Conversation
|
Review the following changes in direct dependencies. Learn more about Socket for GitHub.
|
tamirms
added this pull request to stack #1044
September 25, 2026 19:41
Each cold events read opened its chunk's index.hash by reading the whole file into memory. It now memory-maps the file instead. The pages come from the kernel's page cache, which every reader of the chunk shares, so an open costs a map plus the pages its lookups touch, rather than a copy of a file that grows with the chunk's term count. roaring moves from v2.18.2 to v2.26.0, which adds vectorized container kernels. Its serialized form is roaring's portable format, so existing artifacts read the same. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QHnF5BhsuoxpWxGmatQzmt
tamirms
force-pushed
the
tamirms/roaring-2.26-mphf-mmap
branch
from
September 25, 2026 20:35
ac4db2a to
8816035
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
index.hash(streamhash.Open) instead of reading the whole file into memory (os.ReadFile+streamhash.OpenBytes).Why
A cold events read opens its chunk's index for every request. Reading it whole copies a file whose size grows with the chunk's term count, about 18.5 MB per request at high term cardinality. Mapped, the pages come from the kernel's page cache, which every reader of the chunk shares, so a request pays only for the pages its lookups touch.
roaring v2.26.0 adds vectorized container kernels and drops a clone from its bulk AND and OR, which the match engine in the next PR runs on every slab. Its serialized form is roaring's portable format, so existing artifacts read the same.
This is the first of four PRs split out of #968 and
tamirms/cold-index-parts, stacked on #1034.Known limitations
N/A
🤖 Generated with Claude Code