MLPerf
The industry-standard benchmark suite for AI training and inference performance; the closest thing to apples-to-apples numbers across accelerator vendors.
Current numbers
Unet3D ≥90%, RetinaNet ≥85%Minimum accelerator utilization for a valid MLPerf Storage v3.0 training result, by workload
~every 30 minFailure interval in MLPerf Storage's illustrative model of 100k accelerators at full utilization