AMD and Nvidia have joined the COVID-19 HPC Consortium, whose members have allocated the resources at their disposal for research related to COVID-19.
The consortium, created at the end of March by IBM, Amazon Web Services, Google Cloud and Microsoft, with the participation of the White House Science and Technology Policy Department, the US Department of Energy and other structures and academic institutions, now has more than 30 supercomputers with a total capacity of 402 PFOLPS. This is less than the Folding @ Home project, which holds a record with 1,500 PFOLPS, but more than the performance of any single supercomputer.
The creators of the consortium report that they received 35 requests from those who need computing resources to fight the virus, and docked them with 19 offers from owners of supercomputers. Moreover, in 11 cases, projects are already working – just two weeks after the idea of a consortium came up. Projects, in particular, are aimed at creating a vaccine and medicine against COVID-19.
Returning to the technical side of the issue, we add that in the systems provided by the consortium participants, there are 3,539,044 CPU cores and 41,286 GPUs.