| name | gpullama-ncu-analysis |
| description | Nsight Compute workflow for analyzing one specific GPULlama3/TornadoVM CUDA kernel. Use after nsys or TornadoVM profiler output identifies the hot kernel.
|
| tags | ["gpullama","ncu","kernel-profiling"] |
| license | Apache-2.0 |
GPULlama3 Nsight Compute Analysis
When To Use
Use ncu only for a specific hot CUDA kernel identified by nsys, TornadoVM profiler output, or generated-kernel investigation.
Do not run broad ncu collection over an entire decode loop without a kernel target.
Workflow
- Identify the target kernel name or regex.
- Use a focused GPULlama3 command that launches the kernel repeatedly.
- Skip warmup/JIT launches.
- Start with SpeedOfLight.
- Add targeted sections based on the first classification.
Start with:
ncu --section SpeedOfLight --csv --kernel-name regex:"<kernel>" --launch-skip <N> --launch-count <M> -- <command>
Escalate:
- memory-bound:
MemoryWorkloadAnalysis
- compute-bound:
ComputeWorkloadAnalysis
- latency/occupancy issue:
LaunchStats, Occupancy
- scheduler/warp issue:
WarpStateStats, SchedulerStats
Classification
Use measured SOL metrics:
- high memory throughput, low compute throughput: memory-bound
- high compute throughput, lower memory throughput: compute-bound
- both low with short kernels or low occupancy: latency/occupancy-bound
For the target hot kernel, pay special attention to:
- DRAM throughput
- L1/L2 behavior
- achieved occupancy
- register count
- shared/local memory usage
- global load width and instruction count
Report
Include:
- target kernel regex
- command
- raw relevant
ncu output or exported metrics path
- SOL classification
- occupancy/register/shared-memory observations
- recommended kernel or TornadoVM codegen change
Do not present invented or estimated metrics.
