PolarDB for PostgreSQL provides a high degree of compatibility with the Amazon DynamoDB API, allowing your applications to connect seamlessly by using standard DynamoDB SDKs, such as Python Boto3 and the AWS SDK for Go, without modifying your application code. While ensuring API compatibility, this feature fully inherits the core enterprise-grade capabilities of the PolarDB cloud-native database, including scaling in seconds, massive data storage, and financial-grade high availability. This provides a cost-effective and high-performance NoSQL solution.
How it works
PolarDB for PostgreSQL achieves compatibility with the DynamoDB API through the interaction between the database proxy layer and compute nodes. Understanding how it works can help you make better use of this feature.
Database proxy (Proxy) The proxy layer is key to protocol compatibility. It accepts DynamoDB HTTP requests from clients and performs authentication, protocol parsing, and request routing. The proxy then converts DynamoDB operations into JSONB query statements that PostgreSQL can process and intelligently routes them to the most appropriate backend compute node. This design completely decouples the compatibility logic from the database kernel. This ensures the underlying PostgreSQL kernel remains clean and stable, while allowing you to reuse PolarDB's connection pool, load balancing, and read/write splitting capabilities.
Compute nodes
The compute nodes feature a high-availability architecture with one primary node and multiple read-only nodes (up to 15 for Enterprise Edition and 7 for Standard Edition). The primary node handles all write operations, while the read-only nodes handle read requests. The compute layer is based on the standard PostgreSQL kernel and uses only built-in plugins to adapt to subtle DynamoDB differences in areas like collation rules, operators, and data types. This minimal-modification design ensures the feature is maintainable and can quickly keep pace with major releases from the PostgreSQL community. The database abstracts away the traditional concepts of databases and schemas, providing a table management experience consistent with DynamoDB.
Advantages
Seamless migration: Highly compatible with the DynamoDB API, letting you connect existing applications directly without refactoring code.
Unified data platform: Access the same data with both NoSQL for fast operations and standard SQL for complex analytics.
Elasticity in seconds and cost-effectiveness: Compute and storage are decoupled, and storage is pay-as-you-go. Compute nodes can be scaled up or down in minutes, delivering significant cost savings.
Enterprise-grade reliability: Built on the high-availability architecture of PolarDB to ensure data security and business continuity.
Requirements
Before using the DynamoDB compatibility feature, ensure your cluster meets the following requirements:
Kernel version:
PostgreSQL 16, with a minor kernel version of 2.0.16.13.16.0 or later.
Database proxy version: 2.3.63.1 or later.
Go to the PolarDB console and check your cluster's version on the page. If the version does not meet the requirements, upgrade the minor version.
Billing
The DynamoDB compatibility feature is available at no additional cost. You are only charged the related fees for your PolarDB for PostgreSQL cluster.
Next steps
Get started: After reviewing the feature overview, follow the Usage Instructions to create a cluster, configure an account, and verify your connection.
Compatibility and limitations: This feature is highly compatible with the DynamoDB API, but some behavioral differences and limitations apply. For example, some deprecated parameters are not supported, and its error handling for failed transactions differs from that of native DynamoDB. For more information, see Compatibility and Limitations.
(Optional) Data migration: If you need to migrate data from an existing Amazon DynamoDB instance, see the Data Migration Guide for detailed steps and tool usage.
Performance testing
PolarDB has benchmarked this feature using the industry-standard Yahoo! Cloud Serving Benchmark (YCSB) tool. The results show stable and excellent performance across different data volumes and read/write models. For details, see the Performance Test Report.