You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
ZonedStrategyBuilder: per-field zone maps, zones cut inside the writer - #10928
Adds ZonedStrategyBuilder, which picks each column's zone map (zone length and aggregates) by field path, and makes ZonedStrategy cut zones itself instead of relying on the 8k RepartitionStrategy above it. This is the first step towards multi-resolution zone maps (#7939): one zone map per column, in the existing vortex.zoned format.
Changes
ZonedStrategyBuilder: with_zones, with_field_zones(path, zones) (the longest matching path wins; an empty list writes no zone map), with_concurrency, build(child, stats).
ZoneOptions { zone_len, aggregate_fns } describes one zone map. Zone lists already take several entries, but build rejects more than one per column until vortex.zoned.v2 exists.
LayoutWriterContext carries the field path, and StructStrategy appends each field's name.
ZonedStrategy cuts each chunk at zone boundaries, aggregates each piece on its own task, and combines the partials of zones that span chunks.
WriteStrategyBuilder::with_zoned. By default the zone length is still the row block size, and the 8k repartition stays.
Breaking:ZonedLayoutOptions and ZonedStrategy::new are replaced by the builder.
Testing
New tests: zones that don't line up with input chunks (smaller, larger, unaligned), per-field overrides (nested, empty), more than one zone map rejected.
Default output is byte-identical to develop: vx convert of TPC-H SF1 customer, orders, lineitem and ClickBench hits_0 gives the same SHA-256 on both builds.
System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.
30 benchmarks were run, but are now archived. If they were deleted in another branch, consider rebasing to remove them from the report. Instead if they were added back, click here to restore them. ↩
This PR has been marked as stale because it has been open for 14 days with no activity. Please comment or remove the stale label if you wish to keep it active, otherwise it will be closed in 7 days
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
changelog/breakA breaking API changestaleThis PR is stale and will be auto-closed soon
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds
ZonedStrategyBuilder, which picks each column's zone map (zone length and aggregates) by field path, and makesZonedStrategycut zones itself instead of relying on the 8kRepartitionStrategyabove it. This is the first step towards multi-resolution zone maps (#7939): one zone map per column, in the existingvortex.zonedformat.Changes
ZonedStrategyBuilder:with_zones,with_field_zones(path, zones)(the longest matching path wins; an empty list writes no zone map),with_concurrency,build(child, stats).ZoneOptions { zone_len, aggregate_fns }describes one zone map. Zone lists already take several entries, butbuildrejects more than one per column untilvortex.zoned.v2exists.LayoutWriterContextcarries the field path, andStructStrategyappends each field's name.ZonedStrategycuts each chunk at zone boundaries, aggregates each piece on its own task, and combines the partials of zones that span chunks.WriteStrategyBuilder::with_zoned. By default the zone length is still the row block size, and the 8k repartition stays.ZonedLayoutOptionsandZonedStrategy::neware replaced by the builder.Testing
vx convertof TPC-H SF1customer,orders,lineitemand ClickBenchhits_0gives the same SHA-256 on both builds.cargo nextest run -p vortex-layout -p vortex-file(886 passed),cargo clippy ... -D warnings,cargo +nightly-2026-09-10 fmt.Follow-up
DictStrategyprobes a fixed row sample (DictStrategyprobes a fixed row sample instead of the first chunk #10927), then drop the 8k repartition aboveZonedStrategy, with benchmarks.vortex.zoned.v2: several zone maps per column.