Dell EMC and NVIDIA Expand Collaboration to Deliver Flexible Deployment Options for Artificial Intelligence Use Cases

As organizations strive to gain a competitive edge in an increasingly digital global economy, Artificial Intelligence (AI) is garnering a lot of attention. Not surprisingly, AI initiatives are springing up in various business domains, such as manufacturing, customer support, marketing and sales. In fact, Gartner predicts that AI-derived global business value is forecast to reach $3.9 trillion by 2022. As companies scramble to determine how to turn the promise of AI into reality, they are faced with a multitude of complex choices related to software stacks, neural networks and infrastructure components, with significant implications on the time-to-value associated with these initiatives.

In such a complex environment, it is critical that organizations be able to rely on vendors that they trust. Over the last few years, Dell EMC and NVIDIA have established a strong partnership to help organizations accelerate their AI initiatives. For organizations that prefer to build their own solution, we offer Dell EMC’s ultra-dense PowerEdge C-series, with NVIDIA’s TESLA V100 Tensor Core GPUs, which allows scale-out AI solutions from four up to hundreds of GPUs per cluster. For customers looking to leverage a pre-validated hardware and software stack for their Deep Learning initiatives, we offer Dell EMC Ready Solutions for AI: Deep Learning with NVIDIA, which also feature Dell EMC Isilon All-Flash storage.  Our partnership is built on the philosophy of offering flexibility and informed choice across a broad portfolio.

“We’re now in the age of AI, where every industry will be enabled and powered by AI,” said Ian Buck, vice president and general manager of Accelerated Computing at NVIDIA. “Working together, NVIDIA and Dell EMC are bringing some of the most powerful GPU-accelerated solutions to enterprises so they can quickly integrate AI into their business.”

Today our companies are expanding our collaboration by announcing a new reference architecture for AI featuring NVIDIA DGX-1 servers complemented with the high-performance of Dell EMC Isilon All-Flash network-attached storage (NAS). This new offer pairs the NVIDIA DGX-1, purpose-built for the unique demands of AI and deep learning, powered by eight advanced data center accelerators – the NVIDIA Tesla V100 Tensor Core GPU – with Dell EMC Isilon scale-out all-flash storage. This combination is designed to give customers more flexibility in how they deploy AI, and delivers breakthrough performance for large-scale deep learning. As Dell EMC often takes the role of trusted IT advisor with a broad solution portfolio, this enables us to support customers looking to deploy NVIDIA DGX-1 with Dell EMC’s scale-out storage and do so more simply and with less risk.

Key components of this offering include:

◈ NVIDIA DGX-1 servers which integrate up to eight NVIDIA Tesla V100 Tensor Core GPUs fully interconnected in a hybrid cube-mesh topology. Each DGX-1 server can deliver 1 petaFLOPS of AI performance, and is powered by the DGX software stack which includes NVIDIA-optimized versions of the most popular deep learning frameworks, for maximized training performance.

◈ Dell EMC Isilon All-Flash scale-out NAS storage delivers the scale (up to 33 PB), performance (up to 540 GB/s), and concurrency (up to millions of connections) to eliminate the storage I/O bottleneck and accelerate AI workloads at scale.

To validate the new reference architecture, we ran multiple industry-standard image classification benchmarks using 22 TB datasets to simulate real-world training and inference workloads (check out the details in this joint white paper).  This testing was done on systems ranging from one DGX-1, all the way to eight DGX-1 servers connected to eight Isilon F800 nodes. The results from two of the key benchmarks are shown below in Figures 1 and 2.  Just as we proved out with the Ready Solution for Deep Learning with C4140 servers, Isilon was able to keep pace and linearly scale to achieve the maximum performance of the compute layer.

Figure 1: ResNet-50 Model Training Benchmark

Figure 2: Inception-v4 Large Image Classification

Here are some of the key findings from our testing of the Isilon and NVIDIA DGX-1 reference architecture:

◈ Achieved compelling performance results across industry standard AI benchmarks from eight through 72 GPUs without degradation to throughput or performance.

