- Home
- Time-Series Databases
- OpenTSDB
OpenTSDB
What it is
OpenTSDB is a free, self-hosted system for storing and serving large volumes of time-series data while preserving the supplied granularity. It supports millisecond precision and can retain data indefinitely. Running on Hadoop and HBase, it scales by adding nodes and is described as capable of millions of writes per second. Data can enter through Telnet or HTTP APIs, batch import, or a Java library. Command-line tools and an HTTP API support OpenTSDB query syntax, with tag filters, aggregation, downsampling, rate conversion, functions, and expressions. Results can be viewed as GnuPlot graphs. Grafana and Bosun are among the open-source tools that can access its data. A plugin framework includes authentication, search, and real-time publishing types. OpenTSDB runs on Linux and Windows, uses the GNU LGPL v2.1 or later, and is currently in maintenance mode.
For and against
For
- Free open-source software with self-hosted deployment
- Millisecond precision and indefinite data retention
- Horizontal scaling by adding nodes
- Multiple ingestion methods, including APIs and Java
- Rich query operations and plugin framework
Against
- No built-in authentication or security review
- HBase outages can cause incoming data loss
- TSD crashes may lose buffered data
- No SQL support or retention policies
The verdict
OpenTSDB covers high-volume time-series ingestion, querying, visualization, and horizontal scaling without a software charge. Its maintenance status and documented outage, buffering, and authentication limitations require careful operational planning.
Who it suits
It suits engineers and operations staff managing large computer systems who want self-hosted, horizontally scalable time-series storage. It also fits teams using APIs, imports, Java, or compatible open-source tools.
Plans and prices
OpenTSDB
Open-source software · self-hosted
Read 8 Oct 2026 · opentsdb.net
On the record
- Analysis
- Queries support tag filters, aggregation, downsampling, rate conversion, functions, and expressions.opentsdb.net · read 8 Oct 2026
- Durability
- The FAQ says a TSD crash can lose up to about one second of buffered data by default before it is persisted to HBase.opentsdb.net · read 8 Oct 2026
- Extensibility
- OpenTSDB has a plugin framework with documented plugin types including authentication, search, and real-time publishing.opentsdb.net · read 8 Oct 2026
- Integrations
- The docs say open-source tools including Grafana and Bosun can access OpenTSDB data.opentsdb.net · read 8 Oct 2026
- Intended users
- The FAQ says OpenTSDB was written for internal use by engineers and operations staff managing large computer systems.opentsdb.net · read 8 Oct 2026
- License
- The FAQ says OpenTSDB uses the GNU LGPL v2.1 or later.opentsdb.net · read 8 Oct 2026
- Purpose
- OpenTSDB stores and serves large volumes of time-series data while preserving granularity.opentsdb.net · read 8 Oct 2026
- Querying
- Data can be queried with CLI tools or an HTTP API and viewed as a GnuPlot graph.opentsdb.net · read 8 Oct 2026
- Reliability limit
- The FAQ says prolonged HBase outages can fill TSD buffers, after which incoming data points may be discarded.opentsdb.net · read 8 Oct 2026
- Scale
- The homepage says OpenTSDB runs on Hadoop and HBase, scales by adding nodes, and can scale to millions of writes per second.opentsdb.net · read 8 Oct 2026
- Security
- The FAQ says OpenTSDB has not undergone a security review and does not include authentication, and recommends restricting untrusted access through a secured HTTP proxy.opentsdb.net · read 8 Oct 2026
- Storage
- The homepage says data is stored as provided, supports millisecond precision, and can be kept indefinitely.opentsdb.net · read 8 Oct 2026
- Support status
- The FAQ says Yahoo maintained OpenTSDB until 2021 and that it is currently in maintenance mode.opentsdb.net · read 8 Oct 2026
