps customers become a center of excellence. >> Yeah. So Dell, like Maurizio has talked about, has a lot of the Dell products today. And in fact, he mentioned about the powered servers, the Power Scale F 900 is actually based on a powered server. So you can see. So a lot of these technologies are sort of interlinked with each other. They talk to each other, they work together and that sort of helps customers manage their entire ecosystem life cycle, data life cycle together versus as piece spots, because we have solutions that solve all aspects of our customer, like Maurizio's needs, right? So, yeah, I'm glad Maurizio is leveraging Dell and I'm happy we are able to help Maurizio solve all his use cases and when. >> Lisa: Excellent. Maurizio, last question, are you going to be using AI machine learning powered by Dell to determine if the tower of Pisa is going to continue to lean or if it's going to stay where it is? >> The leaning tower is an engineering miracle. Some years ago, an incredible engineering worker was able to fix the leaning for a while, and let's hope that the tower of Pisa stay there because it's one of our beauty that you can come to visit. >> And that's one part of Italy I haven't been to. So post pandemic, I got to add that to my travel plans. Maurizio and Kaushik, it's been a pleasure talking to you about how Dell is partnering with the University of Pisa to really help you power AI machine learning workloads to facilitate many use cases. We are looking forward to hearing what's next. Thanks for joining me this morning. >> Kaushik: Thank you. >> Maurizio: Thank you. For my guests, I'm Lisa Martin. You're watching theCUBE's coverage of Dell technologies world, the digital event experience. (upbeat music)

>> From around the globe, it's theCUBE, with digital coverage of VMworld 2020, brought to you by the VMworld and its ecosystem partners. >> I'm Stu Miniman, and welcome back to theCUBES coverage of VMworld 2020, our 11th year doing this show, of course, the global virtual event. And what do we love talking about on theCUBE? We love talking to customers. It is a user conference, of course, so really happy to welcome to the program. From the University of Pisa, the Chief Technology Officer Maurizio Davini and joining him is Thierry Pellegrini, one of our theCUBE alumni. He's the vice president of worldwide, I'm sorry, Workload Solutions and HPC with Dell Technologies. Thierry, thank you so much for joining us. >> Thanks too. >> Thanks to you. >> Alright, so let, let's start. The University of Pisa, obviously, you know, everyone knows Pisa, one of the, you know, famous city iconic out there. I know, you know, we all know things in Europe are a little bit longer when you talk about, you know, some of the venerable institutions here in the United States, yeah. It's a, you know, it's a couple of hundred years, you know, how they're using technology and everything. I have to imagine the University of Pisa has a long storied history. So just, if you could start before we dig into all the tech, give us our audience a little bit, you know, if they were looking up on Wikipedia, what's the history of the university? >> So University of Pisa is one of the oldest in the world because there has been founded in 1343 by a pope. We were authorized to do a university teaching by a pope during the latest Middle Ages. So it's really one of the, is not the oldest of course, but the one of the oldest in the world. It has a long history, but as never stopped innovating. So anything in Pisa has always been good for innovating. So either for the teaching or now for the technology applied to a remote teaching or a calculation or scientific computing, So never stop innovating, never try to leverage new technologies and new kind of approach to science and teaching. >> You know, one of your historical teachers Galileo, you know, taught at the university. So, you know, phenomenal history help us understand, you know, you're the CTO there. What does that encompass? How, you know, how many students, you know, are there certain areas of research that are done today before we kind of get into the, you know, the specific use case today? >> So consider that the University of Pisa is a campus in the sense that the university faculties are spread all over the town. Medieval like Pisa poses a lot of problems from the infrastructural point of view. So, we have bought a lot in the past to try to adapt the Medieval town to the latest technologies advancement. Now, we have 50,000 students and consider that Pisa is a general partners university. So, we cover science, like we cover letters in engineering, medicine, and so on. So, during the, the latest 20 years, the university has done a lot of effort to build an infrastructure that was able to develop and deploy the latest technologies for the students. So for example, we have a private fiber network covering all the town, 65 kilometers of a dark fiber that belongs to the university, four data centers, one big and three little center connected today at 200 gigabit ethernet. We have a big data center, big for an Italian University, of course, and not Poland and U.S. university, where is, but also hold infrastructure for the enterprise services and the scientific computing. >> Yep, Maurizio, it's great that you've had that technology foundation. I have to imagine the global pandemic COVID-19 had an impact. What's it been? You know, how's the university dealing with things like work from home and then, you know, Thierry would love your commentary too. >> You know, we, of course we were not ready. So we were eaten by the pandemic and we have to adapt our service software to transform from imperson to remote services. So we did a lot of work, but we are able, thanks to the technology that we have chosen to serve almost a 100% of our curriculum studies program. We did a lot of work in the past to move to virtualization, to enable our users to work for remote, either for a workstation or DC or remote laboratories or remote calculation. So virtualization has designed in the past our services. And of course when we were eaten by the pandemic, we were almost ready to transform our service from in person to remote. >> Yeah, I think it's, it's true, like Maurizio said, nobody really was preparing for this pandemic. And even for, for Dell Technologies, it was an interesting transition. And as you can probably realize a lot of the way that we connect with customers is in person. And we've had to transition over to modes or digitally connecting with customers. We've also spent a lot of our energy trying to help the community HPC and AI community fight the COVID pandemic. We've made some of our own clusters that we use in our HPC and AI innovation center here in Austin available to genomic research or other companies that are fighting the the virus. And it's been an interesting transition. I can't believe that it's already been over six months now, but we've found a new normal. >> Detailed, let's get in specifically to how you're partnering with Dell. You've got a strong background in the HPC space, working with supercomputers. What is it that you're turning to Dell in their ecosystem to help the university with? >> So we are, we have a long history in HPC. Of course, like you can imagine not to the biggest HPC like is done in the U.S. so in the biggest supercomputer center in Europe. We have several system for doing HPC. Traditionally, HPC that are based on a Dell Technologies offer. We typically host all kind of technology's best, but now it's available, of course not in a big scale but in a small, medium scale that we are offering to our researcher, student. We have a strong relationship with Dell Technologies developing together solution to leverage the latest technologies, to the scientific computing, and this has a lot during the research that has been done during this pandemic. >> Yeah, and it's true. I mean, Maurizio is humble, but every time we have new technologies that are to be evaluated, of course we spend time evaluating in our labs, but we make it a point to share that technology with Maurizio and the team at the University of Pisa, That's how we find some of the better usage models for customers, help tuning some configurations, whether it's on the processor side, the GPU side, the storage and the interconnect. And then the topic of today, of course, with our partners at VMware, we've had some really great advancements Maurizio and the team are what we call a center of excellence. We have a few of them across the world where we have a unique relationship sharing technology and collaborating on advancement. And recently Maurizio and the team have even become one of the VMware certified centers. So it's a great marriage for this new world where virtual is becoming the norm. >> But well, Thierry, you and I had a conversation to talk earlier in the year when VMware was really geering their full kind of GPU suite and, you know, big topic in the keynote, you know, Jensen, the CEO of Nvidia was up on stage. VMware was talking a lot about AI solutions and how this is going to help. So help us bring us in you work with a lot of the customers theory. What is it that this enables for them and how to, you know, Dell and VMware bring, bring those solutions to bear? >> Yes, absolutely. It's one statistic I'll start with. Can you believe that only on average, 15 to 20% of GPU are fully utilized? So, when you think about the amount of technology that's are at our fingertips and especially in a world today where we need that technology to advance research and scientistic discoveries. Wouldn't it be fantastic to utilize those GPU's to the best of our ability? And it's not just GPU's , I think the industry has in the IT world, leverage virtualization to get to the maximum recycles for CPU's and storage and networking. Now you're bringing the GPU in the fold and you have a perfect utilization and also flexibility across all those resources. So what we've seen is that convergence between the IT world that was highly virtualized, and then this highly optimized world of HPC and AI because of the resources out there and researchers, but also data scientists and company want to be able to run their day to day activities on that infrastructure. But then when they have a big surge need for research or a data science use that same environment and then seamlessly move things around workload wise. >> Yeah, okay I do believe your stat. You know, the joke we always have is, you know, anybody from a networking background, there's no such thing as eliminating a bottleneck, you just move it. And if you talk about utilization, we've been playing the shell game for my entire career of, let's try to optimize one thing and then, oh, there's something else that we're not