Make numeric segmenter chunk size and head graph configurable (#603)#603
Open
Victor-C-Zhang wants to merge 2 commits into
Open
Make numeric segmenter chunk size and head graph configurable (#603)#603Victor-C-Zhang wants to merge 2 commits into
Victor-C-Zhang wants to merge 2 commits into
Conversation
|
@Victor-C-Zhang has exported this pull request. If you are a Meta employee, you can view the originating Diff in D99729917. |
Summary: Add a segmenter to all XLDB numeric graphs (U8/U16/U32/U64/F32) that auto-chunks input at ~16MB boundaries, aligned to element width. This is implemented inside makeXldbGenericIntCGraphImpl so it applies to all int/float graph variants automatically. Fallback (ZSTD) and Store graphs are not segmented. There is an existing numeric segmenter standard graph, but we can't use it because it doesn't allow us to specify the chunk size or inner graph. The next diff in this stack will add support in the numeric segmenter. When a new OpenZL release is issued, we will migrate to the numeric segmenter. Differential Revision: D99729801
meta-codesync
Bot
changed the title
Victor-C-Zhang
added a commit
to Victor-C-Zhang/openzl-1
that referenced
this pull request
Apr 7, 2026
…ok#603) Summary: Make the numeric segmenter accept local params for the chunk byte size and head graph, replacing the previously hardcoded values. - Chunk size: read via ZL_Segmenter_getLocalIntParam() using ZL_SEGM_NUMERIC_CHUNK_BYTE_SIZE_MAX_PID. Falls back to 16MB default if not provided. - Head graph: read via ZL_Segmenter_getCustomGraphs(). Falls back to ZL_GRAPH_NUMERIC_COMPRESS if no custom graph is provided. Both changes are backward-compatible — existing callers that don't provide these params get identical behavior to before. Differential Revision: D99729917
Victor-C-Zhang
force-pushed
the
export-D99729917
branch
from
2dd7410 to
2c3634c
Compare
…ok#603) Summary: Pull Request resolved: facebook#603 Make the numeric segmenter accept local params for the chunk byte size and head graph, replacing the previously hardcoded values. - Chunk size: read via ZL_Segmenter_getLocalIntParam() using ZL_SEGM_NUMERIC_CHUNK_BYTE_SIZE_MAX_PID. Falls back to 16MB default if not provided. - Head graph: read via ZL_Segmenter_getCustomGraphs(). Falls back to ZL_GRAPH_NUMERIC_COMPRESS if no custom graph is provided. Both changes are backward-compatible — existing callers that don't provide these params get identical behavior to before. Differential Revision: D99729917
Victor-C-Zhang
force-pushed
the
export-D99729917
branch
from
2c3634c to
f84df91
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
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:
Make the numeric segmenter accept local params for the chunk byte size
and head graph, replacing the previously hardcoded values.
ZL_SEGM_NUMERIC_CHUNK_BYTE_SIZE_MAX_PID. Falls back to 16MB default
if not provided.
ZL_GRAPH_NUMERIC_COMPRESS if no custom graph is provided.
Both changes are backward-compatible — existing callers that don't
provide these params get identical behavior to before.
Differential Revision: D99729917