Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Default max concurrent search req. numNodes * 5 #31171

Merged
merged 4 commits into from
Jun 8, 2018
Merged

Default max concurrent search req. numNodes * 5 #31171

Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1c47301
Default max concurrent search req. to the avg shards size per index
s1monw Jun 7, 2018
aafae7a
fix devision by zero
s1monw Jun 7, 2018
62ee2d1
move to a constant
s1monw Jun 7, 2018
d63d3c8
Merge branch 'master' into fix_max_con_search_request
s1monw Jun 7, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 8 additions & 2 deletions server/src/main/java/org/elasticsearch/action/search/TransportSearchAction.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -340,13 +340,19 @@ private void executeSearch(SearchTask task, SearchTimeProvider timeProvider, Sea
}
return searchTransportService.getConnection(clusterName, discoveryNode);
};
if (searchRequest.isMaxConcurrentShardRequestsSet() == false) {
if (searchRequest.isMaxConcurrentShardRequestsSet() == false && shardIterators.size() > 0) {
/*
* We try to set a default of max concurrent shard requests based on the node count but upper-bound it by 256 by default to keep
* it sane. A single search request that fans out to lots of shards should not hit a cluster too hard while 256 is already a
* lot.
* Yet we are still trying to guarantee some concurrency ie. if you have an index size of N shards and you are searching a
* single index on a single node we still search it concurrently.
*/
searchRequest.setMaxConcurrentShardRequests(Math.min(256, nodeCount));
final int numShards = shardIterators.size();
final int numIndices = remoteClusterIndices.size() + concreteIndices.length;
assert numShards >= numIndices : "found schroedingers index numShards: " + numShards + " vs. numIndices: " + numIndices;
assert numIndices > 0;
searchRequest.setMaxConcurrentShardRequests(Math.min(256, nodeCount * (numShards / numIndices)));
}
boolean preFilterSearchShards = shouldPreFilterSearchShards(searchRequest, shardIterators);
searchAsyncAction(task, searchRequest, shardIterators, timeProvider, connectionLookup, clusterState.version(),
Expand Down