5건 · 관련도순
Systematically identify whether a GPU kernel is compute-bound, memory-bound, or latency-bound using roofline analysis, occupancy calculations, compute/load ratio per tile, and SASS instruction inspection. Produces a decision matrix for optimization strategy selection (cp.async, warp interleaving, tiling, double-buffering, or CuAssembler hand-tuning).
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Think and work like an expert Computer Architecture Researcher. Use when a task calls for Computer Architecture Researcher judgment. Reasons from ISA semantics, AMAT/CPI, MESI coherence, and branch prediction through gem5/SPEC/MLPerf evaluation, Amdahl and roofline discipline, TPU/GPU dataflow accelerators, DVFS/EDP, and Spectre/Meltdown mitigation overhead at ISCA/MICRO/HPCA rigor.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when benchmarking, profiling, or A/B-testing CUDA kernels or end-to-end perf in the imp inference engine on RTX 5090 (sm_120), including refreshing tests/perf_baseline.json or publishing numbers to docs/BENCHMARKS.md and the README. Triggers on "benchmark kernel", "profile cuda", "ncu", "nsys", "kernel timing", "kernel sum", "occupancy", "bandwidth bound", "compute bound", "roofline", "perf baseline", "is this regression real", "decode dropped", "aggregate throughput", "two-image A/B", "prefill kernel A/B". Do NOT use for writing/optimizing kernel code (sm120-cuda-expert) or output-quality checks (check-degeneration).
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when profiling with Intel VTune or AMD uProf for hotspots, top-down pipeline stalls, memory-bound analysis, or roofline data. Not for raw perf stat counters: use linux-perf.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when profiling CUDA with Nsight Systems or Nsight Compute, reading roofline and occupancy metrics, or annotating phases with NVTX. Not for correctness: use cuda-debugging.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.