◈ Linear scalability from 8-72 GPUs delivering up to 19.9 GB/s while keeping the GPUs pegged at >97% utilization.

◈ The Isilon F800 system can deliver up to 96% throughput of local memory, bringing it extremely close to the maximum theoretical performance limit an NVIDIA DGX-1 system can achieve.

With superior capabilities for low latency, high throughput and massively parallel I/O, Dell EMC Isilon and NVIDIA DGX-1 servers complement each other extremely well to address deep learning workloads, effectively compressing the time needed for training and testing analytical models for multi-petabyte data sets on AI platforms.

This new reference architecture extends the commitment of both Dell EMC and NVIDIA to make AI simple and more accessible to organizations wishing to apply AI and deep learning techniques to develop new applications or solve complex data-centric problems. This pre-validated reference architecture adds to our unmatched set of joint offerings across Dell Precision workstations, Dell EMC PowerEdge servers, Dell EMC Ready Solutions and Dell EMC Networking.

Stay tuned for more details on the performance benchmarks of this reference architecture and for upcoming announcements as our server, storage, and software innovations for AI and deep learning use cases continue to evolve.

Continue reading

Dell EMC and the 5G Network Transformation

Applying Digital, IT, Security and Workforce Transformation to 5G Networks

I’ve blogged about 5G and the mobile network transformation a few times, and have spent countless hours, at this point, speaking with many of Dell EMC’s top service provider customers and partners on how the 5G network transformation will play out. To date, I’ve refused to walk into meetings with canned slides that capture the enormity of 5G in orders-of-magnitude (more devices, more data, lower latency, etc.), nor with specific 5G solutions, because I’ve wanted to focus specifically on how 5G is driving different behaviors:

◈ Different architectural methodologies
◈ Different operational models
◈ Different procurement processes
◈ Different product and services offerings

Across the board, in all my conversations, it is becoming increasingly apparent that 5G is the perfect storm of once-in-a-decade infrastructure modernization that will bring together the key trends we, as an industry, have been speaking about for several years. Bear with me as I name-drop some of these key trends – Software-Defined Everything, Cloud Native, DevOps, Artificial Intelligence / Machine & Deep Learning, APIs, Automation, Orchestration etc….

More importantly, though, I think there are four higher-layer transformations occurring, and these technologies are just an ephemeral set of technologies and architectures which form the initial foundation of those transformation.  Dell Technologies has not been shy about its viewpoint on the importance of technology to help drive organizational innovation and success. I believe, in some ways, the term “5G” has become the aggregate of the four approaches to technology transformation – Digital Transformation, IT Transformation, Workforce Transformation and Security Transformation – applied to the mobile network.

Let me share some thoughts on how these apply to 5G:

◈ Digital Transformation – monetization of data analytics and artificial intelligence. Whether we talk about autonomous vehicles, Augmented Reality, or IoT – at the core of just about every 5G network transformation business case is the need to transport, aggregate, process, and react to massive amounts of data collected from “things.” For the first time in any of the “G” evolutions, services are top-of-mind for service providers, and we are witnessing a use-case driven network upgrade focused on Digital Services forming the foundation of future revenues.

◈ IT Transformation – modernizing infrastructure to support automated processes. New use-cases are great, but they cannot be realized by doing what has always been done – building proprietary, vertically-integrated networks that are opaque to services and infrastructure operations. Virtualization (both hypervisors and containers) and open networking, and storage, provide the foundation for a 5G multi-cloud future. This future is where Cloud Radio Access Networks (C-RAN), Edge Clouds, Telco Clouds, and Public Clouds are fully integrated, and real-time telemetry and automation provide the tools to more efficiently build and manage services.

◈ Workforce Transformation – providing the tools and skills for operations to take advantage of the new infrastructure capabilities. IT Transformation is only as successful as the workforce capable of taking advantage of it. Such a workforce is skilled in more than just mobile networking and organized to take advantage of integrated development and operations (“DevOps”). Cloud native, horizontal 5G networks won’t be operated in the same way that proprietary, vertically-integrated 4G networks were. Mobility and Networking experience are no longer sufficient – operational “excellence” will integrate this domain knowledge data science and computer science skill sets (data modeling, API scripting/programming, application development).

