High Performance Computing

Does your enterprise need to simplify management, reduce costs, and improve reliability and performance for high-performance computing (HPC) and AI workloads?

Built for the exascale era, the HPE Apollo 6500 Gen10 Plus System accelerates performance with NVIDIA® HGX A100 Tensor Core GPUs and AMD Instinct™ MI100 with Infinity Fabric™ accelerators to take on some of the most complex HPC and AI workloads. This purpose-built platform provides enhanced performance with premier graphics processing units (GPU), fast GPU interconnect, high-bandwidth fabric, and configurable GPU topology, providing rock-solid reliability, availability, and serviceability (RAS). Configure with single or dual processor options for a better balance of processor cores, memory, and I/O. Improve system flexibility with support for 4, 8, 10, or 16 GPUs and a broad selection of operating systems and options, all within a customized design to reduce costs, improve reliability, and provide leading serviceability.

Do you need a density-optimized, scalable system to meet the changing demands of your digital transformation journey? The HPE Apollo 2000 Gen10 Plus System is a shared infrastructure chassis with flexible support for up to 4 ProLiant XL225n Gen10+ servers (AMD) or up to 4 ProLiant XL220n Gen10+ servers (Intel®) or 2 XL290n Gen10+ servers (Intel), helping you increase your rack space density. Server nodes can be serviced without impacting operation of other nodes in the same chassis for increased server up-time. It delivers the flexibility to tailor the system to the precise needs of your demanding high-performance computing (HPC) workloads with the right compute, flexible I/O, and storage options. The system can be deployed with a single server, leaving room to scale as customer's needs grow, bringing the power of supercomputing to data centers of any size. It is ideal for HPC applications in industry verticals like manufacturing, oil and gas, life sciences, and financial services.

Have Cloud or IT-as-a-Service (ITaaS) environments forced you to settle for Tier-1 storage that compromises your data center? HPE 3PAR StoreServ 20000 Storage is an enterprise flash array with more than 3.8M IOPS, sub-millisecond latencies, a 16x density advantage over the competition¹, and scalability of over 20 PB of usable capacity for massive consolidation – for block and file workloads as well as object access – with Quality of Service (QoS). The flash-optimized architecture features the HPE 3PAR Gen5 ASIC for silicon-based hardware acceleration, including inline deduplication, compression, data packing, thin technologies and other compaction technologies that can reduce acquisition and operational costs without compromising performance. Configuration options are available for all-flash models and combined flash and HDD models. HPE 3PAR StoreServ 20000 also includes at no extra charge data protection with HPE Recovery Manager Central.

The HPE Performance Cluster Manager software is a fully integrated system management solution for all HPE high performance computing (HPC) clusters and supercomputers.
The software offers fast system setup from bare-metal, comprehensive hardware monitoring and management, image management and software updates as well as power management and more.
HPE Performance Cluster Manager reduces the time and resources you need to spend administering your systems, keeps them resilient and running as close to maximum efficiency as possible so you can achieve better return on your hardware investments.

Are your High Performance Compute (HPC) and AI environments struggling with file management?
HPE Data Management Framework 7 (DMF7) delivers centralized data management across HPC and AI storage systems and protects scalable, parallel file systems like Lustre and Spectrum Scale. Namespace reflection is used to create an independent snapshot of file system state, allowing you to recover file systems in a known good state. This system maintains file versions, allowing users to recover files from previous successful job runs.

HPE DMF7 automates data movement between tiers in a storage hierarchy, e.g. between flash and disk. Administrators and users can also use HPE DMF7 to move files between file systems, e.g. when files must be moved from storage that is being retired. HPE DMF7 improves utilization of expensive, high performance storage by automatically moving files to lower cost storage tiers, creating a virtual storage space that appears to scale beyond the physical capacity.

Is your storage slowing down your HPC compute cluster?

The Cray ClusterStor E1000 Storage System is purpose-engineered to meet the demanding input/output requirements of supercomputers and HPC clusters in a very efficient way. The E1000 parallel storage solution typically achieves the given HPC storage performance requirements, significantly reducing the number of storage drives. That means HPC users with a fixed budget for the HPC system can spend more of their budget on CPU/GPU compute nodes, accelerating time-to-insight. The E1000 Storage System embeds the open-source parallel file system Lustre to deliver this efficient performance. Hewlett Packard Enterprise provides enterprise-grade customer support in-house for Lustre that scales out (nearly) linearly, without software licensing for the file storage systems per terabyte capacity or per storage drive. This allows customers to reap the benefits of the open-source movement while getting enterprise-grade support.

Are you looking to equip your enterprise environment with the performance, reliability, and security needed for the most demanding workloads?

The HPE Superdome Flex 280 is a highly reliable server that starts at two and scales up to eight 3rd generation Intel® Xeon® Scalable processors. Its modular architecture scales cost-efficiently to meet future growth. Six UPI links per processor result in higher bandwidth and faster data rates than prior generations.¹ Designed to provide 64 GB to 24 TB of shared memory using DRAM or in combination with persistent memory, it is an ideal choice for real-time analytics. Extreme HPE Superdome RAS features such as advanced memory resiliency, firmware-first approach, analysis engine, and self-healing provide increased system uptime. Superior security, including support for Silicon Root of Trust, protects your critical workloads. As-a-service consumption with HPE GreenLake provides flexibility while maintaining on-premises control.

Is your HPC networking solution able to meet your converged workload needs for today and tomorrow?

HPE Slingshot provides a modern, high-performance interconnect for HPC and AI clusters that delivers high-bandwidth and low-latency for HPC, ML, and analytics applications by bringing together the specialized requirements of HPC-optimized fabrics with the ubiquity of Ethernet. This delivers a converged infrastructure with high-performance on both HPC simulation codes and native IP applications, with efficient scalable access to data sources.

Building on Cray's specialized silicon, HPE Slingshot delivers consistent performance and low latency under load and at scale. This prepares you to efficiently service increasingly diverse users taking advantage of your HPC resources, and do so without overprovisioning bandwidth or deploying multiple systems to avoid congestion on your most demanding workloads.

Do you need an operating system specifically designed to run demanding HPC applications?
HPE Cray Operating System is a suite of high-performance software designed to run large, complex applications at scale.
Based on standard SUSE Enterprise Linux, the software provides many features designed to improve application efficiency, reliability, management, and extend data access.

How do you enable a system to perform like a supercomputer, but run like a cloud?

HPE Cray System Management for HPE Cray supercomputers is a solution enabling system administrators to manage large-scale supercomputers leveraging the architecture and advances of hyper-scalers and cloud providers. While offering the familiar capabilities of high performance computing (HPC) system management software, HPE Cray System Management enables customers to go beyond the traditional and enable new services, deploy broad ranges of workloads, and drive towards the as-a-service experience. Built to manage systems scaling to Exascale, HPE Cray System Management offers everything needed for manageability, reliability, and interoperability for your HPE Cray supercomputers.