AWS opensearch-service documentation change
Summary
Added documentation for new Optimized engine metrics including NativeMemoryPressure, NativeRuntimeResidentMemory, NativeSearchRuntimeCPUUtilization, ThreadpoolNativeSearchCPUQueue, and ThreadpoolNativeSearchCPUThreads. Also noted that certain OpenSearch Dashboards metrics don't apply to Optimized domains.
Security assessment
The change introduces new monitoring metrics for performance optimization but contains no evidence of addressing security vulnerabilities or weaknesses. Metrics focus on resource utilization and query execution, not security controls.
Diff
diff --git a/opensearch-service/latest/developerguide/managedomains-cloudwatchmetrics.md b/opensearch-service/latest/developerguide/managedomains-cloudwatchmetrics.md index a9fa185a5..6bce298e1 100644 --- a//opensearch-service/latest/developerguide/managedomains-cloudwatchmetrics.md +++ b//opensearch-service/latest/developerguide/managedomains-cloudwatchmetrics.md @@ -7 +7 @@ -Viewing metrics in CloudWatchInterpreting health charts in OpenSearch ServiceCluster metricsDedicated master node metricsDedicated coordinator node metricsEBS volume metricsInstance metricsWarm metricsCold storage metricsOpenSearch Optimized Instances (OR1) metricsAlerting metricsAnomaly detection metricsAsynchronous search metricsAuto-Tune metricsMulti-AZ with Standby metricsPoint in time metricsSQL metricsk-NN metricsCross-cluster search metricsCross-cluster replication metricsLearning to Rank metricsPiped Processing Language metrics +Viewing metrics in CloudWatchInterpreting health charts in OpenSearch ServiceCluster metricsDedicated master node metricsDedicated coordinator node metricsEBS volume metricsInstance metricsWarm metricsCold storage metricsOpenSearch Optimized Instances (OR1) metricsOptimized engine metricsAlerting metricsAnomaly detection metricsAsynchronous search metricsAuto-Tune metricsMulti-AZ with Standby metricsPoint in time metricsSQL metricsk-NN metricsCross-cluster search metricsCross-cluster replication metricsLearning to Rank metricsPiped Processing Language metrics @@ -30,0 +31,2 @@ OpenSearch Service publishes the following metrics to CloudWatch: + * Optimized engine metrics + @@ -339,0 +342,29 @@ Metric | Description +## Optimized engine metrics + +Amazon OpenSearch Service provides the following metrics for domains running the Optimized engine (log analytics). For more information, see [Optimized for log analytics](./optimized-log-analytics.html). + +Metric | Description +---|--- +`NativeMemoryPressure` | The percentage of native (off-heap) memory in use on the node. This metric is analogous to `JVMMemoryPressure` but measures memory consumed by the native analytics engine rather than the Java heap. Applicable to hot, warm, and coordinator nodes. Relevant statistics: Maximum +`NativeRuntimeResidentMemory` | The amount of resident memory, in bytes, consumed by the native analytics engine on the node. Applicable to hot, warm, and coordinator nodes. Relevant statistics: Maximum, Average +`NativeSearchRuntimeCPUUtilization` | The CPU utilization, as a percentage, of the DataFusion query execution engine on the node. Applicable to hot, warm, and coordinator nodes. Relevant statistics: Maximum, Average +`ThreadpoolNativeSearchCPUQueue` | The number of queued tasks in the native search CPU thread pool. If the queue size is consistently high, consider scaling your cluster. Applicable to hot, warm, and coordinator nodes. Relevant statistics: Maximum +`ThreadpoolNativeSearchCPUThreads` | The size of the native search CPU thread pool. Applicable to hot, warm, and coordinator nodes. Relevant statistics: Maximum + +###### Note + +The following metrics don't apply to Optimized domains because OpenSearch Dashboards is not available: + + * `OpenSearchDashboardsHealthyNodes` + + * `OpensearchDashboardsReportingFailedRequestSysErrCount` + + * `OpensearchDashboardsReportingFailedRequestUserErrCount` + + * `OpensearchDashboardsReportingRequestCount` + + * `OpensearchDashboardsReportingSuccessCount` + + + +