◈ Security Transformation – integrating natively into all services and processes, rather than as an overlay. These foundational shifts to the services, infrastructure, and workforce cannot be done in isolation of a security transformation. Networking and security are evolving in tandem, and 5G is a beneficiary of that evolution. I see security as both an opportunity to monetize the 5G network with unique Security-as-a-Service offerings but also as a foundational overlay to all services, infrastructure, and processes. Point products, such as firewalls and intrusion detection systems (IDS) will give way to data-driven, automated security frameworks that leverage the underlying programmable infrastructure for both visibility and enforcement.

Needless to say, each of these transformations and how they relate to 5G is complex. That is why Dell EMC is not sitting still in 5G. We continue to work closely with industry leaders to define how infrastructure, operations, and services are realized through 5G network transformation, and how we define and realize 5G ready infrastructure at-scale, with the performance and ease-of-deployment needed to capture new Digital Service opportunities profitably.

Continue reading

Accelerating OpenStack Deployment with Dell EMC and Red Hat

Service providers everywhere are challenged by growing CAPEX and OPEX for existing network infrastructure, long time to market for new or updated services, price and margin erosion in existing businesses models and slowing innovation to capture new revenue opportunities. To address these challenges, service providers globally are looking to transform their datacenters from proprietary and rigid architectures to a disaggregated model with open architectures which allows them to increase infrastructure flexibility, efficiency and scalability.

Dell EMC and Red Hat have been working together for over a decade to provide jointly engineered and fully-validated solutions to help service providers with this transformation. At the OpenStack Summit in Berlin, we are excited to announce the latest version of Dell EMC Ready Architecture for Red Hat OpenStack Platform v13.0 release for Communications Service Providers as well as Managed & Hosting Service Providers and high-end enterprises.

At its core, we have added several new enhancements to help service providers rapidly deploy OpenStack. These include:

1. Support for Red Hat OpenStack 13 Queens Release: Red Hat OpenStack 13 is the latest release and includes several key enhancements:

◈ All Red Hat OpenStack Platform services are now containerized including OpenStack Networking (Neutron), OpenStack Block Storage (cinder), and OpenStack Shared File Systems (Manila).

◈ Red Hat OpenStack Director supports fast forward upgrade from Red Hat OpenStack Platform 10 to Red Hat OpenStack Platform 13.

◈ Real-Time Kernel-based Virtual Machine (KVM) integration and Real-Time KVM Compute role for NFV workloads to support guests with stringent latency requirement.

◈ OpenDaylight is a flexible, modular and open SDN platform that is now fully supported with this release.

2. Support for Dell EMC S5428-ON Switch: The S5248-ON is a fixed form factor top-of-rack switch offering multiple options of 25GbE SFP28 ports for in-rack server and storage. The S5248-ON key features include enhanced buffering, higher forwarding tables and hardware support for Layer 3 VXLAN while continuing to support Open Networking.

3. Cumulus Linux Networking OS: To support our service providers who tend to be Linux savvy and want to use Layer 3 Border Gateway Protocol (BGP) Ethernet VPN (EVPN) fabric for data centers, this release  supports Cumulus Linux as the networking OS running on top of Dell EMC S5248 Open Networking Switches.

4. NFV Enhancements: We continue to enhance our NFV features and add new features that many of our customers are asking for. These include enhancements such as OVS-DPDK, Neutron enabled SR-IOV, MTU support starting with jumbo frames, and Neutron DVR.

5. Additionally, Dell EMC continues to have the one of the most comprehensive of Ironic driver coverage for bare-metal provisioning on Dell EMC servers and covers IPMI, iDRAC and RedFish drivers. Dell EMC also is the first server vendor to have Red Hat Bare Metal certification for PowerEdge servers (in addition to Red Hat OpenStack Platform, Linux and GlusterFS).

