Why it matters
Intensity shapes perf regime. Understanding shapes optimization.
Advertisement
The architecture
M x K by K x N: 2 M N K FLOPs + M K + K N + M N bytes.
Intensity high for M+N+K large.
Advertisement
How it works end to end
H100 balance ~300 FLOP/byte bf16.
Large matmul compute-bound.
Small matmul BW-bound.
Batching helps small.