flâneur

Modern GPU Programming For MLSys — Modern GPU Programming For MLSys

mlc.ai · 547 words · saved by 2 readers

Machine learning systems sit at the heart of modern AI workloads. In these systems, performance often comes down to the quality of a small number of GPU kernels. Attention kernels, LLM prefill and decode kernels, low-precision block-scaled GEMMs, fused MoE layers, and other large fused kernels all directly shape end-to-end speed in both training and serving.

Modern GPU Programming For MLSys# Machine learning systems power many of today’s AI workloads. As models grow and deployment settings become more complex, end-to-end performance increasingly depends on a small number of critical GPU kernels. Attention, LLM prefill and decode, low-precision block-scaled GEMM, fused MoE layers, and other large fused kernels directly affect both training and serving speed. Making these kernels fast requires more than a list of optimization tricks. Recent GPU architectures introduce richer memory spaces, new data-movement mechanisms, and increasingly…

saved by

related reading