While Dell EMC Ready Architecture for Red Hat OpenStack Platform is intended as a baseline conversation starter, we fully understand that service providers may have standardized on an architecture that includes their best practices. Dell EMC makes customization of the solution easy with our Professional Services offerings targeted specifically to address their requirements.

Looking ahead, Dell EMC and Red Hat together will continue to integrate capabilities that service providers want to reduce installation complexity, automate operations, enable seamless delivery new services.

Continue reading

New 200TB Fast Track Solution Proves Dell EMC is the Go-To Platform for SQL Server Success

Dell EMC now offers the LARGEST, SMALLEST and MOST array-based solutions in Microsoft’s SQL Server Fast Track catalog – what else would you expect from the leader in midrange storage?

Today’s business thrives on a steady diet of data – and Microsoft SQL Server is one of the premier technologies helping companies of all sizes capitalize on the value of that data, delivering real-time operational intelligence to create competitive advantage in nearly every industry.

Some assembly required?

But before you can achieve the benefits of SQL Server, you’ll need more than just software – an entire ecosystem is involved.  Whether the goal is traditional high-scale OLTP, advanced analytics, or the foundation of an AI/ML solution, successful deployments also require:

◈ Powerful compute resources to efficiently drive millions (or billions) of transactions

◈ Fast, flexible storage to cost-effectively handle a growing tsunami of input

These essential elements must work together with the host operating system, network infrastructure and SQL Server itself to deliver a throughput flow optimized for the specific database use desired.  If key components are sized incorrectly or misconfigured, you may end up spending too much, wasting personnel resources – or worst of all, not achieving the performance or flexibility required to meet your business objectives.  The necessary multi-disciplinary planning and coordination can be extremely complex.  

Take the SQL Fast Track

That’s where Dell EMC’s Microsoft Data Warehouse Fast Track solutions come in.  Fast Track is a joint effort between Microsoft and its infrastructure partners to deliver validated, pre-configured deployment guides that reduce the complexity of advanced data warehouse projects. These detailed solution “cookbooks” include ready-to-order hardware specs for balanced compute and storage systems, test-validated OS and software settings, cabling diagrams, and everything else required to ensure your deployment goes quickly and correctly the first time.

As a long-standing Microsoft partner, it’s not surprising Dell EMC takes a leadership position in the Fast Track program, where we now offer the LARGEST (200TB), the SMALLEST (8TB) and the MOST array-based solutions (we’ve documented 14 diverse configurations and counting).*

Large-scale SQL powerhouse

Our latest Fast Track solution, certified by Microsoft this week, is the highest-capacity from any vendor in the catalogue. This 200TB scalable blueprint is suitable for leading innovators in industries such as banking and healthcare – or anyone else with mission-critical, performance-intensive application needs.

◈ 200TB Reference Architecture

◈ 200TB Deployment and Configuration Guide

Featuring our Dell EMC SC9000 array and Dell EMC PowerEdge R940 servers, the new configuration provides true enterprise-class capabilities.  SC9000 scales to 6PB per array, with multi-array federation, intelligent deduplication/compression, automated RAID tiering and auto-failover for built-in disaster recovery. The SCOS 7.3 array firmware update (see video overview) recently boosted SQL transactions-per-second by 44% (!), allowing the 200TB Fast Track system to hit 1 million IOPS, with max IOPS for the array itself now over 2.2 million.

PowerEdge R940 also drives phenomenal SQL performance with quad Intel Xeon processors and up to 112 cores, 6TB of memory and 12 NVDIMMs.  Automated server management simplifies IT operations, built-in security features protect your assets – and together, these best-in-class Dell EMC products enable high availability architectures with Windows Server, Hyper-V and VMware.

Affordable SQL entry

But big companies aren’t the only ones who need state-of-the-art business intelligence.  Fortunately, an SC Series/PowerEdge combination also powers the smallest array-based solution in the Fast Track portfolio. Our 8TB SQL SCv3020 array / R740 server configuration delivers high-value services surprisingly similar to the 200TB solution described above, allowing businesses of all sizes to start small without sacrificing the advantages of enterprise-class technology.

◈ 8TB Reference Architecture

◈ 8TB Deployment and Configuration Guide

And everything SQL in between

If neither of the above “starting point” configs suits your unique needs, there are literally a dozen other Dell EMC solutions in the Fast Track catalogue, including a 100TB configuration using the same affordable SCv3020 / R740 combination. Easy evolution between solutions is a hallmark of SC Series storage – but all the Dell EMC Fast Track configurations offer the assurance of single-vendor simplicity and integration from the market leader in both storage and servers. Once you start with Dell EMC, you know we can take you wherever your business needs to go.

Entry to enterprise. Trusted. Validated. It’s why thousands of customers worldwide choose Dell EMC as their one-stop SQL shop. Because in the end, neither Dell EMC nor Microsoft will be satisfied until we make your SQL Server deployment a key enabler of your company’s success.

Continue reading

Did Software Eat the World? Or was it Data Capital?

Marc Andreessen coined the phrase “software is eating the world” almost ten years ago. His proclamation was clairvoyant because it predicted the rise of the digital age and its corresponding impact including the advantageous position the companies who were leading the charge would have. It strikes me that we’ve transitioned from software to data – novel data – that now defines the value organizations will have. Moreover, I believe the value has always rested within the data, it was just waiting for software and hardware innovation in order to gain access to that value.

Today it is the users who have been utilizing data the most, but in the future it will be machines. This change will be similar to the way AI, IoT, and data analytics currently dwarf applications in their consumption of data. This rise will require a massive shift, not only in technology investments but also how these companies are structured and operate. We are only just beginning to understand the business value of data and how it can be leveraged effectively to create a digital business. Technology companies inadvertently created a blueprint when they started disrupting industries, and now it is easy to find examples in almost any sector where data is used.

Software eating the world in the consumer space

Perhaps the best way to showcase the potential of data comes from how it affects our everyday lives as consumers, since this has been an area where so many of us have seen the most extensive change.

I recently spent two weeks in Japan, which despite barely speaking Japanese, was one of the easiest travel experiences I’ve ever had. My tour guide was a smartphone and its applications offered insights on every level. It was easier to get around cities of magnitude far larger than the city where I live. Consumers are trained that “there’s an app for that,” and each application offers a unique value that made the entire trip a breeze. I couldn’t imagine how difficult this trip would have been without my smartphone applications.

From finding my way around a complex series of train-lines, to remotely checking into residences, to finding the best food in town – the experience was seamless and none of the details got in the way of our vacation as we traversed city after city. Probably the most telling moment was when I recalled there was a temple I had visited 19 years ago that I really wanted to see again but didn’t know how to find it again. I searched using three words: Phoenix, temple, and Kyoto. The first hit was Byodoin temple or Phoenix Hall. Using just two data points – the location and a partial description had yielded the destination, and those were not even the most accurate descriptions as Uji is the exact location, and the actual name was different.

We live in a time where getting lost, picking the wrong restaurant, or getting stuck somewhere have become optional due to our consumption and leveraging of data in our personal lives. What could that same approach yield us in business?

The rise of Data Capital as a business asset

With Smartphone adoption estimates topping 80% market penetration, we demand these types of experiences in our daily lives. Why are we still running businesses without access to that same level of data? Yes, it requires a new approach and investments, but as we’ve seen in our personal lives it would be unthinkable to continue operating a business without being data-centric. Put simply, every company is now a technology company. Where will we see data having the most significant impact on companies? In four major areas:

Digital Experiences

As we noted, digital experiences have been the way in which we have most commonly leveraged data, and these will continue to persist and grow as every industry becomes infused with technology. We’ve already seen massive disruption from digital firms challenging incumbents and disrupting industries. As data becomes connected with every aspect of our lives, these offerings will be a strong differentiator for businesses that embrace this change. The physical world will merge with the online world, offering new connected possibilities – whether it’s a web application served up to a phone or an AR/VR experience that has the potential reshape nearly every activity.

