# ElastiCache for Valkey adds OpenTelemetry metrics and optional 15-second publishing > AWS adds attribute-based PromQL queries and optional 15-second metric publishing for node-based Valkey clusters, where CloudWatch supports OpenTelemetry metrics. Canonical URL: https://www.devobs.io/news/news-event-46ddfb0dab62ba0fd25fe7a5/ By: Samira Haddad Published: 2026-10-04T09:26:45.427Z Updated: 2026-10-04T09:26:45.427Z Event date: 2026-10-02 Section: Cloud AWS announced OpenTelemetry metrics and a detailed monitoring mode for node-based Amazon ElastiCache for Valkey clusters on [October 2, 2026](https://aws.amazon.com/about-aws/whats-new/2026/10/amazon-elasticache-valkey-opentelemetry-metrics-detailed-monitoring). The metrics publish to Amazon CloudWatch and carry attributes that operators can filter and aggregate using Prometheus Query Language (PromQL). The key distinction is cadence and metric selection: standard monitoring publishes a core set every 60 seconds, while detailed monitoring lets operators choose metrics for 15-second publishing. ## What each monitoring mode includes **Standard monitoring** continues to publish CloudWatch Metrics (Classic) every 60 seconds and adds a core set of OpenTelemetry metrics at the same interval, at no additional charge. That core set also powers ElastiCache Insights, a pre-built CloudWatch dashboard. **Detailed monitoring** lets operators select from the full OpenTelemetry metric set and publish their selections every 15 seconds. AWS says operators can query these metrics in CloudWatch and Grafana using PromQL, with use cases including finding nodes approaching connection limits, breaking errors down by type and forecasting memory exhaustion. These capabilities are described in the [AWS announcement](https://aws.amazon.com/about-aws/whats-new/2026/10/amazon-elasticache-valkey-opentelemetry-metrics-detailed-monitoring). ## What the shorter interval means for operations At the stated cadence, detailed monitoring schedules four metric publications per minute rather than one. That provides finer reporting granularity; it should not be read as a guarantee that every brief latency spike will be captured. A practical starting point is to select detailed metrics around a specific operational question—connection pressure, error patterns or memory headroom—rather than automatically selecting the entire set. This ties diagnostic value to the metrics being purchased. ## Availability, setup and billing AWS lists availability for **node-based Valkey clusters in AWS Regions where CloudWatch supports OpenTelemetry metrics**. To configure detailed monitoring, open a cluster’s **Metrics** tab in the ElastiCache console, select **Detailed**, then choose **Configure detailed metrics**. According to the [announcement](https://aws.amazon.com/about-aws/whats-new/2026/10/amazon-elasticache-valkey-opentelemetry-metrics-detailed-monitoring), ElastiCache adds no charge for detailed monitoring, but CloudWatch OpenTelemetry pricing applies to the detailed metrics selected, alarms created and PromQL API queries run. For budgeting, no additional ElastiCache fee does not mean the monitoring pipeline is free. ## Source references -