What is this?
StarRocks is an open-source, high-performance query engine for sub-second, ad-hoc analytics on data warehouses and data lakes, built as a Linux Foundation project.
StarRocks is an open-source, high-performance query engine for sub-second, ad-hoc analytics on data warehouses and data lakes, built as a Linux Foundation project.Is this a database I deploy myself, or a hosted service?It is a self-hosted analytical database you run on your own cluster, made of two components: a Frontend (FE) for metadata and query planning, and a Backend (BE) for storage and execution.
It is a self-hosted analytical database you run on your own cluster, made of two components: a Frontend (FE) for metadata and query planning, and a Backend (BE) for storage and execution.How do I get it running?Follow the Quick Starts and Deploy guides in the docs; you can compile StarRocks with Docker or deploy a cluster manually.
Follow the Quick Starts and Deploy guides in the docs; you can compile StarRocks with Docker or deploy a cluster manually.Do I need Hive, Iceberg, or Delta Lake already set up to use it?No, those are optional - StarRocks works as a standalone database, but it can also query data directly in Hive, Iceberg, Delta Lake, and Hudi without importing it first.
No, those are optional - StarRocks works as a standalone database, but it can also query data directly in Hive, Iceberg, Delta Lake, and Hudi without importing it first.How complex is it to deploy and maintain a cluster?It's a distributed system, but the architecture is designed to be simple: it tunes query plans automatically, rebalances resources when nodes are added or removed, and recovers replicas after node failure on its own.
It's a distributed system, but the architecture is designed to be simple: it tunes query plans automatically, rebalances resources when nodes are added or removed, and recovers replicas after node failure on its own.Who is this built for?Teams running multi-dimensional, real-time, or ad-hoc analytics at scale; it's used in production by companies like Airbnb, Pinterest, Tencent, Trip.com, and Xiaohongshu.
Teams running multi-dimensional, real-time, or ad-hoc analytics at scale; it's used in production by companies like Airbnb, Pinterest, Tencent, Trip.com, and Xiaohongshu.What makes StarRocks' queries so fast?It uses a native vectorized SQL engine that fully exploits CPU parallelism, giving sub-second query returns that are 5 to 10 times faster than previous-generation systems.
It uses a native vectorized SQL engine that fully exploits CPU parallelism, giving sub-second query returns that are 5 to 10 times faster than previous-generation systems.Can I connect my existing BI tools to it?Yes - StarRocks supports ANSI SQL and is compatible with the MySQL protocol, so most SQL clients and BI software can connect to it directly.
Yes - StarRocks supports ANSI SQL and is compatible with the MySQL protocol, so most SQL clients and BI software can connect to it directly.How does real-time data updating work?StarRocks uses a primary-key update model that supports efficient upsert and delete operations even while queries run concurrently.
StarRocks uses a primary-key update model that supports efficient upsert and delete operations even while queries run concurrently.What changed with the shared-data architecture in v3.0?Starting from version 3.0, StarRocks added a shared-data architecture (separating storage and compute) to provide better scalability and lower costs than the original shared-nothing design.
Starting from version 3.0, StarRocks added a shared-data architecture (separating storage and compute) to provide better scalability and lower costs than the original shared-nothing design.Which topics is it in?lakehouse, realtime-database, olap, data-warehouse, distributed-database, big-data, cloud-native, analytics, enterprise, mysql, sql, real-time
lakehouse, realtime-database, olap, data-warehouse, distributed-database, big-data, cloud-native, analytics, enterprise, mysql, sql, real-time