Search
Homepage This page's url is: -crn- Rankings and Research Companies Channelcast Marketing Matters CRNtv Events WOTC Jobs HPE Zone Intel Partner Connect Digital Newsroom Dell Technologies World Newsroom Dell Technologies Newsroom HP Reinvent 2020 Newsroom IBM Newsroom The IoT Integrator Lenovo Newsroom NetApp Data Fabric Intel Tech Provider Zone

AMD, Penguin Computing Supply HPC Systems For COVID-19 Research

'AMD and our technology partners are proud to provide researchers around the world with these new systems that will increase the computing capability available to fight COVID-19 and support future medical research,' AMD CEO Lisa Su says of the new deals to supply supercomputers to three U.S. universities.

AMD and Penguin Computing have reached new deals to supply supercomputers to three U.S. universities while also setting up a new cloud-based system for COVID-19 research.

Penguin Computing said Monday it will supply high-performance computing systems equipped with AMD's second-generation EPYC processors and Radeon Instinct MI50 GPUs to New York University, the Massachusetts Institute of Technology and Rice University in Houston, Texas. AMD is also contributing a system for its cloud HPC service, Penguin Computing On Demand.

[Related: How Penguin Computing Is Fighting COVID-19 With Hybrid HPC]

With each university system capable of performing more than one petaflop and Penguin Computing's new cloud system capable of four petaflops, these are the first systems to be funded by AMD's $15 million COVID-19 HPC fund, which kicked off in April with a deal to supply hundreds of Radeon Instinct MI50 GPUs for the U.S. Department of Energy's Corona supercomputer.

"High performance computing technology plays a critical role in modern viral research, deepening our understanding of how specific viruses work and ultimately accelerating the development of potential therapeutics and vaccines," AMD CEO Lisa Su said in a statement. "AMD and our technology partners are proud to provide researchers around the world with these new systems that will increase the computing capability available to fight COVID-19 and support future medical research."

HPC is seen as a critical tool in accelerating the discovery of drugs and vaccines for COVID-19, as demonstrated by the formation of the White House-led COVID-19 High Performance Computing Consortium, which counts AMD and Nvidia as well as OEMs and cloud service providers like Hewlett Packard Enterprise and Microsoft as members. The effort is also receiving support from Folding@Home, a distributed computing application that lets anyone with a PC or server contribute.

New York University said it will use its new HPC system for discovering drugs that could target COVID-19 and future SARS virus mutations, analyzing medical imaging for patient screening, retrieving research results from a vast library of biomedical literature and "analyzing political attitudes and voting behavior in response to financial hardships."

MIT, on the other hand, will use its new system for modeling and testing as well as the discovery of vaccines. Rice University's Center for Theoretical Biological Physics will use its system to focus its research on the coronavirus' surface proteins, in understanding how they enter human cells and identifying drug molecules that could bind effectively to them.

The chipmaker is soliciting proposals for COVID-19 research from researchers who want remote access to the new AMD-based system available on the Penguin Computing On Demand cloud service.

AMD said the donated systems are equipped with G290-Z21 compute nodes supplied by Gigabyte, which each contain a single 48-core EPYC 7642 processor and eight Radeon Instinct MI50 GPUs. The system also includes R182-291 management nodes from Gigabyte, which use two 16-core EPYC 7302 processors each. The systems are also using Mellanox HDR 200-Gbps InfiniBand interconnects supplied by Nvidia, which recently acquired Mellanox.

In a recent interview with CRN, Penguin Computing President Sid Mair said the company's strategy of using both on-premise and cloud servers allows researchers to offload jobs into the cloud when there isn't enough resources to deploy new on-premise servers.

They can't upgrade quick enough in order to continue to do their research, so being able to walk in and move their workflow over into an HPC environment that works and acts and implements just like they would do it on-premise but they're doing it in the cloud is becoming very, very beneficial to our researchers," Mair said in April.

Back to Top

Video

 

sponsored resources