lance-format / lance-format/lance
Ensure Datafusion queries are constrained to Datafusion target partitions
@wjones127 is already working on this.
Since Jun 1, 2026.
- Dominant language
- Rust
- Stars
- 7.1k
- Forks
- 852
- Avg merge
- 3d 18h
- Merged PRs (30d)
- 272
Description
We should make sure that the maximum number of CPUs used by a query is dictated by the target number of partitions in DataFusion. This lets a system that wants to use fewer than the number of available CPUs for a query. For example, in a multi-tenant system restricting each query to use 8 vCPUs even if 128 are available.
Some nodes don't use partitions yet; even if they use separate thread pools we should make sure they respect this setting in their max concurrent tasks. Moving those nodes to use partitioning is out-of-scope for this issue.
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.