[BUG] IVFPQ Benchmarks crash with >=8 threads on Ampere and above

Open
#1,995 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
15/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Stale
Domain
machine-learning, performance

Research direction

The issue does not name any files, tests, or entry points. Start by reproducing the IVFPQ benchmark crash with at least 8 threads on Ampere or newer hardware, then compare behavior with the pool memory resource enabled and disabled. Done means the benchmark no longer segfaults while retaining the pool memory resource.

Written by the indexing model from the issue text.

Description

bug

We are still unsure why this is happening. We notice the segfault goes away when we turn off the pool memory resource, but this is not an acceptable solution as it hinders the performance.

cc @divyegala

Dominant language
Cuda
Stars
1k
Forks
251
Avg merge
1d 8h
Merged PRs (30d)
31

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA/raft

All issues in NVIDIA/raft

Similar issues

More Machine Learning issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.