Improve Distributor Batching / Make Everything Better Forever #1075
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
Signed-off-by: Joe Elliott <number101010@gmail.com>
Signed-off-by: Joe Elliott <number101010@gmail.com>
Signed-off-by: Joe Elliott <number101010@gmail.com>
Signed-off-by: Joe Elliott <number101010@gmail.com>
joe-elliott
requested review from
annanay25,
dgzlopes,
yvrhdn,
mapno and
mdisibio
as code owners
mdisibio
approved these changes
Oct 25, 2021
Signed-off-by: Joe Elliott <number101010@gmail.com>
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Outdated suggestions cannot be applied.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.
What this PR does:
Makes everything better. Like literally it makes everything better. While experimenting with service graph generation in the agent we noticed that certain batching patterns exerted significantly more or less load on Tempo. We traced that down to this code:
tempo/modules/distributor/receiver/shim.go
Line 181 in af34e13
This line is amplifying all writes that come in based on the number of batches in the incoming push. By combining these more intelligently we can get massive performance improvements. This translates to a 40% reduction in CPU usage in our largest cluster (currently consuming 2M spans/second):
Which issue(s) this PR fixes:
Fixes #Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX]