The next era of supercomputing
Competitive advantage in the exascale era relies on the power of data unlocked by supercomputing-powered modeling and simulation and AI. HPE Cray XD and EX supercomputers are designed to deliver exascale technologies to unlock the next frontier of discovery, innovation and achievement.
HPE Cray – Supercomputing for the exascale era
Today’s science, technology, and big data questions are bigger, more complex, and more urgent than ever. Answering them demands an entirely new approach to computing -- and HPE Cray supercomputer are up to the challenge. We’re introducing revolutionary capabilities for revolutionary questions, and HPE Cray systems are the foundational building blocks for scientists and researchers around the world.
Flexibility for today and tomorrow
As your workloads rapidly evolve, the ability to choose your architecture becomes critical. With HPE Cray supercomputing, you can incorporate multiple silicon processing choices—or a heterogeneous mix—with a single management and application development infrastructure. Flex from single to multi-socket nodes, GPUs, FPGAs, and other processing options that may emerge, such as AI-specialized accelerators.
Exascale computing era networking
HPE Slingshot is our high-speed, purpose-built supercomputing interconnect. It features Ethernet capability, advanced adaptive routing, unique congestion control, and sophisticated quality-of-service capabilities. Support for both IP-routed and remote memory operations broaden the range of applications beyond traditional modeling and simulation.
Fast, intelligent storage
HPE’s high performance computing storage is as well suited for the unique storage demands of AI as traditional all-flash enterprise file storage and is as scalable and cost effective as traditional parallel storage.
#1
on the TOP500
Built by HPE, ORNL’s Frontier supercomputer is #1 on the Top500
#1
on HPL-MxP
Built by HPE with Intel, ANL’s Aurora supercomputer is #1 on the HPL-MxP
#2
on the GREEN500
Built by HPE, the University of Bristol’s Isambard-AI phase 1 is #2 on the Green500
Our customers
Performs like a supercomputer, runs like a cloud
The HPE Cray software stack is built on literally decades of supercomputing expertise and is now coupled with a container based model to deliver a new standard in manageability, reliability, availability, resiliency, and interoperability. Seamlessly scale from development on their laptop or in the cloud to production on a supercomputer.
Explore the HPE Cray portfolio
HPE Cray XD systems represent the next-generation purpose-built AI technology. HPE Cray EX supercomputers support multiple processor architectures and accelerator options and are architected for forward compatibility with next-generation blades and servers.
- XD
- EX
HPE Cray XD670
The HPE Cray XD670 is designed and optimized to accelerate performance for large AI model training and tuning. Powered by eight NVIDIA H200 or H100 Tensor Core GPUs, HPE Cray XD670 helps you develop and train larger AI models faster and accelerate time-to-results with unprecedented efficiency, precision, and detailed insights.
HPE Cray XD675
HPE and AMD are accelerating the AI era with high performance computing built for AI workloads. The HPE Cray XD675 is powered by eight AMD Instinct™ MI300X accelerators and a portfolio of advanced software to bring your AI applications to life. Let us help you achieve AI performance at scale.
HPE Cray XD665
The HPE Cray XD665 is designed for diverse HPC and AI workloads and accelerated computing applications. This system includes direct liquid-cooling technology that drives sustainability and efficiency at scale. Our systems can be delivered either on-prem or hybrid cloud to empower every customer with simpler deployment and management, increased cost savings, and faster ROI.
HPE Cray XD2000
HPE is bringing breakthrough capacity and agility to achieve real-world results. The HPE Cray XD2000 is purpose-built for supercomputing and AI workloads, powered by latest generation processors to provide unprecedented performance, high bandwidth memory, and built-in acceleration for the fastest-growing workloads. Our density-optimized systems offer a complete, scalable solution with power and cooling options that delivers maximum performance at a lower TCO. Transform your business with more accessible supercomputing.
HPE Cray EX supercomputer
For highest density and efficiency, liquid-cooled cabinetry supports direct liquid cooling of all components in a highly dense, bladed configuration. These cabinets contain eight chassis, each of which supports up to eight compute and eight switchblades. The orthogonal orientation of compute to switchblades enables a cable-less connection.
HPE Cray EX2500
HPE is bringing the components of our landmark HPE Cray EX supercomputers to the enterprise. The HPE Cray EX2500 is the latest iteration in this robust line of systems and one of our most significant technology advancements in decades. The liquid-cooled design delivers a compact system architecture that enables high efficiency and significant cost savings in OPEX without impacting performance. As power levels increase, it is important for companies to consider how much power a machine will require to run demanding workloads. 100% direct liquid cooling (DLC) provides unparalleled energy savings, so you can benefit from a single integrated infrastructure that reinvents what we know about business and enhances your ability to operate and innovate.
Featured products and solutions
HPE GreenLake, the open and secure edge-to-cloud platform
Complete your solution with a hybrid cloud operating model across all your workloads and data. HPE GreenLake - the cloud operating experience you have always wanted – flexible and scalable, combining security, visibility and ease of management.