The MLPerf consortium has released the results for MLPerf Training v0.7 machine learning training performance benchmark suite. And NVIDIA has come up tops by breaking 16 AI performance records, doubling the number of records it smashed in 2019.
The A100 Tensor Core GPU delivered the fastest performance per accelerator on all eight MLPerf benchmarks.
The DGX SuperPOD system, a massive cluster of DGX A100 systems connected with HDR InfiniBand, also set eight new performance milestones for overall fastest time to solution at scale.
This is the third consecutive and strongest showing for NVIDIA in training tests from MLPerf, an industry benchmarking group formed in May 2018.
MLPerf comprises more than 70 companies and researchers from leading universities. Its MLPerf benchmark suites are the industry standard for measuring machine learning performance.
NVIDIA set six records in the first MLPerf training benchmarks in December 2018 and eight in July 2019.