apache/paimon

Health v2 · maintenance only

Adoption not part of the score

3384 stars · 1385 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Apache Paimon is a lakehouse table format that combines lake format with LSM tree structure to support real-time streaming updates. It enables building realtime lakehouse architectures with Apache Flink and Apache Spark for both streaming and batch operations.

Use cases

• build a realtime lakehouse with streaming and batch processing
• store streaming updates in a data lake table format
• query up-to-date tables from Flink and Spark
• replace Flink Table Store with an Apache lake format
• manage changelog data in object storage with LSM merges
• unify offline and online analytics on one lakehouse storage

When to choose

• you need frequent streaming updates and changelog semantics in a lakehouse, which formats like Iceberg handle less naturally
• your stack is Flink- or Spark-based on the JVM
• you want an Apache Foundation project with active community support

When to avoid

• you only need append-only analytics where Iceberg or Delta Lake ecosystems are more mature
• your workloads are outside the JVM/Flink/Spark ecosystem
• you need a full database rather than a table format

Facets

library · maturity active

database streaming etl big-data databases analytics jvm cloud lakehouse table-format apache-flink apache-spark lsm-tree streaming-updates data-lake data-engineering

Member repositories

For agents

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