Release 0.6.4 to main (exists→get lookup reuse + L3 prefetch counters) - #23
Merged
Conversation
SGLang's HiCache prefetch resolves a prefix in two steps -- batch_exists() to find the hit length, then batch_get() to pull those pages -- which issued two directory RPCs for the same keys. batch_exists() now fetches full locations and primes the resident hit prefix into a one-shot handoff cache; the imminent batch_get() consumes (pops) them, skipping the second lookup. Primes are popped on use and short-TTL (1s), so a consumed location is only ever as stale as the exists->get gap -- always on, never serving a location older than one prefetch handoff (unlike the opt-in directory_read_cache_ttl). Adds a directory_lookups_saved counter and a contract test asserting the get reuses primed locations and re-resolves once they are consumed. Co-authored-by: 兰州小红鸡 <flyphp@outlook.com>
read_requests==0 alone can't distinguish 'SGLang never probed L3' from 'probed but didn't fetch'. Add exists_requests (batch_exists calls) and exists_pages_found (pages reported present) so the prefetch path is observable end to end: - exists_requests==0 -> prefetch never reached the backend - exists_pages_found==0 -> probing but directory miss (key/shard) - found>0 but read_requests==0 -> found, not fetched (local hit/policy) Co-authored-by: 兰州小红鸡 <flyphp@outlook.com>
…doff-49da perf+obs(store): reuse exists→get directory lookup; count L3 prefetch probes (0.6.4)
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Lands the 0.6.4 work on
main. PR #22 was merged into this branch (cursor/discovery-poll-logging-49da) rather thanmain, somainis still at 0.6.3 (which itself was never tagged). This PR fast-forwardsmainto the v0.6.4 release commit.Commits over
main:perf(store): reuse directory lookup across exists→get—batch_existsprimes resident hit locations into a one-shot, short-TTL handoff cache thatbatch_getconsumes, skipping the redundant second directory RPC. Newdirectory_lookups_savedcounter.obs(store): count batch_exists probes + pages found—exists_requests/exists_pages_foundcounters make the SGLang L3 prefetch path observable end to end (distinguishes "never probed" vs "probed but didn't fetch").Version → 0.6.4 (already tagged
v0.6.4; the GitHub Release + PyPI publish run from the tag).