What is SereneDB used for?

SereneDB is a Postgres-compatible database that combines Elasticsearch-style full-text search with ClickHouse-style analytics in a single engine. It is a good fit for log and observability analytics with search, and replacing separate search and OLAP systems.

Is SereneDB open source?

Yes. SereneDB is open source under the Apache-2.0 licence. Its source code is on GitHub at serenedb/serenedb and is written mainly in C++.

Yes. SereneDB is open source, so the software itself is free to use.

Can I self-host SereneDB?

Yes. SereneDB can be self-hosted on your own server or infrastructure.

What is SereneDB an alternative to?

SereneDB is an open-source alternative to Elastic. Other open-source alternatives to Elastic include ParadeDB, Infinity and GreptimeDB.

Is SereneDB actively maintained?

Yes. The most recent commit to SereneDB was on 2 October 2026, and the latest release is v26.09.1.3, published on 1 October 2026. The project has 858 stars on GitHub.

SereneDB: open-source alternative to Elastic

About SereneDB

SereneDB is an open-source database that unifies search and analytics. It aims to give the search experience of Elasticsearch and the analytical speed of ClickHouse in one Postgres-compatible system, treating search as a first-class feature of the query engine instead of a separate cluster.

Search is powered by IResearch, a C++ information retrieval library that members of the SereneDB team have developed since 2016. The database supports BM25 ranking, vector and hybrid search, and column-wise updates to keep data fresh. A zero-ETL mode queries Parquet, Iceberg, CSV and JSON data directly on remote storage such as S3 or HDFS without an ingestion pipeline.

The project publishes open benchmarks comparing it with ClickHouse, Elastic, OpenSearch and ParadeDB, which readers can review for themselves, and offers Slack and Telegram communities. SereneDB is written in C++ and licensed under Apache-2.0.

Key features

Postgres-compatible SQL interface
Full-text search with BM25 ranking
Vector and hybrid search
Column-oriented analytics engine
Zero-ETL queries on S3 and HDFS data
Efficient column-wise real-time updates

Good fit for

→Log and observability analytics with search
→Replacing separate search and OLAP systems

Built with: C++; ClickHouse; PostgreSQL
Tags: database; search; analytics; olap; postgresql-compatible; vector-search; cpp; bm25

SereneDB: questions and answers

What is SereneDB used for?: SereneDB is a Postgres-compatible database that combines Elasticsearch-style full-text search with ClickHouse-style analytics in a single engine. It is a good fit for log and observability analytics with search, and replacing separate search and OLAP systems.
Is SereneDB open source?: Yes. SereneDB is open source under the Apache-2.0 licence. Its source code is on GitHub at serenedb/serenedb and is written mainly in C++.
Is SereneDB free?: Yes. SereneDB is open source, so the software itself is free to use.
Can I self-host SereneDB?: Yes. SereneDB can be self-hosted on your own server or infrastructure.
What is SereneDB an alternative to?: SereneDB is an open-source alternative to Elastic. Other open-source alternatives to Elastic include ParadeDB, Infinity and GreptimeDB.
Is SereneDB actively maintained?: Yes. The most recent commit to SereneDB was on 2 October 2026, and the latest release is v26.09.1.3, published on 1 October 2026. The project has 858 stars on GitHub.