NVIDIA / NVIDIA/raft

[BUG] Run raft-ann-bench with faiss_cpu_flat algo on Xeon cpu will fail

Open
#2,201 1 comment 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

Describe the bug
it will raise below error on Xeon CPU.

Error occurred running benchmark: Command '['/home/ubuntu/wwq/miniconda3/envs/neuralchat_rag/bin/ann/FAISS_CPU_FLAT_ANN_BENCH', '--build', '--data_prefix=./', '--benchmark_out_format=json', '--benchmark_counters_tabular=true', '--benchmark_out=./wiki_all_88M/result/build/faiss_cpu_flat,base.json.lock', '--raft_log_level=3', 'wiki_all_88M_faiss_cpu_flat,base,k10,bs10000_afc3d9c8-d53d-11ee-af72-0a7d5625b4dd.json']' died with <Signals.SIGKILL: 9>.

Steps/Code to reproduce bug

python -m raft-ann-bench.run --dataset wiki_all_88M --dataset-path ./ --algorithms faiss_cpu_flat --build

Expected behavior
run benchmark succeed

Environment details (please complete the following information):
Bare-metal installation on Ubuntu
Raft was installed by conda install -c rapidsai -c conda-forge raft-ann-bench-cpu

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.

Research direction

Start by running the reported python -m raft-ann-bench.run command with the faiss_cpu_flat algorithm on Ubuntu and inspect where the FAISS_CPU_FLAT_ANN build process receives SIGKILL. The issue is resolved when the wiki_all_88M build benchmark completes successfully on a Xeon CPU.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, ubuntu
Domain
performance, search
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.