Quickwit is an open-source search engine on object storage with subsecond latency for large datasets.
The project made by the authors of the Rust search engine library tantivy recently raised $2.6 million in a seed round co-led by FirstMark and firstminute with many more participants.
Quickwit 0.2 was launched in January with new features such as the ability to ingest Kafka natively with exactly-one semantics, a search stream API, PostgreSQL metastore, tag pruning, and a proper indexing pipeline.
The search engine uses a different approach than Elasticsearch which relies on document replication. Meanwhile, Quickwit indexes documents on a single node and leaves it to an object storage to replicate the resulting index files.
“We need to replicate documents somewhere, to consider them ingested,” Paul Masurel, co-founder of Quickwit wrote in a blog post. “We make it possible for users to plug Quickwit straight into their favorite distributed queue. Right now Quickwit only supports Kafka.”
Common use cases for Quickwit include searching through logs, adding full-text search capabilities to OLAP databases such as ClickHouse, and searching through backups that sit on cloud storage by adding Quickwit index files on your same storage.