Apache Paimon is a table format for building lakehouse architectures on data lakes. It supports batch and streaming workloads in one storage layer, with streaming updates through a log-structured merge design, time travel, schema evolution, and vector and full-text search for AI pipelines. It is used together with engines such as Apache Flink and Apache Spark. Paimon is a format and library rather than a hosted service, so organisations run it on their own storage and compute.
Paimon is licensed under the Apache License 2.0 and has been a top-level project of the Apache Software Foundation since April 2024, when it graduated from incubation. The foundation is based in the United States, so governance and trademarks sit with a US non-profit. Because Paimon is open source and self-hosted, the data it manages stays in the storage the organisation chooses, whether in the EU or elsewhere. For a European organisation, the choice of storage location and cloud infrastructure is therefore a relevant sovereignty question, alongside the US governance of the foundation.