Observability data pipeline that collects, transforms, and routes logs and metrics as an agent or aggregator
MPL-2.0
- Rust
- CUE
- HTML

About Vector
Vector is an open-source observability data pipeline for collecting, transforming, and routing logs and metrics. It runs end-to-end as an agent or aggregator, so teams can consolidate telemetry flow and send data to current or future vendors. The focus is control over observability data, including cost reduction, enrichment, and data security placement.
Pipeline configuration is built from sources, transforms, and sinks. Sources include Docker logs, files, HTTP, journald, Kafka, and sockets. Transforms include dedupe, filter, remap, Lua, and log-to-metric. Sinks include AWS CloudWatch Logs, Amazon S3, ClickHouse, Elasticsearch, and Google Cloud Storage.
Vector is built in Rust and maintained by Datadog's Community Open Source Engineering team. It is used by Atlassian, T-Mobile, Comcast, Zendesk, Discord, Fastly, Visa, and Instacart, and has over 500 contributors. It is licensed under MPL-2.0 and self-hosted as a single binary.
Key features
- Collect, transform, and route logs and metrics
- Deploy as an agent or aggregator
- Sources include Docker logs, files, HTTP, journald, Kafka, and sockets
- Transforms include dedupe, filter, remap, Lua, and log-to-metric
- Sinks include CloudWatch Logs, S3, ClickHouse, Elasticsearch, and GCS
Details
- First released
- 2018
- Deployment
- Agent · aggregator
- Data
- Logs · metrics beta
- Language
- Rust
- Company
- Datadog
- License
- MPL-2.0
