Skip to content

[ML] Increase default queue_capacity to 10_000 and decrease max queue_capacity to 100_000 #115041

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

Merged
merged 5 commits into from
Oct 23, 2024
Merged

[ML] Increase default queue_capacity to 10_000 and decrease max queue_capacity to 100_000 #115041

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
189ff78
Increase default queue capacity and decrease max queue capacity
Oct 17, 2024
9264645
Update docs/changelog/115041.yaml
maxhniebergall Oct 17, 2024
e045cf3
Update tests to match new constants
Oct 18, 2024
1701999
Merge branch 'trainedModelsQueueCapacity' of github.com:elastic/elast…
Oct 18, 2024
4ece775
Merge branch 'main' of github.com:elastic/elasticsearch into trainedM…
Oct 18, 2024
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
6 changes: 6 additions & 0 deletions docs/changelog/115041.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
pr: 115041
summary: Increase default `queue_capacity` to 10_000 and decrease max `queue_capacity`
to 100_000
area: Machine Learning
type: enhancement
issues: []
4 changes: 2 additions & 2 deletions ...c/main/java/org/elasticsearch/xpack/core/ml/action/StartTrainedModelDeploymentAction.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,7 +71,7 @@ public class StartTrainedModelDeploymentAction extends ActionType<CreateTrainedM
public static final AllocationStatus.State DEFAULT_WAITFOR_STATE = AllocationStatus.State.STARTED;
public static final int DEFAULT_NUM_ALLOCATIONS = 1;
public static final int DEFAULT_NUM_THREADS = 1;
public static final int DEFAULT_QUEUE_CAPACITY = 1024;
public static final int DEFAULT_QUEUE_CAPACITY = 10_000;
public static final Priority DEFAULT_PRIORITY = Priority.NORMAL;

public StartTrainedModelDeploymentAction() {
Expand All @@ -89,7 +89,7 @@ public static class Request extends MasterNodeRequest<Request> implements ToXCon
/**
* If the queue is created then we can OOM when we create the queue.
*/
private static final int MAX_QUEUE_CAPACITY = 1_000_000;
private static final int MAX_QUEUE_CAPACITY = 100_000;

public static final ParseField MODEL_ID = new ParseField("model_id");
public static final ParseField DEPLOYMENT_ID = new ParseField("deployment_id");
Expand Down
10 changes: 5 additions & 5 deletions .../java/org/elasticsearch/xpack/core/ml/action/StartTrainedModelDeploymentRequestTests.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ public static Request createRandom() {
request.setNumberOfAllocations(randomIntBetween(1, 8));
}
if (randomBoolean()) {
request.setQueueCapacity(randomIntBetween(1, 1000000));
request.setQueueCapacity(randomIntBetween(1, 100_000));
}
if (randomBoolean()) {
request.setPriority(randomFrom(Priority.values()).toString());
Expand Down Expand Up @@ -168,7 +168,7 @@ public void testValidate_GivenQueueCapacityIsNegative() {

public void testValidate_GivenQueueCapacityIsAtLimit() {
Request request = createRandom();
request.setQueueCapacity(1_000_000);
request.setQueueCapacity(100_000);

ActionRequestValidationException e = request.validate();

Expand All @@ -177,12 +177,12 @@ public void testValidate_GivenQueueCapacityIsAtLimit() {

public void testValidate_GivenQueueCapacityIsOverLimit() {
Request request = createRandom();
request.setQueueCapacity(1_000_001);
request.setQueueCapacity(100_001);

ActionRequestValidationException e = request.validate();

assertThat(e, is(not(nullValue())));
assertThat(e.getMessage(), containsString("[queue_capacity] must be less than 1000000"));
assertThat(e.getMessage(), containsString("[queue_capacity] must be less than 100000"));
}

public void testValidate_GivenTimeoutIsNegative() {
Expand Down Expand Up @@ -234,6 +234,6 @@ public void testDefaults() {
assertThat(request.getNumberOfAllocations(), nullValue());
assertThat(request.computeNumberOfAllocations(), equalTo(1));
assertThat(request.getThreadsPerAllocation(), equalTo(1));
assertThat(request.getQueueCapacity(), equalTo(1024));
assertThat(request.getQueueCapacity(), equalTo(10_000));
}
}
Loading