◆ NeutralMSFTNVDA
MLCommons Releases New MLPerf Storage v3.0 Benchmark Results
GlobeNewswire Inc.·
MLCommons released MLPerf Storage v3.0 benchmark results, introducing new KV Cache and Vector Database tests for AI inference workloads, and adding S3 object storage support alongside POSIX. Nineteen organizations submitted results, including eleven first-time submitters. The benchmark reveals significant variations in power efficiency across storage solutions, with on-premises checkpointing achieving up to 201 GB/second per watt.
Read Full Article at GlobeNewswire Inc. →