perf-nsight-compute-analysis

Official

Profile CUDA kernels with Nsight Compute.

AuthorNVIDIA
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Analyze Nsight Compute profiling outputs to diagnose CUDA kernel bottlenecks.

Core Features & Use Cases

  • SOL% based bottleneck classification to identify compute-bound, memory-bound, or latency-bound kernels.
  • Roofline and memory-hierarchy analysis to guide optimization priorities.
  • Section-guided deep dives (SpeedOfLight, ComputeWorkloadAnalysis, MemoryWorkloadAnalysis) on .ncu-rep reports.
  • Reproducible guidance for profiling workflows across Nsight Compute versions and GPU architectures.

Quick Start

Profile a kernel with Nsight Compute to quickly identify bottlenecks and recommended next steps.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: perf-nsight-compute-analysis
Download link: https://github.com/NVIDIA/skills/archive/main.zip#perf-nsight-compute-analysis

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.