lance-format / lance-format/lance
Migrate LanceScanExec to native DataFusion partitions
@wjones127 is already working on this.
Since Jun 4, 2026.
- Dominant language
- Rust
- Stars
- 7.1k
- Forks
- 852
- Avg merge
- 3d 18h
- Merged PRs (30d)
- 272
Description
LanceScanExec outputs a single partition and manages I/O parallelism internally via try_buffered. It should be migrated to output N partitions (where N = target_partitions from the session config).
Rather than statically splitting fragments across partitions upfront, each partition should pull from a shared work queue of fragment reads. This avoids stragglers from uneven fragment sizes — a partition that finishes a small fragment quickly picks up the next available one rather than sitting idle.
This follows the same shared-queue pattern already used by FilteredReadExec in MultiplePartitions mode (Arc<AsyncMutex<task_stream>>).
Depends on #7081 (EnforceDistribution) to handle exchange nodes downstream automatically.
Part of the migration tracked in #6967.
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.
Assessment
This issue has not been assessed yet.