starrocks

Why we included this project

Data teams that need sub-second answers over very large analytical datasets will find a serious workhorse here. StarRocks is a distributed, vectorized SQL engine built on MPP and columnar storage, a combination that suits multi-dimensional reporting, real-time updates, and ad-hoc exploration. Its external catalog layer reads Hive, Iceberg, Hudi, or Delta Lake tables in place, so you can run fast analytics on the lake without migrating data into a proprietary store. Asynchronous materialized views and caching speed up repeated dashboard queries against those external sources, while native tables absorb high-ingest real-time workloads. Whether you want a faster compute layer over an existing data lake or a dedicated OLAP store, this covers both without forcing you to pick a side.