Performance improvement 3: ngram profiles #203
Draft
Conversation
This file contains 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
msm-code
changed the title
WIP on a very promising approach to query planning Close to a workable solution - ngram profiles Make it production ready Fix the comment location
msm-code
force-pushed
the
fix/performance3-ngram-profiles
branch
from
b962ce7
to
6fe18ec
Compare
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Outdated suggestions cannot be applied.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.
I'm not super happy with this change, but it gives good results.
So what we have here is a heuristic to estimate how many files match for a given ngram. We prefer to start with queries that will return a smaller number of files (because there is a chance that we can "fail fast" and return from a sub query without doing all the work).
So this PR introduces a class called NgramProfile, that stores this information. This is additional 128MB of memory footprint per database, but it shouldn't matter too much in the real world (I hope).
TODO: maybe we should make this optional?
TODO: do we need a way to regenerate the profile?
TODO: measure the real world impact (on cold and warm RAM) and assess the results. It complicates the code significantly, so I think we need at least 25% speedup to consider merging it.