Reduce the number of Kafka clients used by producers #8088
Merged
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: Marco Pracucci <marco@pracucci.com>
Signed-off-by: Marco Pracucci <marco@pracucci.com>
Signed-off-by: Marco Pracucci <marco@pracucci.com>
pstibrany
approved these changes
May 8, 2024
Signed-off-by: Marco Pracucci <marco@pracucci.com>
pkg/storage/ingest/writer.go
Outdated
| @@ -195,7 +202,7 @@ func (w *Writer) newKafkaWriter(partitionID int32) (*kgo.Client, error) { | |||
| kgo.DefaultProduceTopic(w.kafkaCfg.Topic), | |||
|
|
|||
| // Use a static partitioner because we want to be in control of the partition. | |||
| kgo.RecordPartitioner(newKafkaStaticPartitioner(int(partitionID))), | |||
| kgo.RecordPartitioner(newKafkaStaticPartitioner()), | |||
There was a problem hiding this comment.
kgo has ManualPartitioner() which does the same as the newKafkaStaticPartitioner() now
There was a problem hiding this comment.
You're absolute right. Done in 32c5ba5.
dimitarvdimitrov
approved these changes
May 8, 2024
Signed-off-by: Marco Pracucci <marco@pracucci.com>
Done, thanks! |
Signed-off-by: Marco Pracucci <marco@pracucci.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Outdated suggestions cannot be applied.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
Currently we run 1 Kafka client per partition. This means that if we have 100 partitions, 100 distributors and a metadata refresh interval of 10s, distributors will issue
100 * 100 / 10 = 1000 req/sto update cluster metadata. The cluster metadata is an expensive operation, and can put extra pressure to Kafka. The current approach doesn't scale.In this PR I propose to run 1 Kafka client per producer by default, but allow to increase it via a new config option. When > 1 clients are configured, the
Writerwill simply shard partitions between clients.Which issue(s) this PR fixes or relates to
N/A
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX].about-versioning.mdupdated with experimental features.