lance-format / lance-format/lance
Spill IVF shuffle/partition state through SpillStore
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 7.1k
- Forks
- 852
- Avg merge
- 3d 18h
- Merged PRs (30d)
- 272
Description
Wire the IVF shuffle (shuffle_data.lance / shuffle_offsets.lance) and per-partition state to SpillStore (#7300). Add a codec for shuffle runs. Shuffle output is single-use scratch (consumed once, partition-by-partition, never re-read), so a streaming spill contract is sufficient.
Part of #7302.
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
Start by reading the SpillStore work in #7300 and the broader context in #7302, then trace the IVF shuffle state associated with shuffle_data.lance and shuffle_offsets.lance. Identify how per-partition state and shuffle runs are represented before defining the codec and streaming spill path. Done means the shuffle data and partition state flow through SpillStore and can be consumed once partition by partition.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- data-engineering
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100