quickwit-oss / quickwit-oss/quickwit
Add some random prefix to split ids.
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 11.7k
- Forks
- 597
- Avg merge
- 2d 22h
- Merged PRs (30d)
- 37
Description
S3 creates partitions based on a prefix of the object keys.
Because we use ULID, splits coming from the same day end up on the same servers.
On the search side, most of the queries tend to target the last few days, and end up hitting the same servers.
For this reason, in some case, we can hit the per-prefix rate limiting.
Contributor guide
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
No files or tests are named. Start by tracing how split IDs become S3 object-key prefixes and how search requests locate recent splits; determine how a random prefix would be generated and interpreted. Done means recent data is distributed across prefixes without breaking split lookup or search behavior, with coverage for the resulting placement and lookup.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- aws, rust
- Domain
- backend, cloud, distributed-systems
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100