How Nvidia built Selene, the world's seventh-fastest computer, in three weeks


Five years ago, Nvidia set out to design a supercomputer-class system powerful enough to train and run its own AI models, such as models for autonomous vehicles, but flexible enough to serve just about any deep-learning researcher. After building multiple iterations of its DGX Pods, Nvidia learned valuable lessons about building a system with modular, scalable units. The COVID-19 outbreak brought new challenges for Nvidia, as it set out to build Selene, the fourth generation of its DGX SuperPODs. Reduced staff and building restrictions complicated their task, but Nvidia managed to go from bare racks in the data center to a fully operational system in just three and-a-half weeks. Selene is now a Top 10 supercomputer, the fastest industrial system in the US and the fastest MLPerf machine that's commercially available.

Duplicate Docs Excel Report

None found

Similar Docs  Excel Report  more

None found