blockstore: replace single held-open reader with parallel span reads for content retrieval#683
Open
parkan wants to merge 1 commit into
Open
blockstore: replace single held-open reader with parallel span reads for content retrieval#683parkan wants to merge 1 commit into
parkan wants to merge 1 commit into
Conversation
file-backed blocks are now fetched as spans of consecutive blocks via ranged reads into a bounded LRU cache, with sequential access detection prefetching the next spans in parallel. removes the global mutex and the single active stream that serialized all /ipfs traffic. concurrent backend reads are capped by a semaphore (connection budget), duplicate span fetches collapse via singleflight, and the handler map gets a proper RWMutex. callers detach from in-flight fetches on cancellation (the fetch completes into the cache), and each backend read is bounded by a timeout so a hung backend cannot pin connection slots. adds (file_id, file_offset) composite index for the span query.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
this does real span-based reads with approximately 14x throughput improvement (median case) over the naive mutex implementation and solves the memory amplification on random small reads into large spans by reading the relevant data directly into a buffer; in theory this should make an arbitrary rclone-compatible backend into a proper trustless gateway to the data while maintaining full compatibility with normal filecoin onboarding