Metrics

OpenObserve provides comprehensive metrics collection, storage, and visualization capabilities for monitoring your applications and infrastructure performance in real-time.

Overview

Metrics in OpenObserve enable you to track key performance indicators, monitor system health, and gain insights into your application's behavior over time. Designed for high-throughput environments, OpenObserve efficiently handles metrics ingestion, storage, and querying at scale while maintaining cost-effectiveness.

Key Features

Metrics Ingestion

• Multiple Formats: Support for multiple formats including Prometheus remote-write and structured JSON metrics via HTTP ingestion.
• Push & Pull Models: Supports push-based ingestion and integration with systems like Prometheus for pull-based scraping.
• High Throughput: Handle millions of metrics per second with minimal latency.
• Optimized Processing: Optimized ingestion pipelines for high-throughput environments.

Metric Structure

• Time Series Data: Native support for time-series metrics with timestamp precision.
• Multi-dimensional: Handle metrics with multiple labels and dimensions for detailed analysis.
• Aggregation Functions: Built-in support for common aggregation functions (sum, avg, min, max, count).
• Custom Metrics: Flexible schema for application-specific metrics and KPIs.

Query & Analysis

• PromQL Support: Full compatibility with Prometheus Query Language for familiar querying.

• SQL Interface: Use SQL syntax for complex metrics analysis and reporting.

• Time Range Selection: Flexible time range queries with support for relative and absolute time periods.

• Mathematical Operations: Perform calculations and transformations on metrics data.

Visualization & Dashboards

• Real-time Charts: Interactive time-series visualizations with multiple chart types.

• Custom Dashboards: Create comprehensive dashboards with multiple metrics panels.

• Alerting Integration: Set up alerts based on metrics thresholds and conditions.

• Share Metrics Views: Share an exact metrics view with your team by copying a URL from the Share button in the metrics toolbar. The URL encodes your current panel configuration — chart type, queries (PromQL or SQL), stream selection, time range, and refresh interval — so recipients see the same chart and data. Opening a shared link auto-restores the view and runs the query immediately.

The share URL carries the full panel state as a versioned base64 blob in the metrics_data query parameter. The metrics page also accepts lightweight deep-link overrides (chart_type, query_type, stream_name, query) for programmatic linking from alerts, dashboards, and other pages.

Storage & Performance

• Optimized Compression: Advanced compression algorithms reduce storage costs.

• Efficient Indexing: High-performance indexing for fast query execution across large datasets.

• Downsampling: Automatic data downsampling for long-term storage optimization.

• Retention Policies: Configurable retention settings to balance storage costs and data availability.

Integration & Compatibility

• Prometheus Compatible: Full compatibility with Prometheus ecosystem and exporters.
• OpenTelemetry Ecosystem: Full compatibility with OpenTelemetry collectors and instrumentation libraries.
• API Access: RESTful APIs for programmatic access to metrics data.
• Standard Exporters: Support for popular metrics exporters (Node Exporter, cAdvisor, etc.).

Get started with metrics

Ready to send your first metrics to OpenObserve?

• Quickstart: get OpenObserve running in 5 minutes.
• Prometheus remote-write: the most common path for metrics ingestion.
• Telegraf: collect metrics from Telegraf agents.

Next steps

• OpenTelemetry / OTLP: unified ingestion for logs, metrics, and traces.
• Dashboards: visualize metrics alongside logs and traces.
• Alerts: notify on metric thresholds and anomalies.
• Downsampling: keep long-term metrics affordable.

Need some help?

• Join our Community Slack
• Or Contact support