Section 8 of 8
How models actually run on hardware: GPUs, memory bandwidth and the roofline, kernels, LLM inference and serving, multi-GPU parallelism, capacity planning.
2 items in this section.
Why the same GPU is nineteen times faster than the CPU at one operation, eleven times at the next, and slower at a third — worked out on a real Tesla T4, from two numbers per machine: how fast it computes and how fast it moves bytes. Along the way: warps and SIMT, latency hiding, the memory hierarchy, arithmetic intensity, the roofline, and a power limit that moves the ceiling.
Most of the arithmetic in a neural network is matrix multiplication. Written the obvious way it is correct, and fifty times slower than torch.mm on the same GPU. Where that difference comes from, one change at a time on a Tesla T4: coalescing, tiles, shared memory, registers, tensor cores and precision — and why almost all of it is about where the numbers are kept rather than how they are multiplied.