Meeting AI/ML Compute Needs with the HPE Apollo 6500 Gen10 Plus

August 2, 2021

Companies across most verticals, including manufacturing, financial services, retail and the life sciences, are in need of high-octane compute power. They need solutions that can analyze massive amounts of data and compute like supercomputers do. They need workloads that use high-performance computing (HPC), artificial intelligence (AI), and machine learning (ML) to deliver real-time insights that help their businesses operate more efficiently, respond to changes faster, and innovate.

Hewlett Packard Enterprise (HPE) offers advanced solutions that provide industry-leading performance and flexibility with enterprise reliability. The HPE Apollo 6500 Gen10 Plus Systems deliver exceptional performance, fast insights, and advanced security to fuel better business outcomes. This provides high-density solutions for the most demanding HPC and AI workloads, and utilizes accelerated computing to process data-intensive workloads on GPU architectures.

The HPE Apollo 6500 Gen10 Plus System leverages the processing capacity of the latest NVIDIA A100 Tensor Core GPUs and both 2nd and 3rd Gen AMD EPYC™ processors and AMD Instinct™ MI100 accelerators to empower data-driven innovation and provide choice. The flexible design of the Apollo 6500 Gen10 Plus chassis allows for support of either two 1P HPE ProLiant XL645d Gen10 Plus servers or a single 2P HPE ProLiant XL675d Gen10 Plus server so that the right solution can be designed to fit your unique workload requirement.

The AMD EPYC™ 7003 Series processors provide outstanding performance across the wide spectrum of industry-standard applications. Built on the “Zen 3” core and AMD Infinity Architecture, the AMD EPYC™ 7003 Series processors provide a full feature set across the stack with industry-leading I/O bandwidth, 7nm x86 CPU technology, and embedded security features that help you mitigate risk.  AMD EPYC™ 7003 series CPUs provide up to 32MB of L3 cache per core, 4-6-8 memory channel interleaving designed for better economies and performance in multiple DIMM configurations, plus synchronized clocks between fabric and memory, all driving better, faster time to results.

AMD EPYC™ has been engineered for data centers that rely on CPU performance and throughput. From hyper-converged infrastructure to database to big data analytics to high-performance computing, workloads have more cores to work with. AMD EPYC 7003 Series processors scale from 8 to 64 cores (or 16 to 128 threads per socket). No other x86 server processor achieves this level of core density.[i]

Given the importance of today’s workloads, security is a top concern. AMD Infinity Guard is a suite of advanced security features built into the processor die. It is designed to defend against internal and external threats attacking data and to decrease potential attacks that surface as software is booted and executed.  Featuring Secure Encrypted Virtualization technologies strengthened with Secure Nested Paging, the AMD EPYC processor is intended to help business power breakthrough data security advancements, such as confidential computing. Secure Memory Encryption helps to safeguard data in certain physical attacks without an impact to overall system performance.

AMD Instinct MI100 GPUs, powered by the AMD CDNA™ architecture and AMD Infinity Fabric™ technologies, provide industry-leading double precision computational performance for HPC with up to 11.5 TFLOPS peak FP64 performance for your most demanding workloads. The MI100 can support your largest data sets and help eliminate traditional bottlenecks in data movement with 32 Gb HBM2 memory and an ultra-high ~1.2 TB/s of memory bandwidth.

The MI100 accelerator also brings all-new Matrix Core Technology supporting a full range of mixed precisions for AI & Machine Learning with superior performance for whatever combination of workloads you need to deploy in your data center. In addition, AMD Instinct™ GPUs are supported by the hyperscale class ROCm™ open software platform bringing developers and customers choice and an open stack for programing innovation.

NVIDIA A100 GPUs ensure low latency at high throughput to enhance these powerful accelerated computing solutions. In addition, with third-generation Tensor Cores, the NVIDIA A100 can efficiently scale up to thousands of GPUs, or, with NVIDIA Multi-Instance GPU (MIG) technology, can be divided into seven isolated GPU instances to accelerate diverse workloads.

