Content

Target customers: Enterprise users who have requirements in data analytics scenarios, including multi-dimensional online analytical processing (OLAP) analysis, real-time data analysis, high-concurrency queries, and unified data analysis. StarRocks is widely used in industries such as new retail, Internet finance, and Internet entertainment. Features released: EMR StarRocks is a next-generation, high-speed data analytics engine that is built based on the Massively Parallel Processing (MPP) framework and is used for all data analytics scenarios. StarRocks is an enterprise-level service that provides a variety of new features. The service uses an architecture that is optimized and updated based on best practices in the industry and excellent research results of relational OLAP databases and distributed storage systems in the era of big data. StarRocks allows enterprise users to analyze data at a high speed and in a unified manner. StarRocks is suitable for various data analytics scenarios. The service supports a variety of data models and import methods, and can be connected to multiple existing services. The data models include detail, aggregate, and update models. The import method can be batch data import or real-time data import. The existing services include Spark, Flink, Hive, and Elasticsearch. StarRocks is compatible with the MySQL protocol. You can use a MySQL client or a common BI tool to access StarRocks for data analysis. StarRocks adopts a distributed architecture to horizontally divide data tables and replicate them into multiple copies. StarRocks adopts the MPP framework. StarRocks supports flexible scaling, provides error tolerance and a multi-copy mechanism, and accelerates parallel computing. It can be used to analyze 10 petabytes of data. StarRocks uses relational models, strict data types, and columnar storage engines. The service adopts the encoding and compression technologies to optimize read and write amplification. In addition, StarRocks uses a vectorized execution plan to make full use of the parallel computing power of multi-core CPUs to significantly improve query performance.