feat(nimble): Add shared integer dictionary encoding (#18637) - #18637
Open
xiaoxmeng wants to merge 1 commit into
Open
feat(nimble): Add shared integer dictionary encoding (#18637)#18637xiaoxmeng wants to merge 1 commit into
xiaoxmeng wants to merge 1 commit into
Conversation
✅ Deploy Preview for meta-velox canceled.
|
meta-cla
Bot
added
the
CLA Signed
Contributor
|
@xiaoxmeng has exported this pull request. If you are a Meta employee, you can view the originating Diff in D112768733. |
Selective Build Plan
Selective build plan |
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 22, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 22, 2026 13:53
f53c14e to
fe4bf01
Compare
meta-codesync
Bot
changed the title
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 00:08
fe4bf01 to
61920ad
Compare
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
duxiao1212
approved these changes
Aug 24, 2026
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 04:20
61920ad to
1c23044
Compare
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 05:12
1c23044 to
3e6fea1
Compare
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 05:17
3e6fea1 to
6da8a5e
Compare
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 06:34
6da8a5e to
69eb876
Compare
xiaoxmeng
added a commit
to xiaoxmeng/velox
that referenced
this pull request
Aug 24, 2026
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 06:39
69eb876 to
4995f40
Compare
…or#18637) Summary: Pull Request resolved: facebookincubator#18637 Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values. Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe. File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented. Reviewed By: tanjialiang Differential Revision: D112768733
xiaoxmeng
force-pushed
the
export-D112768733
branch
from
August 24, 2026 07:08
4995f40 to
12403dc
Compare
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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:
Add shared dictionary encoding for integer Nimble value streams. Encoded chunks store the normal prefix plus dictionary indices, while the alphabet is resolved from a stripe stream, a file catalog, or an external resolver. This lets selected value streams share alphabets even when their physical stream ids vary by stripe, including FlatMap values.
Wire dictionary resolution through tablet readers and bind the resolved alphabet before regular, selective, and index-reader decoding. Writers configure eligible regular columns and flat-map values explicitly, compare shared dictionary encoding with direct encoding for stripe scope on the first chunk, and keep that choice for the rest of the stripe.
File and external dictionaries use prebuilt alphabets and avoid per-chunk scope or dictionary-id payload. Unsupported dictionary lifecycles remain rejected until they are implemented.
Reviewed By: tanjialiang
Differential Revision: D112768733