Distributed Database Middleware (DDM)

Distributed Database Middleware (DDM) solves distributed expansion issues of databases. DDM breaks the capacity and performance bottleneck of traditional databases and achieves high concurrent access to mass data.

Join the open beta test to claim a limited free trial. Learn More

Product Advantages


Automatic horizontally partitioned, smooth capacity expansion, and one-click linear capacity expansion.


Data access in petabyte-scale volume; ten-fold database connections compared with a single-node database; million-scale-level high concurrency.


High available clusters with second-level automatic fault recovery; strong consistent transactions and eventually consistent transactions; multiple security policies to protect privacy of the database and users.


Compatible with MySQL protocols; read/write splitting requires no code modification; DDM console simplifies the OM.

Product Architecture

Distributed Database Middleware (DDM)

DDM interconnects with multiple RDSs in the backend, achieving transparent distributed access of databases.

  • Applications access DDM through standard MySQL protocol. DDM automatically sends queries to shards in RDS according to routing rules, and then returns aggregated results to applications.
  • DDM automatically identifies SQL types, and allocates write operation to RDS active instances and read operation to RDS standby instances based on the read/write splitting policy.
  • DDM console enables you to manage and maintain DDM instances, logical databases, logical tables, and sharding rules.
  • Smooth capacity expansion function enables you to easily add new RDS instances when the storage nodes do not have enough storage capacity. Data redistribution is automatically achieved.
  • Multiple security policies such as VPC, subnet, and security group are provided to ensure data security and user privacy. Only applications, DDMs, and RDSs in the same VPC can access each other.

Application Scenarios

Large-scale Applications

In industries such as e-commerce, financial services, O2O, retailing, and social media, the response of core trading systems become slow due to mass data generated by large user bases and frequent marketing activities, impairing the development of services.
DDM provides data linear horizontal expansion capability to improve database performance and access efficiency in a real-time manner, helping you copy with high-concurrent real-time trading scenarios.

IoT Data

In the Internet of Things (IoT) scenarios such as industrial monitoring, remote control, smart cities, smart homes, and the Internet of Vehicles, there are numerous sensors and monitoring devices, high sampling frequency, and a large amount of data. Data generated in these scenarios may exceed the storage capacity of single-node databases, causing capacity bottlenecks.
DDM provides horizontal capability expansion to store mass data in an affordable manner.

File Indices

There may be a hundred million to trillion records of figures, files, and videos on the Internet and social media applications. Indices of these data are usually stored in databases, which must enable real-time operations such as adding, modification, reading, and deleting.
With high performance and distributed expansion capability,DDM enables efficient index-based search.

Cost-efficient Database Solutions

Industries such as government organizations, large-scale enterprises, and banks rely on expensive commercial solutions based on mid-range computers and high-end storage.
Compared with traditional commercial database solutions, DDM, deployed on clusters of common servers, provides cost-efficient database solutions with same or even higher performance.

Function Description

Horizontal Splitting

After a distributed database is created, data horizontal splitting can be achieved by specifying the split keys and split rules.

Smooth Capacity Expansion

RDS instances can be added to extend the storage capacity and automatically redistribute data.

Distributed Transactions

One-phase transaction submission model is supported.

Read/Write Splitting

Read/Write splitting is available to applications. You can configure Read/Write splitting using the DDM console, sparing efforts on changing code.

Global Sequence

Global unique sequence in ascending order is supported in distributed scenarios, meeting requirements on scenarios where primary keys or unique keys are required or other scenarios.

Online Monitoring

Help you to detect resource and performance bottlenecks by monitoring read and write ratio and SQL statements that are executed slowly.

DDM Console

DDM console is provided for you to manage and maintain DDM instances, logical databases, logical tables, and sharding rules.