Extremely fast query performance in all scenarios
Running queries against a single table or multiple tables, querying local tables or data stored in data lakes, with StarRocks, experience 3 to 10 times better performance than other popular solutions.
Fully vectorized execution engine
With a columnar storage engine and fully vectorized opterators, StarRocks makes full use of CPU processing power and SIMD instructions to boost performance.
Cost-based optimizer (CBO)
Query planning is just as or even more important than query execution in boosting query performance. StarRocks' CBO seamlessly helps you find the most optimal plan to execute your queries through multi-table join reorder, sub-query rewrite, Join distributed execution selection, and more, giving StarRocks its class-leading query performance, especially with complex multi-table queries.
Pipeline execution framework
Engineered for multicore scheduling, the query parallelism in StarRocks can be adaptively adjusted without the need to set parameters, boosting performance in highly concurrent and hybrid workload.
Intelligent materialized view
To further boost performance and reduce load, you can perform transformations on your data using StarRocks' materialized view. With pre-computed joins and data aggregations, the query execution time can be further reduced, enabling more demanding scenarios such as highly concurrent analytics and exploratory ad-hoc analysis against a massive amount of data.
Real-time Analytics: Guaranteed data freshness
From streaming data to data capture, StarRocks guarantees the latest data and is ready to provide fresh insight.
Primary Key table
StarRocks' Primary Key table is specifically engineered to deliver uncompromised query performance when having concurrent real-time data change operations. The primary key index enables data change operations to be resolved at data ingestion, leaving only the latest version of records. Hence, at query, the merge-on-read operation is not required, predicates & index pushdowns can also be used to boost query performance.
Intelligent materialized view
The materialized view of StarRocks can be incrementally updated at data ingestion and perform query rewrite at query execution. With guaranteed consistency between the base table and the MV, StarRocks' MV further accelerates real-time analytics.
StarRocks Flink-cdc connector
Easily sink data into StarRocks in real-time.
Analytics with flexibility: Unleash the full potential of your data
A database that adapts to your use cases, StarRocks provides the flexibility to scale your business on demand with ease.
Use StarRocks as a unified query interface and store your historical data in data lakes while having real-time streaming data ingested into StarRocks.
Star or Snowflake schema
StarRocks supports Star and Snowflake schema for extreme performance even on multi-table queries.
ANSI SQL syntax and MySQL protocol support- StarRocks is compatible with a wide range of clients and BI tools.
StarRocks' resource group can efficiently provision resources for multiple tenants within a cluster that has heterogeneous workloads - implements intra-process resource isolation, and protects mission-critical queries while making full use of resources. This is achieved with StarRocks' pipeline execution framework: its ability to perform time-quantum-based user-mode scheduling to enable efficient CPU resource reuse.