The next era of compute is here, and every industry is preparing to digitally transform. HPE and AMD are empowering organizations to unleash higher compute performance than ever before with purpose-built solutions tuned to modernize your data center.

Comprehensive technologies, tools, and advisory services from HPE enable supercomputing capabilities from edge to core to cloud. We combine groundbreaking HPE systems with the extreme processing capacity of AMD EPYC™ CPUs and AMD Instinct™ GPUs to create an environment today that will remain resilient for years.

Building a robust supercomputing infrastructure

As your workloads rapidly evolve, your architecture becomes critical. The HPE Cray supercomputing line, powered by AMD processors, enables you to address today’s supercomputing challenges. These powerful systems are the firepower of computing environments, allowing you to streamline workflows from modeling and simulation to analytics and AI. By combining the best compute, interconnect, storage, and software these systems deliver reliable and exceptional performance for all your applications to deliver insights and innovation on a scale that has never been seen before.

For increased acceleration, build your business on the world’s highest performing server processor[1] and the world’s fastest HPC GPU.[2] AMD EPYC CPUs and AMD Instinct GPUs are designed together to speed compute workloads, data-intensive models, and analysis techniques. Used by some of the world’s fastest, most scalable data centers and supercomputers, AMD processors can help generate rapid innovation and results. With a flexible processor portfolio with consistent features across the stack, you can choose the right processor to match your workload. In addition, AMD Instinct GPUs help accelerate your discoveries with nearly 3.5x the boost for HPC and nearly 7x boost for AI, compared with prior-gen AMD accelerators.[3] This powerful combination represents a major advancement in the speed and flexibility of computing. A key component of these industry-defining systems is HPE Slingshot, the latest version of scalable interconnect that provides a high-performance network. HPE Slingshot is the new interconnect that allows HPE Cray systems to bridge the worlds of supercomputing, cloud, and the data center with state-of-the-art small packet performance and scalability.

Cray ClusterStor E1000 high-performance storage is well-suited for the unique demands of your supercomputing infrastructure. This new breed of parallel storage is engineered to overcome your biggest data center challenges. Cray ClusterStor E1000 is a balanced, zero bottleneck design that rips through any roadblocks. You can gain unprecedented throughput for breakneck time to insight.

HPE offers an extensive software portfolio along with source offerings, so you can choose the right mix of these solutions to fit your exact needs. The HPE Cray software stack includes system development, system management, and data management tools to turn even the most complex hardware into your ideal solution. Coupled with a container-based model, the software stack has a proven track record of helping organizations simplify deployments to get results faster.

AMD works with a broad set of technology partners and open source developers to build a robust software ecosystem. Our efforts include extensive contributions to operating system and virtualization environments, compilers libraries, tools, runtimes, and more. These engagements have produced more than 100 solutions across a wide spectrum of business needs. This allows for a more powerful user experience. Delivering performance and scalability for key workloads provides choice and innovation with optimized solutions and fast time to value.

Entering the future of compute

HPE Pointnext Services offer modern solutions to help you transform, all while training your workforce and identifying tools that will help you to excel in supercomputing. Our experts work with you to deliver on your vision to automate, optimize, and scale your operations.

You can consume your ideal solution as fully managed, pay-per-use services from HPE GreenLake. HPE aims to democratize the use of supercomputing by eliminating the concerns of system and operational costs and limited technology expertise through consumption-based options. New HPE GreenLake services for HPC removes the complexity and cost associated with supercomputing deployments by delivering fully managed, pre-bundled services that come in small, medium, or large options.

HPE and AMD are committed to helping you maximize on supercomputing innovation. Our comprehensive solutions give you the tools and expertise to accelerate your transformation, based on the same feature sets as the largest supercomputing systems.

It’s time to free your resources from day-to-day heavy lifting, so you can stay ahead of what’s next. Let’s get started.

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 provide the opportunity to connect with your peers.

---

[1] Results as of 04/14/2021 using SPECrate®2017_int_base. The AMD EPYC 7763 scored 839, http://www.spec.org/cpu2017/results/res2021q1/cpu2017-20210219-24936.html which is higher than all other 2P scores published on the SPEC® website. SPEC®, SPECrate® and SPEC CPU® are registered trademarks of the Standard Performance Evaluation Corporation. See www.spec.org for more information.

[2] Calculations conducted by AMD Performance Labs as of Sep 18, 2020 for the AMD Instinct™ MI100 (32GB HBM2 PCIe® card) accelerator at 1,502 MHz peak boost engine clock resulted in 11.54 TFLOPS peak double precision (FP64), 46.1 TFLOPS peak single precision matrix (FP32), 23.1 TFLOPS peak single precision (FP32), 184.6 TFLOPS peak half precision (FP16) peak theoretical, floating-point performance. Published results on the NVidia Ampere A100 (40GB) GPU accelerator resulted in 9.7 TFLOPS peak double precision (FP64). 19.5 TFLOPS peak single precision (FP32), 78 TFLOPS peak half precision (FP16) theoretical, floating-point performance. Server manufacturers may vary configuration offerings yielding different results. MI100-03

[3] Calculations performed by AMD Performance Labs as of Sep 18, 2020 for the AMD Instinct™ MI100 accelerator at 1,502 MHz peak boost engine clock resulted in 184.57 TFLOPS peak theoretical half precision (FP16) and 46.14 TFLOPS peak theoretical single precision (FP32) matrix floating-point performance. The results calculated for Radeon Instinct™ MI50 GPU at 1,725 MHz peak engine clock resulted in 26.5 TFLOPS peak theoretical half precision (FP16) and 13.25 TFLOPS peak theoretical single precision (FP32) matrix floating-point performance. Server manufacturers may vary configuration offerings yielding different results. MI100-04