What We Mean by Software-Defined Storage

first_imgIt’s day two of EMC World 2013 in Las Vegas. Everybody here is talking about or asking about software-defined storage. Let me give you a couple of analogies to help you understand what we mean by software-defined storage and what we at EMC are doing. There are two main components to keep in mind.Virtualize Everything, Compromise NothingIn my living room at home, I have a half-dozen control devices for my TV, my set top box for streaming Internet content to my TV, a DVR, a video game that my kids use more than I do, a DVD player, and an amplifier. There is nothing more annoying than not being able to watch a program because my kid has hidden the remote. I’d rather not have to search for it and fiddle with it. I’d rather just watch the program. It’s the same in the data center. You need dedicated operators for different infrastructures. And it’s a real pain.To control the devices in your living room, or to control the different devices in your data center, you need a common way of doing it. It’s more efficient. The folks in the data center can be way more productive. And it’s more reliable, because you tend to do things in a standard way.So if the universal remote control is what controls your home entertainment devices in your living room, software-defined storage controls the physical arrays in your data center. That’s the first part of what we are doing with software-defined storage.But that’s only half of it. In addition to providing a new model for storage management, software-defined storage also provides a new architecture for new kinds of workloads. And this is where the story really gets interesting.To continue the home entertainment analogy, a lot of the programming that was created for your traditional television set really didn’t produce a lot of high demands. Not all that long ago, you could only view about 525 lines or so of a TV picture if you were in the U.S., or 625 lines, I believe, if you were in Europe. But, either way, there really wasn’t any killer content to make TV come more alive.Over time, the entertainment industry went a long way to make movies and programming look phenomenal. But the regular, traditional TV just wouldn’t do it justice. So, in order to view the new movies and programming in all its glory, we needed something to interpret that content type. Without an HDTV and an HD DVR, you just didn’t get the full experience. So, people went out and got the wide screen, HDTV to watch the Super Bowl. And they kept the smaller TV they had in the kitchen for convenience. Different devices for different types of content.We see the same thing happening in data centers. A lot of new applications are being built in a different way, and we need new capabilities to interpret the new content types. So think of software-defined storage as enriching the underlying storage arrays to interpret these new content types that many of these new applications will be building.Now, traditional applications are not going away. Industry analysts estimate that traditional workloads will expand by 70% between 2012-2016. These workloads are typically block and file workloads. They require a high degree of transactional consistency. And a lot of the resiliency is built into the hardware itself. But we think the real opportunity going forward is going to be in next-generation cloud applications. These are a relatively small number of apps today, but they are growing explosively — ten times faster than traditional workloads! Typically, they are not going to require a block or file storage infrastructure. Object storage is what they are going to demand.We think we have solved many of these challenges with a product we call EMC ViPR. It is not just a point product. It is a platform for software-defined storage – both a new model for storage management, as well as the data services that provide a new architecture for new kinds of workloads. To learn more about it, see yesterday’s post, What is Software-Defined Storage?, by my colleague Amitabh Srivastava, head of EMC’s Advanced Storage Division.last_img read more

Read More →

Building a Better Cloud Starts with Getting the Best People

first_imgAs we close in on the end of our fiscal year, it’s amazing to see how much success we’ve had on the Dell Technologies Cloud team in our inaugural year. This isn’t a blog post to pat ourselves on the back or take the victory lap; it’s an acknowledgement that we still have a long way to go to achieve our goals, and that means growing the team.In this new decade, what organizations require from their cloud solutions partners has substantially changed. No one is clamoring for yet another public cloud vendor to further splinter things. No, what organizations need is a partner who can drive consistency across these disparate environments. And who better than Dell Technologies, a leader not only in infrastructure solutions, but also virtualization and cloud software, with a strong commitment to building the solutions that organizations have come to rely on so heavily.With that in mind, we launched Dell Technologies Cloud, the first set of solutions that show the path forward for Dell Technologies: Embracing Cloud Operating Models everywhere. This direction permeates every aspect of our business from engineering, to sales, to marketing. And because of this we are actively expanding our team, looking for the best and the brightest to join us on this rocketship.Here is why you should believe in us, and why this is a great organization to come into:Market Acceptance – You might be saying, “why should I select Dell Technologies versus one of the hyperscalers?” The truth is we are one of the largest and most trusted infrastructure and software companies in the world, and our launch of Dell Technologies Cloud has been praised by analysts and customers alike.Ability to Make an Impact – Since the solution is relatively new to market and we’re rapidly transforming to meet new markets and new customers’ needs across the board, we have a startup-like culture, backed with the investments only the largest companies can make.Smart and Pragmatic Leadership – We enter spaces with an obvious market need and where we hold a competitive advantage, versus getting into protracted and messy battles where either side’s success looks more like mutually assured destruction. Because of this, even during challenging economic conditions we’ve executed well and taken share.Strong Culture – I’ve spent time in many different organizations but Dell’s commitment to work/life balance and ability to foster a collaborative “one team” culture is truly unique. I can’t stress enough how amazing it is to be able to ask for help and see folks willing to support as opposed to having them say “that’s not in my KPIs.”Career Trajectory – From what I’ve observed in my almost three years here, and looking at this team specifically, we’re in a high growth area that can lead to great career growth. From my own experience I would say it’s been the most I’ve ever been rewarded for the work I’ve done.Come join us! Take on the challenge of educating an entire market that the next decade must deliver consistent experiences across all the clouds. It’s a great company to work for.Please visit the Dell Technologies career site and search for “cloud” jobs there. We have many openings including Product Marketing, Sales Enablement, Engineering, and Product Management roles that we’re actively filling.last_img read more

