Metrics for incremental bulk splits #116765
Merged
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
|
Pinging @elastic/es-distributed-indexing (Team:Distributed Indexing) |
|
Hi @ankikuma, I've created a changelog YAML for you. |
Tim-Brooks
reviewed
Nov 18, 2024
|
|
||
| metrics.add( | ||
| registry.registerLongAsyncCounter( | ||
| "es.indexing.high_watermark_splits.total", |
There was a problem hiding this comment.
Let's include coordinating in this name since it is at the coordinating level currently.
es.indexing.coordinating.low_watermark_splits.total
es.indexing.coordinating.high_watermark_splits.total
Tim-Brooks
reviewed
Nov 19, 2024
| @@ -160,6 +183,14 @@ public void writeTo(StreamOutput out) throws IOException { | |||
| if (out.getTransportVersion().onOrAfter(TransportVersions.V_8_15_0)) { | |||
| out.writeVLong(totalCoordinatingRequests); | |||
| } | |||
|
|
|||
| if (out.getTransportVersion().onOrAfter(TransportVersions.INDEXING_PRESSURE_THROTTLING_STATS)) { | |||
| out.writeVLong(lowWaterMarkSplits); | |||
There was a problem hiding this comment.
These two branches can be joined.
| @@ -85,6 +91,18 @@ public IndexingPressureStats(StreamInput in) throws IOException { | |||
| } else { | |||
| totalCoordinatingRequests = -1L; | |||
| } | |||
|
|
|||
| if (in.getTransportVersion().onOrAfter(TransportVersions.INDEXING_PRESSURE_THROTTLING_STATS)) { | |||
There was a problem hiding this comment.
These two branches can be joined.
| return (currentUsage >= lowWatermark && size >= lowWatermarkSize) || (currentUsage >= highWatermark && size >= highWatermarkSize); | ||
| if (currentUsage >= highWatermark && size >= highWatermarkSize) { | ||
| highWaterMarkSplits.getAndIncrement(); | ||
| logger.trace(() -> Strings.format("Split bulk due to high watermark: currentUsage [%d] and size [%d]", currentUsage, size)); |
There was a problem hiding this comment.
Perhaps this should either be labelled "current bytes". Or the currentUsage variable should be printed in a bytes representation.
| @@ -65,7 +65,7 @@ protected Settings nodeSettings(int nodeOrdinal, Settings otherSettings) { | |||
| .put(super.nodeSettings(nodeOrdinal, otherSettings)) | |||
| .put(IndexingPressure.SPLIT_BULK_LOW_WATERMARK.getKey(), "512B") | |||
| .put(IndexingPressure.SPLIT_BULK_LOW_WATERMARK_SIZE.getKey(), "2048B") | |||
| .put(IndexingPressure.SPLIT_BULK_HIGH_WATERMARK.getKey(), "2KB") | |||
| .put(IndexingPressure.SPLIT_BULK_HIGH_WATERMARK.getKey(), "4KB") | |||
| .put(IndexingPressure.SPLIT_BULK_HIGH_WATERMARK_SIZE.getKey(), "1024B") | |||
| .build(); | |||
| } | |||
There was a problem hiding this comment.
There are tests in this class testIncrementalBulkLowWatermarkBackOff and testIncrementalBulkHighWatermarkBackOff. Can we modify them to ensure that IndexingPressure indexingPressure properly accounts for the splits? We check indexing pressure stats for other things in this IT so it should be a small addition.
rjernst
pushed a commit
to rjernst/elasticsearch
that referenced
this pull request
Nov 20, 2024
Add metrics to track incremental bulk request splits due to indexing pressure. Resolves ES-9612
ankikuma
added a commit
to ankikuma/elasticsearch
that referenced
this pull request
Nov 21, 2024
Add metrics to track incremental bulk request splits due to indexing pressure. Resolves ES-9612
This was referenced Nov 21, 2024
ankikuma
added a commit
that referenced
this pull request
Nov 21, 2024
alexey-ivanov-es
pushed a commit
to alexey-ivanov-es/elasticsearch
that referenced
this pull request
Nov 28, 2024
Add metrics to track incremental bulk request splits due to indexing pressure. Resolves ES-9612
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.
Adding metrics to track incremental bulk request splits due to indexing pressure. Also added tests to test the metrics.
While writing tests I found that the low and high watermark thresholds in IncrementalBulkIT.java need a small tweak (without this tweak all splits are caused by high watermark thresholds and the low watermark thresholds don't kick in).
Resolves ES-9612