apache/fluss

Health v2 · maintenance only

Adoption not part of the score

2113 stars · 611 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Apache Fluss is a lakehouse-native streaming storage built for real-time analytics and AI, serving as the real-time data layer for Lakehouse architectures. It combines a message broker, online KV store, stream-processing state backend, and lakehouse cold store into one system, integrating with compute engines like Apache Flink and Apache Spark.

Use cases

• build real-time analytics dashboards with sub-second data freshness
• serve as the real-time data layer for a lakehouse with Paimon, Iceberg, or Hudi
• power real-time fraud detection and alerting systems
• run streaming ETL and real-time data warehouses with Flink SQL
• build feature stores and AI/ML pipelines with low-latency lookups
• ingest CDC streams from Postgres, MySQL, Oracle, or MongoDB
• track real-time shipment updates and geolocation services

When to choose

• you need sub-second fresh data feeding lakehouse analytics
• you want to unify message queue, KV store, and lakehouse cold storage
• your stack uses Apache Flink or Spark for stream processing
• you need updatable primary-key tables plus append-only log tables

When to avoid

• you only need batch analytics with no real-time requirements
• you want a simple managed message queue without lakehouse integration
• your infrastructure cannot run a distributed cluster with coordinator and tablet servers

Facets

service · maturity active

streaming database message-queue caching etl analytics big-data analytics databases jvm self-hosted cloud streaming-storage lakehouse real-time-analytics apache-flink apache-spark data-lake cdc log-tables primary-key-tables hot-tier data-engineering real-time docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="apache/fluss")