ZeroShotMind

Series

GPU and Kernel Optimization

From CUDA thread blocks and shared memory to custom Triton kernels, fused operators, and profiling with nsys — the implementation layer behind fast LLM inference.

Fundamentals
gpucudatritonkernelsinferenceoptimization
  1. 1

    CUDA Fundamentals for ML Engineers

    CUDA exposes GPU parallelism through a three-level thread hierarchy: grid, block, and warp. Understanding how these map to hardware — SMs, register files, shared memory — is the prerequisite for writing fast kernels.

    2024-06-20

  2. 1

    CUDA Fundamentals for ML Engineers

    CUDA exposes GPU parallelism through a three-level thread hierarchy: grid, block, and warp. Understanding how these map to hardware — SMs, register files, shared memory — is the prerequisite for writing fast kernels.

    2024-06-20

  3. 2

    Writing Fused Kernels with Triton

    Kernel fusion eliminates the HBM round-trips between chained operations. Triton makes this practical in Python. This post builds a fused online softmax from scratch, then extends it to a fused RMSNorm + linear projection — the kind of kernel that actually speeds up LLM inference.

    2024-06-20

  4. 2

    Writing Fused Kernels with Triton

    Kernel fusion eliminates the HBM round-trips between chained operations. Triton makes this practical in Python. This post builds a fused online softmax from scratch, then extends it to a fused RMSNorm + linear projection — the kind of kernel that actually speeds up LLM inference.

    2024-06-20