quickwit-oss / quickwit-oss/quickwit

Multilang tokenizer is slow on long text in JPN/CMN/KOR

Open
#3,684 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Rust
Stars
11.7k
Forks
597
Avg merge
2d 22h
Merged PRs (30d)
37

Description

For short/medium texts, throughput is around a few MB/s, but for long Wikipedia articles, it goes under 1MB/s.

Contributor guide

Open the contributing guide

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

The issue names no files, tests, or entry points. Start by reproducing throughput on long Wikipedia text in Japanese, Chinese, and Korean, then trace the multilang tokenizer path. Done means the long-text throughput bottleneck is addressed and the reported performance is verified without regressing short or medium texts.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
performance, search
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.