HPE systems leverage CPU-GPU heterogeneous compute for a broad range of mission-critical HPC/AI applications. The HPE Apollo 6500 Gen10 Plus System maximizes GPU utilization with up to 16 GPUs per server to rapidly capture, analyze, and operationalize intelligence, regardless of the HPC and AI workload requirements. NVIDIA NVLink establishes a seamless connection between GPUs to work together as a single robust accelerator. NVLink interconnects provide dedicated communication, which enables memory to migrate from GPU to GPU. A single A100 supports up to 12 NVLink connections for a total bandwidth of 600 gigabytes per second.

Enter MLPerf

Businesses need a way to compare solutions for their mixed workloads. MLPerf Inference is an emerging benchmark that has garnered wide industry approval to asset solution performance. Developed by MLPerf.org, a group comprised of 50+ technology and academic leaders, the MLPerf Inference is a machine learning inference performance benchmark suite that measures how fast systems can process inputs and produce results using a trained model.

Two HPE Apollo 6500 Gen10 Plus Systems were recently tested in a quarterly industry roundup coordinated by MLPerf.org. The tests evaluated performance for common AI/ML tasks, including image classification, object detection, speech-to-text, natural language processing, recommendations, and more. The HPE Apollo 6500 Gen10 Plus with the HPE ProLiant XL645d Gen10 Plus Server outperformed all other similarly-equipped systems across all categories.

Contributing to the results, 2nd Gen AMD EPYC™ processors offer tremendous bandwidth and a high core count to continuously feed data-hungry GPUs. These high-frequency processors integrated with NIVIDIA Mellanox HDR InfiniBand add up to 200 gigabits per second of bandwidth for every two GPUs, so even businesses operating at the cluster level can communicate at twice the speed. HPE’s first optimized 4-GPU server, is a single AMD EPYC processor system which delivers better price-performance than ever for HPC. Also, with a refreshed 8-GPU offering supporting up to two AMD EPYC processors per server, enterprises can harness up to 16 PCIe GPUs, more than twice as many as HPE has supported in the past.

In addition, the HPE Apollo 6500 Gen10 Plus System offers extensive storage options, with up to 16 front-accessible storage devices, SAS/SATA solid-state drives, and up to six NVMe drives to access data faster while consuming less power. HPE has plans to roll out new solutions through H1 2021, which will feature a staggering 16 NVMe drives for almost 6X greater bandwidth. Now, HPE customers can harness the robust power, frequency, and processing capacity of accelerated computing to drastically reduce time-to-insight.

Selecting a Technology Partner

There are a growing number of choices in ways to increase HPC and AI/ML capabilities. Today’s systems are designed to be optimized for specific workflows – regardless of varying needs. To make full use of the available technology, users should look for a technology partner that offers complete solutions and services.

HPE has a strong track record of delivering top-performing systems for business and scientific computing and provides a comprehensive solution including:

  • A full software stack including a programming environment, systems management, and workflow management tools
  • Support services to avoid downtime, ensure availability, and maintain performance
  • Applications engineering help with deep industry expertise and support
  • Financial flexibility to deploy HPC and AI/ML processing where you need it and how you consume it.
  • Building solutions based on the best available technology like AMD EPYC processors and AMD Instinct accelerators for faster time to value.

As technology leaders in HPC, together HPE and AMD can deliver real-world expertise for the most demanding applications.

HPE has a long and rich history in providing leading-edge systems for the HPC market. Over the last few years, the company has made strategic acquisitions that allowed it to expand its offerings and have access to innovative technologies that have since been integrated into its product lines. Notable deals included the acquisitions of Silicon Graphics Inc. (SGI) and Cray Computing, both known for their capabilities in high-performance computing.

Additional factors to consider when evaluating solutions

When looking for an HPC solution for today’s most demanding workloads, raw processing power alone is not enough. Many other factors should be considered when evaluating a solution. And in fact, many of these additional factors that should go into an evaluation are areas where HPE excels. Some of these include:

HPE Pointnext Services: HPE Pointnext Services advises businesses on IT strategy to transform hybrid cloud environments, take advantage of AI, and innovate at the edge. IT experts design and build solutions with the latest software-defined technology and help businesses operate and optimize their environments across hybrid cloud and traditional IT. This offloads the tasks of running IT from internal staff, so IT can be consumed as a service. HPE Pointnext Services also fills the missing skills gap that many companies have when implementing new technologies. Pointnext services include:

  • Cloud services to bring agility and manageability to your technology environment.
  • Edge services to achieve better insights and automation at the edge.
  • Modernization services to transform your IT and data center with automation and container technologies.
  • Data-driven services help you streamline AI adoption and migrate to unified data platforms.

