HPE wins contract for Kestrel supercomputer with US renewable energy lab
The company will support the organisation’s research and development into planning for future energy systems
HPE has won a contract with the US Department of Energy’s National Renewable Energy Laboratory (NREL) to build a new supercomputer to support the organisation’s ongoing research and development into energy.
Dubbed “Kestrel”, in a continuation of NREL's practice of naming systems after various bird species, the supercomputer will be built on the HPE Cray EX high-performance computing (HPC) platform, which is advertised as being able to scale performance on demand and harness insights more efficiently through advanced modelling, simulation, AI, and analytics capabilities.
Kestrel is set to play a critical role in computing across the Department's Office of Energy Efficiency and Renewable Energy research portfolio, advancing research in computational materials, continuum mechanics, and large-scale simulation and planning for future energy systems.
The supercomputer will contain Intel Xeon Scalable processors, Nvidia A100NEXT Tensor Core GPUs to support AI, and HPE Slingshot, an Ethernet fabric built for next-generation supercomputing to address higher speed and congestion control for larger data-intensive and AI workloads. NREL’s Kestrel will also deliver over 75 petabytes of parallel file system storage using the Cray Clusterstor E1000 storage system.
RELATED RESOURCE
Kestrel is set to be completed in 2023 and will have more than five times greater performance than NREL’s existing system, Eagle, with approximately 44 petaflops of peak performance. It will be hosted in NREL’s Energy Systems Integration Facility data centre in Golden, Colorado.
“We look forward to continuing our relationship with NREL and are honoured to have been selected to deliver an advanced supercomputer with Kestrel that will significantly augment the laboratory’s efforts in making breakthrough discoveries of new, affordable energy sources to prepare for a sustainable future,” said Bill Mannel, vice president and general manager of HPC at HPE.
This is HPE’s second supercomputing announcement of the week, as on 1 December it announced it had won a contract with the government of Thailand to build a supercomputer for the country’s National Science and Technology Development Agency. It is set to be 30 times faster than its existing system and expected to be installed and become operational in 2022.
-
ITPro Best of Show NAB 2026 awards now open for entriesThe awards are a fantastic opportunity for companies to stand out at one of the industry's most attended shows
-
Mistral CEO Arthur Mensch thinks 50% of SaaS solutions could be supplanted by AINews Mensch’s comments come amidst rising concerns about the impact of AI on traditional software
-
HPE ProLiant Compute DL345 Gen12 review: A truly EPYC single socket rack serverReviews The big EPYC core count and massive memory capacity make this affordable single-socket rack server ideal for a wide range of enterprise workloads
-
HPE ProLiant Compute DL340 Gen12 review: An appealing alternative to dual-socket Xeon 6 rack serversReviews Big on core density, storage, and expansion, HPE's new DL340 Gen12 rack server offers an energy-efficient Xeon 6 solution at an affordable price
-
On the ground at HPE Discover Barcelona 2025ITPro Podcast This is a pivotal time for HPE, as it heralds its Juniper Networks acquisition and strengthens ties with Nvidia and AMD
-
HPE promises “cross pollinated" future for Aruba and JuniperNews Juniper Networks’ Marvis and LEM capabilities will move to Aruba Central, while client profiling and organizational insights will transfer to Mist
-
HPE ProLiant Compute DL325 Gen12 review: A deceptively small and powerful 1P rack server with a huge core countReviews The DL325 Gen12 delivers a CPU core density and memory capacity normally reserved for expensive, power-hungry dual-socket rack servers
-
HPE's new Cray system is a pocket powerhouseNews Hewlett Packard Enterprise (HPE) has unveiled new HPC storage, liquid cooling, and supercomputing offerings ahead of SC25
-
Inside Isambard-AI: The UK’s most powerful supercomputerLong read Now officially inaugurated, Isambard-AI is intended to revolutionize UK innovation across all areas of scientific research
-
HPE forced to offload Instant On networking division and license Juniper’s AI Ops source code in DOJ settlementNews HPE will be required to make concessions to push the deal through, including divesting its ‘Instant On’ wireless networking division within 180 days.