doing. So,you know, so important. Retail, I want to hear from your standpoint, you know, virtualization and HPC, you know, AI type of uses there. What value does this bring to you and, you know, and key learnings you've had in your organization? >> So, we as a university are a big users of the VMware technologies starting from the traditional enterprise workload and VPI. We started from there in the sense that we have an installation quite significant. But also almost all the services that the university gives to our internal users, either personnel or staff or students. At a certain point that we decided to try to understand the, if a VMware virtualization would be good also for scientific computing. Why? Because at the end of the day, their request that we have from our internal users is flexibility. Flexibility in the sense of be fast in deploying, be fast to reconfiguring, try to have the latest beats on the software side, especially on the AI research. At the end of the day we designed a VMware solution like you, I can say like a whiteboard. We have a whiteboard, and we are able to design a new solution of this whiteboard and to deploy as fast as possible. Okay, what we face as IT is not a request of the maximum performance. Our researchers ask us for flexibility then, and want to be able to have the maximum possible flexibility in configuring the systems. How can I say I, we can deploy as more test cluster on the visual infrastructure in minutes or we can use GPU inside the infrastructure tests, of test of new algorithm for deep learning. And we can use faster storage inside the virtualization to see how certain algorithm would vary with our internal developer can leverage the latest, the beat in storage like NVME, MVMS or so. And this is why at the certain point, we decided to try visualization as a base for HPC and scientific computing, and we are happy. >> Yeah, I think Maurizio described it it's flexibility. And of course, if you think optimal performance, you're looking at the bare medal, but in this day and age, as I stated at the beginning, there's so much technology, so much infrastructure available that flexibility at times trump the raw performance. So, when you have two different research departments, two different portions, two different parts of the company looking for an environment. No two environments are going to be exactly the same. So you have to be flexible in how you aggregate the different components of the infrastructure. And then think about today it's actually fantastic. Maurizio was sharing with me earlier this year, that at some point, as we all know, there was a lot down. You could really get into a data center and move different cables around or reconfigure servers to have the right ratio of memory, to CPU, to storage, to accelerators, and having been at the forefront of this enablement has really benefited University of Pisa and given them that flexibility that they really need. >> Wonderful, well, Maurizio my understanding, I believe you're giving a presentation as part of the activities this week. Give us a final glimpses to, you know, what you want your peers to be taking away from what you've done? >> What we have done that is something that is very simple in the sense that we adapt some open source software to our infrastructure in order to enable our system managers and users to deploy HPC and AI solution fastly and in an easy way to our VMware infrastructure. We started doing a sort of POC. We designed the test infrastructure early this year and then we go fastly to production because we had about the results. And so this is what we present in the sense that you can have a lot of way to deploy Vitola HPC, Barto. We went for a simple and open source solution. Also, thanks to our friends of Dell Technologies in some parts that enabled us to do the works and now to go in production. And that's theory told before you talked to has a lot during the pandemic due to the effect that stay at home >> Wonderful, Thierry, I'll let you have the final word. What things are you drawing customers to, to really dig in? Obviously there's a cost savings, or are there any other things that this unlocks for them? >> Yeah, I mean, cost savings. We talked about flexibility. We talked about utilization. You don't want to have a lot of infrastructure sitting there and just waiting for a job to come in once every two months. And then there's also the world we live in, and we all live our life here through a video conference, or at times through the interface of our phone and being able to have this web based interaction with a lot of infrastructure. And at times the best infrastructure in the world, makes things simpler, easier, and hopefully bring science at the finger tip of data scientists without having to worry about knowing every single detail on how to build up that infrastructure. And with the help of the University of Pisa, one of our centers of excellence in Europe, we've been innovating and everything that's been accomplished for, you know at Pisa can be accomplished by our customers and our partners around the world. >> Thierry, Maurizio, thank you much for so much for sharing and congratulations on all I know you've done building up that COE. >> Thanks to you. >> Thank you. >> Stay with us, lots more covered from VMworld 2020. I'm Stu Miniman as always. Thank you for watching the theCUBE. (soft music)