HPE Finance Services: HPE Finance Services offers helps accelerate the adoption of needed technology. There is IT financing for SMBs so they can get access to the technology they need to support their business. IT asset lifecycle solutions from HPE Financial Services help companies extract additional value from their existing IT investments or move to new technology more effectively. And IT investment solutions help companies get the technology they need with the budget they have today. The programs make it easy to structure expenses in the way that makes the most sense for a business.

Security and management: The HPE Silicon Root of Trust secures the Gen10 portfolio with firmware protection, malware detection, and firmware recovery down to the silicon. This unrivaled technology ensures that HPE servers will never boot with compromised firmware, making them “The World’s Most Secure Industry Standard Servers.” HPE iLO server management software enables customers to securely configure, monitor, and update HPE systems seamlessly from anywhere in the world, so customers can operate with confidence. AMD Infinity Guard adds layers of protection that help secure your data and mitigate risk.

HPE GreenLake: HPE GreenLake for HPC is a consumption-based payment model that aligns cash to actual usage. With HPE GreenLake, businesses can select from a wide range of pre-configured services, such as containers, VMs, storage, compute, data protection, or SAP HANA. The solution can scale with ease with capacity that is actively monitored, managed, and proactively deployed when needed. And since it relies on expertise from HPE’s IT Operations Centers, it frees up staff to work on other projects.

In conclusion

The new HPE Apollo 6500 Gen10 Plus systems accelerate performance to take on the most complex HPC and AI workloads. The purpose-built platform built on AMD EPYC processors provides unprecedented compute power with industry-leading CPUs, GPUs, fast interconnects, high-bandwidth fabric, and configurable topology.

Simply put, the HPE Apollo 6500 Gen10 System with AMD EPYC processors and AMD Instinct accelerators deliver exceptional performance, faster insights, and advanced security to take your business to the next level. In addition, this platform is engineered to be resilient, so customers can operate with confidence, regardless of their HPC and AI requirements. This accelerated computing system is changing the game for customers around the world by offering:

  • High-density computing for data-intensive tasks
  • Rock-solid availability and reliability at the enterprise-level
  • Easy deployment and manageability
  • Flexibility to meet data center requirements
  • Optimized design for increased simplicity and faster time to value

HPC and AI innovation is revolutionizing business operations. Visit us online to learn how HPE and AMD can help you realize your goals.

Would you like to apply for our HPE Insiders for HPC client-only community? It is a dedicated invitation-only space to keep you informed, give you your seat at the table and have some fun along the way with your peers!!


[i] AMD EPYC has up to 64c/128t and Intel Scalable has up to 56c/112t, per processor. EPYC-10

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance computing and the advanced-scale AI market. Early customers Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerated for AI applications. Now, Amazon Web Services (AWS) is int Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testbed (AQT), which is based at Lawrence Berkeley National Labor Read more…

Graphcore Introduces Larger-Than-Ever IPU-Based Pods

October 22, 2021

After launching its second-generation intelligence processing units (IPUs) in 2020, four years after emerging from stealth, Graphcore is now boosting its product line with its largest commercially-available IPU-based sys Read more…

Quantum Chemistry Project to Be Among the First on EuroHPC’s LUMI System

October 22, 2021

Finland’s CSC has just installed the first module of LUMI, a 550-peak petaflops system supported by the European Union’s EuroHPC Joint Undertaking. While LUMI -- pictured in the header -- isn’t slated to complete i Read more…

AWS Solution Channel

Royalty-free stock illustration ID: 577238446

Putting bitrates into perspective

Recently, we talked about the advances NICE DCV has made to push pixels from cloud-hosted desktops or applications over the internet even more efficiently than before. Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerate Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testb Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the powe Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

US Closes in on Exascale: Frontier Installation Is Underway

September 29, 2021

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, held by Zoom this week (Sept. 29-30), it was revealed that the Frontier supercomputer is currently being installed at Oak Ridge National Laboratory in Oak Ridge, Tenn. The staff at the Oak Ridge Leadership... Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Leading Solution Providers

Contributors

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make i Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire