HPE today introduced two new supercomputing and rack-scale innovations to help customers improve storage economics and enhance cooling efficiencies. New hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 create the world’s fastest hybrid supercomputing storage1, and the HPE Rack Management Controller improves cooling and power of high-density racks.

“When making long-term investments in environments where advanced workloads are rapidly evolving, our customers want scalable, dense, and flexible systems to help them address dynamic shifts in requirements,” said Trish Damkroger, senior vice president and general manager, HPC and AI Infrastructure Solutions, HPE. “Our comprehensive portfolio, which now includes a world-leading hybrid supercomputing storage option, delivers powerful capabilities to scale, while maximizing capacity and optimizing costs.”

HPE delivers the world’s fastest hybrid supercomputing storage

As demand for data-intensive workloads increases, prices for large solid-state drives (SSDs) have increased. To meet the needs of research institutions and national laboratories, HPE is combining flash performance with high-capacity hard disk drives (HDD) to deliver new hybrid configurations that enable customers to spend less on storage and redirect infrastructure investment toward compute node purchases, while delivering up to 41% more data throughput per storage rack2.

The new hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 combine SSDs with new 32-terabyte SAS-4 HDD and new SAS-4 HDD enclosures, supporting 74 or 100 drives.

Purpose-engineered for supercomputing, the new configurations deliver:

  • Improved storage economics: Customers can match storage performance to processing requirements, while spending 82% less than they would on a comparable all-flash system3.

  • More capacity, same data center footprint: Creating up to 32% more usable storage capacity per rack4, helping customers support growing datasets without expanding rack space.

  • High read-and-write performance: Delivering up to 116 GB per second of read throughput and 92 GB per second of write throughput per rack, enabling storage to better match the needs of HPC.

New intelligent rack management for HPC and AI infrastructure growth

As HPC and AI infrastructure pushes power densities to new levels, cooling efficiency and operational visibility play a critical role in supporting customers’ goals for growth. In response, the HPE Rack Management Controller is designed to enable intelligent liquid cooling through real-time monitoring, leak detection, automated control and rack-level visibility. Engineered and informed by HPE’s five decades of liquid cooling innovation, the controller helps organizations optimize cooling efficiency and facilitate the growth of high-density HPC and AI infrastructure.

The controller supports platform-specific design variants tailored to the requirements of different rack-scale systems. HPE Rack Management Controller will initially support HPE Cray Supercomputing GX5000 systems, with future extensions for AMD and NVIDIA rack-scale systems.

The HPE Rack Management Controller integrates with the HPE Supercomputing Management Software suite, including Foundation for infrastructure management, Composable for resource sharing, and Observability for operational monitoring. The integration shares rack-level telemetry and control capabilities, improving visibility and making it easier to operate and expand large, liquid-cooled HPC and AI environments.

HPE continues to advance the infrastructure required for the world’s most demanding scientific, commercial, and HPC workloads. Through its proven expertise across supercomputing, storage, networking, cooling, and systems management, HPE delivers integrated platforms that help customers scale performance, improve efficiency, and accelerate innovation.

Availability

  • Availability for the new hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 is planned for early 2027.

  • HPE Rack Management Controller for HPE Cray Supercomputing GX systems and the Supercomputing Management Software suite will be available in March 2027. Support for AMD and NVIDIA rack-scale systems will follow in June 2027.

About HPE

HPE (NYSE: HPE) is a leader in essential enterprise technology, bringing together the power of AI, cloud, and networking to help organizations achieve more. As pioneers of possibility, our innovation and expertise advance the way people live and work. We empower our customers across industries to optimize operational performance, transform data into foresight, and maximize their impact. Unlock your boldest ambitions with HPE. Discover more at www.hpe.com.

____________________________

1 The new hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 enable customers to get up to 41% faster data throughput in gigabyte per second (GB/sec) per storage rack compared to the next hybrid supercomputing storage competitor (DDN).

2 The new hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 enable customers to get up to 41% faster data throughput in gigabyte per second (GB/sec) per storage rack compared to the next hybrid supercomputing storage competitor (DDN).

3 The new hybrid configurations of HPE Cray Supercomputing Storage Systems E2000 enable customers to get the same capacity for up to 82% lower acquisition cost compared to the next all flash supercomputing storage competitor. The analysis compared the street price of 15 petabytes of usable storage capacity that meets the performance specifications of NVIDIA for a cluster of 1,024 GPUs.

4 As compared to the currently shipping hybrid HPE Cray Supercomputing Storage Systems E2000

Media gallery

About The Author
Business Wire