cluster experiment details #2
Description
It's written in your paper that “We run all experiments on a cluster with 3 compute nodes and 3 memory nodes”, I'm curious about how to operate cluster experiments using your open source code. Can you give a guidance?