Hammerspace MLPerf® Storage v1.0 Benchmark Results

The MLCommons MLPerf Storage benchmark is intended to demonstrate the performance of various storage systems for simulated machine learning workloads, so that technical buyers and decision makers have some criteria when evaluating storage system performance for machine learning, deep learning, and other forms of GPU computing.

This year, Hammerspace submitted results for the MLPerf v1.0 Storage Benchmark for the first time. This technical brief summarizes the results of that testing, including:

• Background on MLCommons and the MLPerf Storage Benchmark
• A summary of Hammerspace’s results relative to other vendors, including the test setups used for the benchmark
• A discussion on the advantages of Hammerspace standards-based parallel file system architecture compared to scale-out NAS and HPC parallel file systems

The results prove the price/performance advantage of Hammerspace for high-throughput, low-latency file and object storage, both on-prem and on-cloud.