At a Glance:
HyperDX is an open-source observability tool for searching, visualizing, and correlating logs, traces, and metrics on top of existing ClickHouse clusters, with native OpenTelemetry support.
Overview:
HyperDX is a self-hosted log and trace exploration platform that operates directly on your existing ClickHouse infrastructure. It provides engineers with a single interface for correlating logs, metrics, traces, and session replays to debug production issues without needing complex query languages. The tool supports schema-agnostic search, full-text queries with property syntax, and live tailing, and it integrates with OpenTelemetry for vendor-neutral instrumentation. It is designed for teams who want a Kibana-like experience backed by ClickHouse, deployable via the ClickStack bundle or as a standalone container.
Key Decision Points:
ClickHouse-native: Requires an existing ClickHouse cluster or can be deployed as part of the ClickStack bundle, which also includes an OpenTelemetry Collector and MongoDB.
Deployment flexibility: Can be run locally as a single container, connected to a hosted ClickHouse Cloud instance, or deployed in production with provided instructions; UI, API, and OTel collector ports must be accessible.
Query model: Supports an intuitive full-text and property-based search syntax (e.g.,
level:err) with optional SQL, plus native JSON string querying.Telemetry ingestion: Accepts data via OpenTelemetry SDKs or HyperDX’s own Browser, Node.js, and Python SDKs, with instrumentation available across multiple languages and Kubernetes.
Resource baseline: A minimum of 4GB RAM and 2 cores is recommended for testing environments.
Core Features:
Unified telemetry search: Correlate logs, traces, metrics, and session replays in a single interface.
Schema-agnostic operation: Works on top of your existing ClickHouse schema without requiring schema changes.
Full-text and property search: Supports partial text search and property filters like
level:errwithout requiring SQL.Alerting: Configure alerts with a few clicks based on telemetry data.
Live tail: Stream logs and traces in real time to view the most recent events.
Native JSON querying: Query and filter on nested JSON string fields directly.
Use Cases:
Engineers troubleshooting production incidents who need to correlate logs, traces, and session replays in one place.
Teams already running ClickHouse who want to add a search and visualization layer for observability without migrating data.
Developers instrumenting applications with OpenTelemetry who need a compatible backend for logs, metrics, and traces.
Open-Source Alternative Value:
HyperDX provides a self-hosted alternative to commercial observability platforms by leveraging an existing ClickHouse data store, which allows teams to avoid per-gigabyte telemetry pricing and retain control over their data infrastructure. Its compatibility with the vendor-neutral OpenTelemetry standard and schema-agnostic design means users can integrate it into diverse environments without being locked into a proprietary ingestion format or query language.
