Elijah's Notes

gpu

9 items with this tag.

  • Jul 31, 2026

    Serving Systems

    • llm-serving
    • inference
    • gpu
    • batching
  • Aug 01, 2026

    From SIMD to SIMT: Vectors, GPUs, and Accelerators

    • simd
    • simt
    • vectors
    • risc-v
    • gpu
    • accelerators
    • systolic-arrays
    • microarchitecture
  • Aug 01, 2026

    GPU Architecture from First Principles

    • gpu
    • architecture
    • simt
    • cuda
    • system-verilog
    • computer-architecture
    • digital-design
  • Aug 01, 2026

    GPU Interconnects and Collective Communication

    • gpu
    • interconnect
    • nvlink
    • nccl
    • collective-communication
    • rdma
    • parallelism
    • machine-learning
  • May 25, 2025

    GPU Architecture and Programming

    • gpu
    • architecture
    • gpu-programming
    • cuda
    • nvidia
    • pytorch
    • triton
  • May 25, 2025

    Optimizing GPU Kernels

    • gpu
    • kernel
    • optimization
    • cuda
    • triton
  • May 10, 2025

    Modeling and Scaling Performance with Roofline

    • roofline
    • performance
    • optimization
    • gpu
    • bandwidth
    • compute-bound
    • arithmetic-intensity
    • gqa
    • kv-cache
  • Apr 02, 2025

    How to write a fast kernel

    • cuda
    • gpu
    • pytorch
    • kernel
  • Apr 02, 2025

    GPU Kernel Programming with Triton and CUDA

    • gpu
    • triton
    • cuda

Created with Quartz v5.0.0 © 2026

  • GitHub