StarRocks is a new generation of extremely fast full-scene MPP databases.
StarRocks has fully absorbed the outstanding research results of relational OLAP databases and distributed storage systems in the era of big data. On the basis of industry practice, StarRocks has further improved, optimized, upgraded the architecture, and added many new functions, forming a new enterprise-grade product.
StarRocks is committed to building an extremely fast unified analysis experience to meet various data analysis scenarios for enterprise users. It supports multiple data models (detailed models, aggregation models, update models), multiple import methods (batch and real-time), and can integrate and connect to various existing systems (Spark, Flink, Hive, ElasticSearch).
StarRocks is compatible with the MySQL protocol. You can use the MySQL client and commonly used BI tools to connect with StarRocks for data analysis.
StarRocks uses a distributed architecture where data tables are horizontally partitioned and stored in multiple copies. The cluster scale can be scaled flexibly, and can support data analysis at the level of 10PB; supports MPP frameworks and accelerates computation in parallel; supports multiple replicas, and has elastic fault tolerance.
StarRocks uses a relational model, uses strict data types and columnar storage engines to reduce read/write amplification through coding and compression technology; uses vectorized execution to fully tap the parallel computing capabilities of multi-core CPUs, thereby significantly improving query performance.