Skip to content

[8.x] Fix scale up for model allocations #115216

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 1 commit into from
Oct 21, 2024
+49 −1
Merged

[8.x] Fix scale up for model allocations #115216

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion .../plugin/ml/src/main/java/org/elasticsearch/xpack/ml/autoscaling/MlAutoscalingContext.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,7 +177,7 @@ public boolean isEmpty() {
return anomalyDetectionTasks.isEmpty()
&& snapshotUpgradeTasks.isEmpty()
&& dataframeAnalyticsTasks.isEmpty()
&& modelAssignments.values().stream().allMatch(assignment -> assignment.totalTargetAllocations() == 0);
&& modelAssignments.values().stream().allMatch(assignment -> assignment.getTaskParams().getNumberOfAllocations() == 0);
}

public List<String> findPartiallyAllocatedModels() {
Expand Down
48 changes: 48 additions & 0 deletions ...rc/test/java/org/elasticsearch/xpack/ml/autoscaling/MlAutoscalingDeciderServiceTests.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,6 +54,7 @@
import static org.elasticsearch.xpack.ml.utils.NativeMemoryCalculator.STATIC_JVM_UPPER_THRESHOLD;
import static org.hamcrest.Matchers.containsString;
import static org.hamcrest.Matchers.equalTo;
import static org.hamcrest.Matchers.greaterThan;
import static org.hamcrest.Matchers.is;
import static org.hamcrest.Matchers.nullValue;
import static org.mockito.ArgumentMatchers.any;
Expand Down Expand Up @@ -331,6 +332,53 @@ public void testScale_GivenModelWithZeroAllocations() {
assertThat(result.requiredCapacity().node().memory().getBytes(), equalTo(0L));
}

public void testScale_GivenTrainedModelAllocationAndNoMlNode() {
MlAutoscalingDeciderService service = buildService();
service.onMaster();

ClusterState clusterState = new ClusterState.Builder(new ClusterName("cluster")).metadata(
Metadata.builder()
.putCustom(
TrainedModelAssignmentMetadata.NAME,
new TrainedModelAssignmentMetadata(
Map.of(
"model",
TrainedModelAssignment.Builder.empty(
new StartTrainedModelDeploymentAction.TaskParams(
"model",
"model-deployment",
400,
1,
2,
100,
null,
Priority.NORMAL,
0L,
0L
),
new AdaptiveAllocationsSettings(true, 0, 4)
).setAssignmentState(AssignmentState.STARTING).build()
)
)
)
.build()
).build();

AutoscalingDeciderResult result = service.scale(
Settings.EMPTY,
new DeciderContext(
clusterState,
new AutoscalingCapacity(AutoscalingCapacity.AutoscalingResources.ZERO, AutoscalingCapacity.AutoscalingResources.ZERO)
)
);

assertThat(result.reason().summary(), containsString("requesting scale up"));
assertThat(result.requiredCapacity().total().memory().getBytes(), greaterThan(TEST_JOB_SIZE));
assertThat(result.requiredCapacity().total().processors().count(), equalTo(2.0));
assertThat(result.requiredCapacity().node().memory().getBytes(), greaterThan(TEST_JOB_SIZE));
assertThat(result.requiredCapacity().node().processors().count(), equalTo(2.0));
}

private DiscoveryNode buildNode(String id, ByteSizeValue machineMemory, int allocatedProcessors) {
return DiscoveryNodeUtils.create(
id,
Expand Down
Loading