Grafana Labs Announces Advancements in Resource Efficiency for Improved Cost Optimizations in Kubernetes Monitoring in the Cloud

Grafana Labs announced updates to Kubernetes Monitoring in Grafana Cloud, the solution for all levels of Kubernetes usage within an organization. These new features target resource utilization and predictions, historical state analysis, and simplified third-party integrations. Additionally, the Grafana Labs team shared new functionality for Kubernetes users in Grafana Incident, its tool for easier incident management, and will present on numerous topics at KubeCon + CloudNativeCon EU 2023, including advances in high-resolution histograms in Prometheus.

New Kubernetes Monitoring features

With Kubernetes Monitoring, the solution introduced to the fully managed Grafana Cloud platform last year, users can automatically ship metrics to Grafana after installing the Grafana Agent into one or more Kubernetes clusters. Once this connection is made, Grafana Cloud users have out-of-the-box access to their Kubernetes metrics, logs, and events via pre-built dashboards and alerts. Grafana Labs continues to expand the solution’s capabilities to give users even more visibility into their Kubernetes fleet:

• Visualize resource utilization efficiency: With this new cost-focused feature, users get a bird’s eye view of all the nodes in a cluster and the condition of those nodes. This helps reduce the deviation between resource allocation and actual resource utilization. 
• Predict CPU and RAM usage: This new ML-powered feature provides resource usage forecasts with a simple click. It helps users assure high availability and avoid performance degradation. With insights of future resource usage peaks, users can identify resource deprivation issues in their Kubernetes infrastructure. This feature is currently available for Grafana Cloud Pro and Advanced users.
• Retroactively visualize the state of your fleet: Kubernetes Monitoring visualizes a fleet’s historical state in respect to user-selected dates and time. Users can navigate to different time frames to see the exact state of all visualizations during that moment. This feature will be available by the end of April.
• Monitor additional services running on Kubernetes: With the recently released Kubernetes Monitoring integrations, users can now monitor additional services running on Kubernetes with no more effort than selecting which service to connect to in the Grafana Connections UI. Current integrations include Grafana Mimir, CockroachDB, cert-manager, and Node Exporter, with NGINX, CoreDNS, and etcd coming soon.

Kubernetes Monitoring is available to all Grafana Cloud users, including those on the generous, forever-free tier.

Grafana Incident Investigations for Kubernetes users

Grafana Incident is a tool available in Grafana Cloud that automates the toilsome tasks of incident management. Its new Investigations feature streamlines incident response in Kubernetes environments by automatically identifying issues in cluster services, such as noisy neighbors and increased log errors, upon incident declaration from an alert. This functionality will soon be available in preview for Grafana Cloud Pro and Advanced users.

Latest developments in Prometheus histograms

As part of the Grafana Labs commitment to supporting the open source community, the team is also presenting learnings on Prometheus histograms during two talks at KubeCon + CloudNativeCon EU 2023. With advances in high-resolution histograms – such as better storage efficiency, higher accuracy of quantile estimations, flexible histogram buckets, simple configuration, and easier adoption – Grafana Labs is driving this technology forward in both Prometheus and OpenTelemetry, and working to maintain compatibility throughout the cloud native ecosystem. Grafana Mimir, Grafana Labs’ horizontally scalable open source TSDB, is one of the first projects to have experimental support for native histograms.