Big Data/Analytics

Another area businesses have already had some success with is big data, which offers new insights into the business. Whether it’s identifying buying patterns, segmenting customers, or identifying inefficiencies in a supply chain – if something can be measured and analyzed then it can likely be improved. It’s hard to imagine a business remaining competitive for long if it is not data-driven, as competitors will run leaner and be able to capture opportunities the data-starved organization will miss. Many industries have already experienced massive disruption as a result of digital companies such as Facebook, Netflix, Airbnb, and Uber entering their markets. With the digital era upon us, data will become a substantial competitive advantage.

The Internet of Things

While this category was coined way back in 1995 at the dawn of the Internet, it has largely remained an elusive target for most organizations. With faster wireless speeds (4G/5G) and more and more data being collected, analyzed, and consumed at the edge by devices, we are finally starting to see IoT redefine how things are done across many industries. Whether it’s enabling new experiences for humans by incorporating a device with a connected service or automating activities with robotics, IoT promises to reshape how business gets done in the future, and it is data that powers these systems. These systems can offer unique experiences and streamline processes, allowing organizations to lock in profitability and offer better service to customers.

 Machine Learning / Deep Learning and AI

One of the hottest areas for tech investment right now is Artificial Intelligence (AI). We are rapidly approaching a time where the sheer volume of data becomes a barrier to analysis by human beings, and this is where AI delivers value. The software and hardware evolution enables advanced algorithms to access enormous datasets in new and differentiated ways, with the ability to turn data into thoughts or actions. AI is often combined with these other three trends we’ve talked about to provide an extension of the value they offer. AI can enable companies to tackle simple tasks or address incredibly advanced use cases where human beings are not well suited to analyze the mass of data in question. Whether it’s an assembly line picker, an autonomous vehicle, or analyzing a human genome – it’s important to note that large stores of data must be collected and high computational power must be applied to achieve accurate results. AI while incredibly complex perhaps represents the pinnacle of data capital and will deliver additional value beyond these other means of leveraging data.

Data Capital is required in the future the primary way to access consumers

With consumers, it’s a choice to use data as a means to enrich an experience. However, for businesses it’s a mandate. So why are consumers so far ahead of the curve in using data compared to so many businesses? It’s because large-scale enterprises tend to have a legacy infrastructure and “data gravity” weighing them down. It’s not as easy as just buying a new Smartphone with VR and facial recognition capabilities. There is a need to modernize and safely navigate the transition of data to modern platforms. Ultimately it is perceived to be a very time consuming and precarious venture to introduce new data capabilities, but it doesn’t have to be.

Dell EMC recognizes this is the critical challenge facing many organizations and is focused on helping them succeed no matter what stage they may be in their journey. Our storage and data protection solutions help organizations unlock their data capital as a way to strengthen and differentiate their businesses. We’ve designed our products with this in mind, and it’s also how we assist organizations to make this transition.

While some vendors suggest a complete “lift and shift” cloud strategy, and others position consolidating to a single storage hub, we understand your business is so much more than any one data platform. Our focus is on delivering the best of breed data storage and data protection solutions that provide coverage for the entire data footprint, and we work with you to determine where data should reside based on its characteristics and your business objectives.

Continue reading

Industry First — Dell EMC Showcases End-to-End NVMe and Storage Class Memory on PowerMax at VMworld Barcelona

I’m excited to share that this week at VMworld Europe, Dell EMC is previewing our latest PowerMax technology combined with another industry leader, VMware. This first of its kind technology preview continues to demonstrate Dell EMC’s commitment to storage innovation for virtual environments and the tight collaboration between Dell EMC and VMware. Our proven track record of providing next-generation technology helps our customers tackle their most demanding and mission-critical applications of today and tomorrow.

The entire engineering team and I are proud that our customers have made Dell EMC PowerMax the de facto standard for applications that require unmatched resiliency, uncompromising data services, and extreme levels of performance. Combined with the efficiencies and automation of PowerMax, IT organizations can modernize their infrastructure to drive massive workload consolidation, shrink hardware footprint, and reduce power consumption. All of which delivers more value to their business while lowering costs and reducing complexity.

