Extend nvbench to measure SOL for compute-bound workloads
Open
Nobody has claimed this yet.
- Dominant language
- Cuda
- Stars
- 927
- Forks
- 123
- Avg merge
- 2d 10h
- Merged PRs (30d)
- 2
Description
Existing nvbench allows to measure SOL for memory bound workloads by providing
state.addGlobalMemoryReads(nbytes)
state.addGlobalMemoryWrites(nbytes)
It would be useful to extend this concept to provide flops such that we can measure how close workload is to the compute roofline.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading nvbench's existing state.addGlobalMemoryReads and state.addGlobalMemoryWrites APIs and the code that reports their measurements. Determine how a FLOP metric should be exposed for compute-bound workloads and how it should support roofline comparisons. Done means nvbench can measure and report FLOPs for such workloads, with coverage for the new behavior.
Written by the indexing model from the issue text.
Assessment
- Domain
- performance, tooling
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100