Optimized #consumeAsync() and couple other small things
#452
Conversation
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
`PullAllResponseHandler` keeps all result records in a queue. It used to always allocate this queue even for queries that return no results. This resulted in unnecessary garbage generation for write-only queries that return nothing. This commit makes `PullAllResponseHandler` start with an empty unmodifiable queue and allocate a real one when first record arrives.
String values from `ServerVersion#toString()` and `BoltServerAddress#toString()` are used to create result summary for every query. Thus it is reasonable to memorize them and not recompute every time.
`LinkedHashMap` has default load factor of 0.75. It is often not enough to specify desired initial capacity in order for the map to not resize. This commit adds a helper method to create `LinkedHashMap` that will not resize.
Each message consists of one or more chunks. Each chunk is prefixed with a fixed-size header which contains two bytes - chunk body size. Messages are separated by a "message boundary" which is a two byte message "00". Code used to keep message boundary and empty chunk header as shared `ByteBuf` constants. These buffers were duplicated (no content copying involved) for every operation. This resulted in creation of multiple lightweight but unnecessary wrapper objects. This commit makes code write both message boundary and empty temporary chunk header directly into IO buffers and removes the need to duplicate shared buffers. So no temporary objects are created.
Previously it was implemented using `#forEachAsync()` and a no-op consumer. This resulted in every record being added to the queue of incoming records, then wrapped in a `CompletableFuture` and only then thrown away. So it did a lot more work than needed. This commit optimizes `#consumeAsync()` and `StatementResult#consume()` that builds on top of it. Now all incoming values will not be transformed into records and will be dropped right away. So no queueing and `CompletableFuture`s will be involved.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.