Skip to content

Joining Sources ​

In rekuiper, JOIN operations correlate data from multiple sources.

rekuiper supports two join combinations:

  • Stream-to-Stream Join: Joins multiple unbounded streams. Stream-to-stream joins require an explicit window definition.
  • Stream-to-Table Join: Joins a stream to a table. Arriving stream events trigger join calculations against current table records.

Supported join types include INNER, LEFT, RIGHT, FULL, CROSS, and temporal ASOF JOIN / LEFT ASOF JOIN (which matches each fast stream event with the most recent reference event at or before its timestamp without requiring a window cross-product).

Streaming Join Topologies in rekuiper1. Stream-to-Stream JoinStream A + Stream B in WindowINNER / LEFT / RIGHT / FULL / CROSSRequires GROUP BY Window ClauseSharedMap Copy-on-Write Buffers2. Stream-to-Table JoinLive Stream + Lookup / Scan TableEnriches Events on ArrivalBacked by Memory, Redis, or SQLO(1)VecDeque Bounded Eviction3. Temporal ASOF JOINASOF JOIN / LEFT ASOF JOINMatches Latest Reference ReadingCondition: s.ts >= r.tsZero Window Cross-Product Overhead

rekuiper is open source (Apache-2.0 / MIT) · For an all-in-one MQTT broker with embedded rekuiper rules, see IndraMQTT.