Stop compromising. Learn how the new HPE Cray Storage Systems C500 bring the value of leadership class storage to entry-level HPC compute clusters.
Since its 2019 launch, the Cray ClusterStor E1000 Storage System has become the de-facto external parallel file system standard for exascale1, pre-exascale2, and national AI supercomputers3 as it allows those leadership supercomputing sites with HPE Cray EX supercomputers to:
Take control of their HPC storage spending, enabled by the embedded open-source Lustre® file system without capacity-based licensing and a cost-effective hybrid architecture.
Remove I/O bottlenecks holding their HPC initiatives back, enabled by the very strong read and write performance of the file system.
Enjoy unified support from the HPE Services team for the whole HPC infrastructure stack and sidestep the “finger pointing” characteristic of third-party storage vendors during complex support cases.
Feel secure in the fact that their valuable data sits in a file system that is owned by a vibrant community and not by a single company.
But not everyone needs leadership-class systems built on HPE Cray EX architecture. As the #1 HPC server vendor globally4, HPE also serves many customers running modeling and simulation (MOD/SIM) workloads on entry-level HPC compute clusters built from HPE Cray XD2000 systems.
These customers still need and want the proven leadership-class storage technologies of Cray ClusterStor E1000 Storage Systems. But they need it at a significantly lower entry price point that fits the size of their compute clusters that are running converged MOD/SIM workloads.
In response, we are announcing HPE Cray Storage Systems C500 which provide those leadership-class storage technologies at a fraction of the entry-price point and with increased ease-of-use.
Discover leadership-class storage at a fraction of the price that is easier to deploy and manage
HPE Cray Storage Systems C500 runs the same software as our high-end Cray ClusterStor E1000 Storage Systems and leverages the same 2U24 storage controllers and 5U84 HDD enclosures. They just do it in a converged and more cost-effective design. Here’s the difference:
Where the E1000 uses a 2U24 storage controller for the System Management Unit (SMU) the C500 uses a more cost-effective HPE ProLiant DL325 Gen11 server
Where the E1000 uses a 2U24 storage controller with 24 NVMe SSD as Metadata Unit (MDU) and one 2U24 storage controller with 24 NVMe SSD as Scalable Storage Unit Flash (SSU-F), the C500 converges both into just one 2U24 converged MDU/SSU-F with 24 NVMe SSD
Where the E1000 only supports fully populated storage enclosures, the C500 supports also half-populated storage enclosures in specific configurations.
The following image shows the entry configuration of HPE Cray Storage Systems C500 which provides between 22 TB and 513 TB usable capacity from 24 NVMe SSDs and can drive up to 80 GB/s aggregate read performance and up to 60 GB/s write performance to the compute nodes.
Additional 2U24 NVMe units or 5U84 HDD enclosures can be added to expand the usable capacity of the file system to 2.6 PB all flash capacity or to 4 PB hybrid (SSD/HDD) capacity5. Beyond those capacity requirements, Cray ClusterStor E1000 Storage Systems are the best way to deploy open source, parallel file systems with enterprise-grade support from the HPE Services Team.
In addition, HPE is rolling out major software improvements and new functionalities that make the storage systems easier to deploy and easier to manage.
HPE Cray Storage Systems C500 bring the value proposition of Cray ClusterStor E1000 Storage Systems to entry-level HPC compute clusters built on HPE Cray XD2000 systems.
With HPE Cray Storage Systems C500 you can:
Take control of your HPC storage spending, enabled by the embedded open-source Lustre® file system without capacity-based licensing and a cost-effective hybrid architecture.
Remove I/O bottlenecks holding your HPC initiatives back, enabled by the very strong read and write performance of the file system.
Enjoy unified support from the HPE Services team for your whole HPC infrastructure stack and sidestep the “finger pointing” characteristic of third-party storage vendors during complex support cases.
Feel secure in the fact that your valuable data sits in a file system that is owned by a vibrant community.
The HPE Cray Storage Systems C500 answers the call for fast and cost-effective parallel storage for entry-level compute clusters running MOD/SIM workloads. And we’re ready to answer yours! Learn more at the HPC storage webpage or request a deep-dive briefing via your HPE account manager.
Meet Uli Plechschmidt, Worldwide Product Marketing Manager for Parallel HPC & AI Storage
Uli leads the product marketing function for high performance computing (HPC) storage. He joined HPE in January 2020 as part of the Cray acquisition. Prior to Cray, Uli held leadership roles in marketing, sales enablement, and sales at Seagate, Brocade Communications, and IBM.
1 Examples for that category are Frontier, Aurora, El Capitan
2 Examples for that category are LUMI, Perlmutter, Adastra
3 Examples for that category are Isambard-AI, Alps, Shaheen III
4 Intersect360 Research, Machine Learning in HPC, September 2023
5 Hyperion Research, HPE Market Update, April 2024
6 Testing limitation, not an architectural limitation