A100 study shows high GPU utilization masks low tensor core efficiency

tokenbender · x · 2026-08-19

Relying on nvidia-smi utilization metrics is misleading; it only indicates time with at least one kernel running. A DATE 2024 paper analyzing TensorFlow workloads on an A100 found that while GPU-level utilization was high, the average instruction issue rate was below 50%, and tensor-core instructions were below 5.2%. This highlights why newer architectures like Blackwell add features like TMEM to optimize data movement.

Original post →

More from Infra

Infra channel →