Penguin Computing Awarded $68 Million Contracts

Penguin Computing, Inc., a division of SMART Global Holdings, Inc., has been awarded 2 contracts totaling $68 million from the Department of Defense HPC Modernization Program (DoD HPCMP) to deliver its TrueHPC supercomputing platform, plus managed services and storage, to the Navy DSRC and AFRL DSRC sites.

TrueHPC platforms will provide HPC capability for users from all services and agencies of the Department of Defense (DoD). The balanced HPC systems and software enhance the DoD’s ability to tackle the most demanding and computationally challenging problems in fluid dynamics, chemistry and materials science, electromagnetics and acoustics, climate/weather/ocean modeling and simulation, among other applications. Penguin’s managed services team will bring additional capability to the DoD in emerging technologies, while also enabling DoD teams to focus on their research.

Equipped with the latest gen memory and processing technologies, these complete and highly dense HPC resources will be among the most powerful HPCs in the DoD HPCMP’s resources, providing a combined total of over 365,000 cores, more than 775TB of memory, and a total of 47PB of storage including over 5PB of flash storage. Combined, these two systems provide a peak performance of over 17.6 petaFLOPs.

“By implementing Penguin’s TrueHPC solution, the DoD HPCMP user community will be able to conduct advanced research for the highly complex problems the user community is tasked with solving,” said Sid Mair, president, Penguin. “Through our work with NVIDIA and AMD, we’re able to provide the Department of Defense with higher density and operational efficiencies, which equates to exceptional value when you’re using very large scale systems that have extremely high power requirements.“

TrueHPC platforms will be installed at 2 out of 4 HPCMP’s DoD Supercomputing Resource Centers (DSRCs):

• The Navy DSRC at Stennis Space Center in Mississippi will receive a TrueHPC platform with 176,128 compute cores from 3rd Gen AMD EPYC processors and 144 NVIDIA A100 Graphics Processing Units (GPUs). The system is interconnected by an NVIDIA HDR 200Gb IB network and supported by more than 26PB of DataDirect Networks, Inc. (DDN) storage, including over 4PB of high-speed NVMe-based solid-state storage and 370TB of system memory, and will provide 8.5 petaFLOPs of peak performance.
• The Air Force Research Lab’s DSRC at Wright-Patterson Air Force Base in Dayton, OH will receive a TrueHPC platform with 189,440 compute cores from 3rd Gen AMD EPYC processors and 152 NVIDIA A100 GPGPUs. This system is interconnected by an NVIDIA HDR 200Gb IB network and supported by more than 20PB of DDN storage, including over 1PB of high-speed NVMe-based solid-state storage and 405TB of system memory, and will provide 9 petaFLOPs of peak performance.

“Increasingly complex problems being tackled by innovative HPC systems, such as Penguin Computing’s TrueHPC, demand unprecedented acceleration and smarter networks to drive performance at any scale,” said Gilad Shainer, SVP of networking, NVIDIA Corp. “NVIDIA’s accelerated computing platform supports the broadest range of supercomputing workloads, powered by the innovation of NVIDIA A100 Tensor Core GPUs and the unparalleled throughput of NVIDIA HDR IB, the world’s only fully off-loadable, in-network computing platform.“

“We are pleased to continue our longstanding technology partnership with Penguin Computing to push the boundaries of HPC and supercomputing,” said Forrest Norrod, SVP and GM, data center and embedded solutions business group, Advanced Micro Devices, Inc. “The combination of our powerful 3rd Gen AMD EPYC processors and Penguin’s Tundra platform will help the Modernization Office’s user community to solve complex problems to achieve their goal of addressing the Department of Defense’s biggest challenges.“

The TrueHPC systems are scheduled to enter production service early in 2022.