MLCommons, the machine learning standards engineering consortium, has produced a machine learning storage benchmark with results from DDN, Nutanix, Weka and others. The benchmark suite measures the performance of storage systems in the context of ML training workloads and has over 28 performance results from five companies. The consortium has more than 50 members including software startups, university researchers, cloud computing and semiconductor giants. David Kanter, executive director of MLCommons, stated that submitting to MLPerf is not trivial and requires real engineering work.