Today we are excited to demonstrate how PowerMax can achieve game changing storage performance across virtualized environments. The demo showcases industry first technology and how Dell EMC will help users transition from today to tomorrow while protecting their current storage investments.

The technology of that tomorrow is NVMe and Storage Class Memory (SCM). SCM is cutting-edge storage media considered to be the next generation workhorse (at well below 100 microseconds latency!). It’s perfect for current and emerging performance hungry workloads. Use cases include supporting real-time and data-intensive analytics applications, as well as super-charging transactional OLTP and database workloads.

Today we are demonstrating the industry’s first storage array to fully exploit NVMe-connected SCM drives combined with end-to-end FC-NVMe networks. The result is a future proof combination that will completely shift the technology landscape. Storage decision makers are bullish about NVMe—79% of IT managers surveyed by ESG who were familiar with NVMe technology said they expect it to eventually replace traditional SAS- or SATA-connected solid-state flash storage.

And here’s why the combination of NVMe networks with SCM drives are a game changer.

NVMe over Fabric does lots of cool things to take full advantage of Flash Arrays. It’s designed for extensive parallelism, concurrency and scalability to unlock the power of today’s multi-core CPUs and high-speed media. It provides end-to-end I/O optimizations, massive que depths, and lower latencies from the NVMe host bus adapters (HBAs) to PowerMax NVMe drives. It all adds to delivering more users to more application transactions with lower response times. And because we can leverage SAN networks, we can help protect your infrastructure investments.

Storage Class Memory represents the future of low latency persistent storage media. The performance of SCM is measured in tens of microseconds. It’s wicked fast and approaches the performance levels of DRAM, but at a lower cost and higher density compared to DRAM. The benefit to users is the ability to provide “in memory” like performance with consistent, predictable I/O latencies.

The ability to leverage the benefits of NVMe and SCM requires the right storage architecture. PowerMax is the world’s only Flash Array that includes an innovative, proven real-time machine learning engine that provides intelligent data placement. This intelligence optimizes NVMe and SCM performance without introducing management overhead. PowerMax allows users to transparently leverage both SCM technology for their highest priority, lowest latency workloads combined with cost efficient NVMe flash drives for their less active data sets. This differentiating capability is essential to helping IT organizations increase application performance, reduce storage costs, and minimize management complexity.

In this unprecedented demo with PowerMax that you’ll find at Dell EMC Booth D401, VM datastores reside on that ultra-low latency SCM drive technology and showcases how VMware recognizes PowerMax storage presented over traditional FC versus FC-NVMe, and how a VM can be moved between two datastores created on two different transport technologies, while on the same overall fabric. The ability to simultaneously run I/O over both FC and FC-NVMe on PowerMax is key to helping customers easily transition existing IT infrastructure to more powerful FC-NVMe networks.

Our close friends at VMware believe this demo is pretty special, too. “The Dell EMC PowerMax platform is ideally suited for large-scale, performance-driven VMware environments that will benefit most from the combination of storage class memory and FC-NVMe”, said Lee Caswell, Vice President Products, Storage and Availability Business Unit, VMware.  “We are seeing an insatiable appetite for performance from high-end storage customers who value the deep integration VMware offers for Dell EMC PowerMax storage innovation.”

For everyone who likes to know the nuts and bolts, the PowerMax demo uses the following hardware and software:

◈ Dell EMC PowerMax 2000

◈ PowerMaxOS and Unisphere for PowerMax pre-GA builds

◈ 32Gb FC I/O modules running NVMe over Fabric pre-GA build

◈ 32Gb FC I/O modules running Fibre Channel pre-GA build

◈ 750GB NVMe Storage Class Memory Drives pre-GA build

◈ 8TB NVMe Flash Drives

◈ VMware vSphere ESXi and vCenter pre-GA builds*

* VMware does not promise any such functionality in its future release of vSphere

Both NVMe over Fabric and SCM online upgrades are planned to be available to PowerMax customers in 2019.

Continue reading