Read More →

Supercharging AI and HPC Engines with Dell Technologies and NVIDIA

first_img Play VideoPlayMuteCurrent Time 0:00/Duration Time 3:17Loaded: 0%Progress: 0%Stream TypeLIVERemaining Time -3:17 Playback Rate1ChaptersChaptersdescriptions off, selectedDescriptionssubtitles off, selectedSubtitlescaptions settings, opens captions settings dialogcaptions off, selectedCaptionsdefault, selectedAudio TrackFullscreenThis is a modal window.Caption Settings DialogBeginning of dialog window. Escape will cancel and close the window.TextColorWhiteBlackRedGreenBlueYellowMagentaCyanTransparencyOpaqueSemi-TransparentBackgroundColorBlackWhiteRedGreenBlueYellowMagentaCyanTransparencyOpaqueSemi-TransparentTransparentWindowColorBlackWhiteRedGreenBlueYellowMagentaCyanTransparencyTransparentSemi-TransparentOpaqueFont Size50%75%100%125%150%175%200%300%400%Text Edge StyleNoneRaisedDepressedUniformDropshadowFont FamilyProportional Sans-SerifMonospace Sans-SerifProportional SerifMonospace SerifCasualScriptSmall CapsDefaultsDoneClose Modal DialogThis is a modal window. This modal can be closed by pressing the Escape key or activating the close button. Play VideoPlayMuteCurrent Time 0:00/Duration Time 1:54Loaded: 0%Progress: 0%Stream TypeLIVERemaining Time -1:54 Playback Rate1ChaptersChaptersdescriptions off, selectedDescriptionssubtitles off, selectedSubtitlescaptions settings, opens captions settings dialogcaptions off, selectedCaptionsen (Main), selectedAudio TrackFullscreenThis is a modal window.Caption Settings DialogBeginning of dialog window. Escape will cancel and close the window.TextColorWhiteBlackRedGreenBlueYellowMagentaCyanTransparencyOpaqueSemi-TransparentBackgroundColorBlackWhiteRedGreenBlueYellowMagentaCyanTransparencyOpaqueSemi-TransparentTransparentWindowColorBlackWhiteRedGreenBlueYellowMagentaCyanTransparencyTransparentSemi-TransparentOpaqueFont Size50%75%100%125%150%175%200%300%400%Text Edge StyleNoneRaisedDepressedUniformDropshadowFont FamilyProportional Sans-SerifMonospace Sans-SerifProportional SerifMonospace SerifCasualScriptSmall CapsDefaultsDoneClose Modal DialogThis is a modal window. This modal can be closed by pressing the Escape key or activating the close button. Dell Technologies and NVIDIA deliver unprecedented acceleration and flexibility for AI, data analytics and HPC to tackle some of the world’s toughest computing challenges.Dell Technologies has long been on the forefront of innovation when it comes to artificial intelligence and high performance computing, and in many cases NVIDIA has been right there with us.This close relationship of two global technology leaders got started well before 2013, when Dell and NVIDIA worked together to advance the concept of a dense GPU platform, which would emerge in the Dell EMC PowerEdge server family. This concept came to market in 2014, when Dell introduced the industry’s first dense 1U four-GPU platform — the Dell EMC PowerEdge C4130 server.This legacy of innovation continued in the days that followed, and in 2016 Dell Technologies became the first OEM to support NVIDIA SXM2 technology. This technology, which is used in the NVIDIA Tesla V100 SXM2 GPU, enables high-speed, direct GPU-to-GPU communication via the NVIDIA NVLink interconnect.The following year, in 2017, Dell Technologies became the first vendor to support the NVIDIA Volta V100 GPU in a unique, balanced CPU-to-GPU dense platform — the Dell EMC PowerEdge C4140 server. This accelerator-optimized, high-density server is built to meet the demands of cognitive computing workloads in a 1U server format that supports four GPUs.In the years since, we have built on our strategic partnership to bring new Dell EMC Ready Solutions for AI to market. These offerings include Deep Learning with NVIDIA, a Dell EMC Ready Solution that provides a GPU‑optimized solution stack that helps organizations shave valuable time from deep learning projects. The Deep Learning with NVIDIA solution is built around NVIDIA Tesla V100 GPUs.Success storiesOver the years, the team of Dell Technologies and NVIDIA has had great success with our common customer base. Here’s a small sample of our mutual successes:We worked closely on a major expansion of the HPC5 supercomputer at Eni, an Italy-based integrated energy company. This system is based on 1,820 Dell EMC PowerEdge C4140 servers, each with four NVIDIA V100 Tensor Core GPUs and two 2nd Generation Intel® Xeon® Scalable processors. With the performance of the HPC5 supercomputer, Eni’s scientists and engineers can use extremely sophisticated in-house algorithms to process subsoil data. We worked in collaboration with the Research Computing Service at the University of Cambridge to develop the Wilkes2 supercomputer, which debuted in 2017 as a TOP500 system and the largest GPU-enabled supercomputer for academic research in the UK. This system is based on Dell EMC PowerEdge servers, each with four NVIDIA GPUs. It supports a wide-range of compute- and data-intensive research, from aeronautical turbulence and automotive combustion simulations to materials modeling and medical applications.center_img We collaborated on the development of one of Canada’s most powerful academic supercomputers, the Super Cedar system at Simon Fraser University. This TOP500 system, launched in 2017, revs up scientific studies with the power of 584 NVIDIA GPUs. It is helping thousands of researchers from British Columbia and across Canada collect, analyze, share and store immense volumes of data. Market leadershipIn developing products and solutions with NVIDIA, Dell Technologies builds on a broad and deep product portfolio that spans from best-in-class servers with the latest processing architectures to leading-edge storage and networking systems, all backed by expert services.In a sign of this leadership, Dell Technologies took the top spot in both server revenue and shipments in 2019, with 20.5 percent market share and 16.3 percent market share, respectively, according to Gartner. Meanwhile, Dell Technologies  maintains “a commanding market share lead” in high-performance storage, according to a 2019 report from Intersect360 Research. The firm noted that Dell Technologies also has the No. 1 revenue share position for storage for HPC applications and  is the overall leader in total HPC solutions revenue. Looking aheadToday, we’re excited about our opportunities to support the new NVIDIA A100 Tensor Core GPU in systems, solutions and services that accelerate workloads, from edge to core to cloud, just as we support other NVIDIA GPU accelerators and technologies in a wide range of our offerings.This new GPU is designed to help users tackle some of the world’s toughest computing challenges. It delivers unprecedented acceleration at every scale for AI, data analytics and HPC. As the engine of the NVIDIA data center platform, A100 can efficiently scale up to thousands of GPUs, or it can be partitioned into isolated GPU instances to accelerate workloads of all sizes.NVIDIA reports that the A100 accelerates major deep learning frameworks and more than 700 HPC applications — and that’s one of the reasons why we at Dell Technologies are excited to incorporate this GPU in our portfolio for HPC, AI and data analytics.To learn moreFor a look at how the San Diego Supercomputer Center is capitalizing on NVIDIA GPUs in its new Expanse supercomputer from Dell Technologies, see the case study “Architecting for Mixed Workloads at Massive Scale.”last_img read more

Read More →