How to think about Hybrid architectures for HPC

Brendan Bouffler
Date : June 22, 2023
Categories : AWS ParallelCluster ,Amazon NICE DCV ,Elastic Fabric Adapter ,Life Sciences
Tags : cpus ,dcv ,ec2 ,efa ,gpus ,hpc ,high performance computing ,lustre ,mpi ,parallel cluster ,schedulers ,storage ,autoscaling ,bioinformatics ,cloud computing ,elastic ,elastic fabric adapter ,hyrbrid ,infiniband ,scientific computing ,technical computing ,tightly coupled ,virtualization ,vizualization ,techshorts

If you’re pondering a a hybrid HPC setup between your on-prem HPC cluster and the cloud, it’s worth exploring how well you understand you local users, daily workflows, and the … rhythm of your environment. Just a few minutes of thinking about this will help you make some crazy good decisions that’ll save you huge amounts of effort trying to solve problems that, frankly, you probably don’t need to solve.

Brian Skjerven is an expert at simplifying this challenge. And so we’re glad we could get the time with him to discuss this just after ISC finished in Hamburg. We recorded some talks in the lobby of our hotel, before we fled to the airport.

If you have ideas for technical topics you’d like to see us cover in a future show, let us know by finding us on Twitter (@TechHpc) and DM’ing us with your idea.