One Cluster to Rule Them All

This presentation addresses the following topic(s):

• Machine learning and HPC: Each worthy of its own investment, or better together? (Alternately: Marriage for the ages, or heading for divorce?)

Machine Learning, especially Deep Learning and HPC, needs tremendous amounts of resources. Many institutes and companies feel tempted to combine the budgets for both workloads into one large cluster.

While GPU accelerated HPC and ML workflows share many similar computational constraints, there are also fundamental differences in their computational needs.

This talk will look at the full scale of cluster deployment from the hardware to the software stack and aims to answer if there is the "one" GPU cluster.