Amazon RDS for Aurora vs TiDB: What are the differences?

Introduction

Amazon RDS for Aurora and TiDB are both popular relational database management systems used for different purposes. While they share some similarities, they also have key differences that make them suitable for different use cases. This article aims to highlight six major differences between Amazon RDS for Aurora and TiDB.

1. Architecture:

   • Amazon RDS for Aurora: It is based on a distributed storage system with separate instances for storage and compute. Data is stored in a distributed manner across multiple availability zones for high availability and fault tolerance. It uses a replication protocol called Aurora Multi-Master, which allows for read and write scalability.

   • TiDB: It is a distributed NewSQL database that combines the scalability of NoSQL systems with the ACID guarantees of traditional SQL databases. TiDB uses a distributed transactional key-value storage engine and supports horizontal scalability by adding more nodes to the cluster.

2. Consistency Model:

   • Amazon RDS for Aurora: It provides strong consistency by default, ensuring that all read operations return the most recent committed data. It uses a distributed consensus protocol to achieve high availability and fault tolerance.

   • TiDB: It provides snapshot isolation, which allows for a higher degree of concurrency. TiDB uses a distributed transaction protocol called Two-Phase Commit (2PC) to ensure consistency across distributed transactions.

3. Compatibility with MySQL:

   • Amazon RDS for Aurora: It is compatible with MySQL, which means it supports the MySQL wire protocol, SQL dialect, and most MySQL features. This makes it easier to migrate existing MySQL applications to Aurora without making major code changes.

   • TiDB: It is also compatible with MySQL, supporting the MySQL wire protocol and SQL dialect. However, not all MySQL features are fully supported in TiDB, so some code modifications may be required when migrating from MySQL.

4. Scaling:

   • Amazon RDS for Aurora: It supports vertical scaling by increasing the instance size or storage capacity. It also supports read scaling by creating read replicas that offload read traffic from the primary instance.

   • TiDB: It supports both vertical and horizontal scaling. Vertical scaling can be achieved by increasing the resources allocated to each node, while horizontal scaling is achieved by adding more nodes to the cluster.

5. Data Sharding:

   • Amazon RDS for Aurora: It automatically shards data across multiple storage instances based on a user-defined shard key. This allows for horizontal scalability and improved performance by distributing the data across multiple nodes.

   • TiDB: It also supports data sharding, but it does not require a user-defined shard key. Instead, it uses a distributed hash index to automatically shard data across multiple nodes. This simplifies the sharding process and eliminates the need for manual configuration.

6. Query Optimization:

   • Amazon RDS for Aurora: It uses an advanced query optimizer that leverages statistics and query hints to generate efficient query execution plans. It also supports indexing and query cache to improve query performance.

   • TiDB: It uses a cost-based optimizer that takes into account the statistical information of the data to generate optimal query execution plans. TiDB also supports indexing and provides various optimization techniques to improve query performance.

In summary, the key differences between Amazon RDS for Aurora and TiDB lie in their architecture, consistency model, compatibility with MySQL, scaling capabilities, data sharding approach, and query optimization techniques.