Auto-sharding in Aurora for MySQL / PostgreSQL

Hello AWS Customer,

Hope you are in good health.

After going through the post, I understand that you would like to know if auto-sharding is supported in Aurora for MySQL/PostgreSQl .

Please correct me if I misunderstood your query.

Answering your Query :

1. Is there any plan to support auto-sharding in Aurora for MySQL / PostgreSQL ? (similar to Citus for PostgreSQL)

Ans : Since sharding is not provided as a managed service from the AWS RDS side, hence, a routing logic needs to be built at the application tier that decides which instance should it connect to based on the partition key.

Please note that Aurora sharding would be like sharding on a traditional MySQL db but the architecture of the Aurora would be different in the way Aurora storage is designed.

Please refer to this document which gives information about sharding [1] . In this post, I describe how to use Amazon RDS to implement a sharded database architecture to achieve high scalability, high availability, and fault tolerance for data storage. I discuss considerations for schema design and monitoring metrics when deploying Amazon RDS as a database shard. I also outline the challenges for resharding and highlight the push-button scale-up and scale-out solutions in Amazon RDS.

If you choose an Aurora DB cluster to build a database shard, you can also achieve high availability by configuring a read replica with the primary instance [2].

For Aurora multi-master clusters, your application typically uses the DB instance endpoints for all or nearly all connections. You decide which DB instance to use for each SQL statement based on the mapping between your shards and the DB instances in the cluster. Each DB instance has one such endpoint. Thus the multi-master cluster has one or more of these endpoints, and the number changes as DB instances are added to or removed from a multi-master cluster. Please also refer to other resources that may help with the configuration[3].

I recommend testing any changes in a non production environment first before applying it in your production environment.

As for supporting the Auto-sharding on Aurora ,Unfortunately, we can not provide you with an ETA or confirmation at the moment on when or how this feature will be available . I request you to keep checking the new feature release updates .

The new features release updates are available regularly on the official AWS Updates sources, Please be on the lookout using the links attached below for your reference.

RDS What's New - https://aws.amazon.com/new/
AWS Database Blog - https://aws.amazon.com/blogs/database/
AWS Blog - https://aws.amazon.com/blogs/aws/

I really hope the above information serves you well. Meanwhile, if you have any questions or concerns please feel free to write back to me. I’ll be happy to assist you.

---

References:

[1]Sharding with Amazon Relational Database Service - https://aws.amazon.com/blogs/database/sharding-with-amazon-relational-database-service/
[2]Amazon Aurora connection management - https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/Aurora.Overview.Endpoints.html
[3]Build highly available MySQL applications using Amazon Aurora Multi-Master - https://aws.amazon.com/blogs/database/building-highly-available-mysql-applications-using-amazon-aurora-mmsr/

A sharding-related feature "Aurora Limitless" is currently in preview for Aurora PostgreSQL:

Announcement: https://aws.amazon.com/about-aws/whats-new/2023/11/amazon-aurora-limitless-database/

Blog post: https://aws.amazon.com/blogs/aws/join-the-preview-amazon-aurora-limitless-database/