Enter your email above, then click "Sign Up" to join the STAC mail list and (optionally) register to access materials on the site. Click for terms.
NVIDIA publish unaudited STAC-AI inferencing benchmark results for GH200
Multiple records in performance and storage efficiency
Worst pairwise error between 94 ports on 2 devices was 799 picoseconds
Solution sets total of 17 records in scale and baseline benchmarks
Sub-30 nanosecond latency for 68-byte frames
Will head up strategy
Any chance to help is worthwhile.
Second set of STAC-M3 results for the Pavilion Hyperparallel Flash Array.
Read the latest about research, events, and other important news from STAC.
