This topic describes the features of DRDS to help you understand DRDS.


DRDS is developed by Alibaba Cloud. It integrates the distributed SQL engine Distributed Relational Database Service (DRDS) and the proprietary distributed storage X-DB. Based on the integrated cloud-native architecture, it supports up to tens of millions of concurrent connections and hundreds of petabytes of data storage. It aims to provide solutions for massive data storage, ultra-high concurrent throughput, large table performance bottlenecks, and complex computing efficiency. It has been tested in each Double 11 Shopping Festival and in the businesses of Alibaba Cloud customers in various industries, boosting the digital transformation of enterprises.

DRDS adopts standard relational database technologies to provide core features. It is deployed with the comprehensive management, O&M, and product-based capabilities, which makes it more stable, reliable, scalable, maintainable, and operable as in a traditional single-instance MySQL database.

DRDS has been used on Alibaba Cloud and Apsara Stack for many years, and experienced core transaction businesses in each Double 11 Shopping Festival and in the businesses of Alibaba Cloud customers in various industries. PolarDB-X supports core online businesses for a large number of users across many industries, such as Internet, finance and payment, education, communications, and public utilities. It is the de facto standard for all core online businesses of Alibaba Group and businesses of Alibaba Cloud customers to access distributed databases.


  • Stability

    For most applications, the responsibility of relational databases is the core foundation of the data management system. This affects user experience on services and protects business data. Therefore, stability is the core factor for database selection.

    Proper use of the time-tested MySQL databases ensures the stability of DRDS. However, single-instance MySQL databases are relatively weak in the high-concurrency, large-volume data storage, and complex computing scenarios.

    DRDS distributes data to multiple ApsaraDB RDS for MySQL instances, with proper concurrency, data storage, and computing loads on each. All ApsaraDB RDS for MySQL instances are stable and distributed logic is processed in DRDS. All together forms a stable, reliable, and highly scalable distributed relational database system.

    Compared with proprietary distributed NewSQL databases, DRDS focuses on continuous stability and O&M availability. By using standard database technologies, PolarDB-X databases can be operated like single-instance databases, allowing users to easily get started and increasing business value.

  • High scalability

    Compared with traditional single-instance relational databases, DRDS uses a hierarchical architecture to ensure linear scaling in concurrency, computing, and data storage. DRDS computing and storage resources are added for scale-out.

    Compared with new cloud-native databases based on distributed storage, DRDS databases can be scaled out without limit, which eliminates the worries and O&M pressures caused by database scalability during rapid business development.

  • Continuous O&M availability

    For most applications, relational databases must stably work around the clock. Therefore, continuous O&M availability is the key capability for relational databases.

    DRDS has been used on Alibaba Cloud and Apsara Stack for many years, providing a variety of product capabilities and a complete O&M system. Services can be automatically scheduled and integrated based on a complete set of API operations.

    • Lifecycle management
      • Instance creation, restart, and release
      • Database creation and deletion
      • Graphical user interface (GUI) operation
    • Capacity management
      • Horizontal partitioning and vertical partitioning
      • Read/write splitting
      • Flexible reconfiguration
      • Sharding
    • Security and audit
      • VPC
      • IP address whitelist
      • Account and permission management
    • Disaster recovery management
      • Integrated backup and recovery
      • SQL flashback
      • Table recycle bin
      • Multi-zone instance disaster recovery
    • Monitoring and alerting
      • Custom layered monitoring
      • CloudMonitor integration and alert management for key metrics
    • Data ecology
      • Data migration, synchronization, and subscription based on Data Transmission Service (DTS)
      • Data integration
      • DMS data management
      • Quick BI integration
      • OpenSearch and Elasticsearch
      • Big data computing and data warehouse