To thrive in the Exascale era, enterprises need a technology solution that delivers exceptional performance today with the capability to support future needs. The HPE Cray XD2000 is the answer.
The term “supercomputing” was coined nearly 100 years ago. In our lifetime, companies have built incredible machines that have transformed what we know about business, humanity, and the world around us. Supercomputers are designed to run impossibly large and complex calculations at breakneck speeds, enabling scientists, researchers, and engineers to solve some of the biggest problems of our time.
But the power of supercomputing means more than deploying massive systems for major corporations and research institutions. Today, breakthrough technologies are enabling greater access to supercomputing power, helping enterprises master the art of the possible.
Compute-intensive applications are pushing the boundaries of traditional infrastructure. The rise of new modeling, simulation, analytics, and AI workflows has put tremendous pressure on enterprises to gain more capacity and more efficiencies from limited IT resources.
To thrive in the age of exascale, enterprises are looking for a type of compute that can check all the boxes. The right solution should deliver exceptional speed, security, and manageability today and be sustainable for the future. For enterprises that are ready to transform, the answer is clear: supercomputing technologies that are accessible at their scale.
Discover supercomputing technologies at your scale
HPE is pioneering a new frontier of supercomputing, providing purpose-built solutions that will transform business operations and accelerate discoveries in areas like climate research, genomics, fraud prevention, and material science. HPE believes this effort will help enterprises leverage supercomputing technologies, whether they have large-scale deployments or are just starting out.
HPE launched its vision by developing the HPE Cray EX supercomputer to address the challenges of explosive datasets and the convergence of supercomputing and AI. These machines are powering the largest and fastest systems in operation, including four of the top 10 supercomputers on the TOP500.
The leading system, Frontier, built for the Department of Energy (DOE) became the first supercomputer to reach exascale at 1.1 exaflops of performance. HPE plans to develop two more HPE Cray EX supercomputers for the DOE over the next two years, culminating in El Capitan, a system capable of 2 exaflops.
Now, HPE is reimagining the exascale system. We have developed an entirely new design merging exascale technologies with an accessible portfolio of solutions that brings supercomputing to the enterprise.
The HPE Cray XD2000 is built from the ground up to handle compute-intensive workloads. It comes with all the components of the HPE Cray EX — including high-speed networking technologies, integrated storage, advanced security features, and an extensive software portfolio to enable faster time to value. Powered by AMD EPYC™ 9004 Series CPUs and AMD Instinct™ accelerators, the HPE Cray XD2000 provides massive performance and efficiency to take on diverse data center challenges. This next-gen system is a complete and scalable solution that fits in standard 19” racks and can scale from one server node to hundreds or thousands of server nodes.
Unlike other supercomputers of its time, the HPE Cray XD2000 offers more flexibility and choice to enable diverse workloads at a broader scale:
- Fabrics: HPE Cray Slingshot, NVIDIA InfiniBand, other standard Ethernet options
- Operating systems: RHEL 9, RHEL 8.6, SLES 15, SLES NEXT, Windows 2022, Windows 2019, VMWare 8.0, VMWare 7.0, Ubuntu 22.04 LTS
- System management: HPE High Performance Cluster Manager, Bright Cluster Manager
- Cooling: Air-cooling, hybrid cooling with ARCS/RDHX, direct liquid cooling (DLC)
- MPI: HPE MPI, Cray MPI, Open MPI, NVIDIA HPC-X
- PCIe cards: Up to 4 LP PCIe cards or 2 LP PCIe cards/2 double-width FHFL PCIe GPUs
- Power supplies: 2, 3, or 4 power supplies to fit different power requirements
- PDUs: 15 metered, switchable, or higher amperage PDUs
Additionally, the system supports industry standard BMC-based on Redfish® server management, so enterprises can simplify and secure their supercomputing environments. These solutions are raising the bar for workload performance by providing all the right tools to achieve better outcomes even faster.
Accelerate business value in the exascale era
HPE and AMD have established leadership in the Exascale era. HPE Cray supercomputing technologies are among our most significant advancements in decades, and the latest system in this industry-leading portfolio is bringing supercomputing capabilities to the enterprise. The HPE Cray XD2000 delivers exceptional compute performance where and how you need to work. This comprehensive and cutting-edge solution has the power to define the next era of science, discovery, and achievement.
HPE and AMD are making supercomputing accessible to more enterprises, and we are committed to pursuing innovation today that will remain relevant for years to come. It’s time to plan your transformation.
Discover your potential in the age of exascale.
Advantage EX Experts
Hewlett Packard Enterprise
twitter.com/hpe_hpc
linkedin.com/showcase/hpe-ai/
hpe.com/info/hpc
AdvEXperts
Our team of Hewlett Packard Enterprise Advantage EX experts helps you dive deep into high performance computing and supercomputing topics.
