WIP Reduce fetch phase heap consumption #127386
Draft
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
This cuts the peak heap consumption for handling large fetch results by up to 50% by avoiding having all results as well as their serialized for on heap at the same time. A follow up can take this to its logical conclusion and pass the network buffer straight down into the actual fetch execution, avoiding the need to even allocate `SearchHits` in the first place or if taken yet another step further, it could even go as far as avoiding the need to allocate the individual hits in full.
original-brownbear
added
WIP
:Search Foundations/Search
|
#127404 makes it possible to do this a little shorter and nicer looking, lets see if that is an option first before un-drafting this one. |
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Outdated suggestions cannot be applied.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.
WIP because even though this passes tests in its current form, it's a little flimsy :) I'll open a dependency for hardening the
RecyclerBytesStreamtoReleasableBytesReferencetransition to be merged before this one. For now I'd just like to get a CI run out of it :)This cuts the peak heap consumption for handling large fetch results by up to 50% by avoiding having all results as well as their serialized for on heap at the same time. A follow up can take this to its logical conclusion and pass the network buffer straight down into the actual fetch execution, avoiding the need to even allocate
SearchHitsin the first place or if taken yet another step further, it could even go as far as avoiding the need to allocate the individual hits in full.