Amazon Relational Database Service (RDS) now support launching read replica in another region with a few clicks. THIS IS HUGE!!! This is the first high level feature that AWS supports across regions. This is the great first step from AWS towards providing better support for cross-region infrastructure setup.
Until now replicas were only allowed in the same Region. And as RDS instances only allow access to the database application and not the server, it was impossible to configure read replicas in another region until now with RDS. With the latest feature, AWS users can, with a few clicks, start read replicas in another region.
This feature is really really important for many use cases:
- Disaster recovery
Although its is very rare for an entire AWS region to go down, it does happen. Many enterprises want to replicate their databases across regions, so that when a catastrophy does occur and the primary region goes down, infrastructure can be quickly setup in another region. Such setup requires database being synced across regions and until now such deployment could not use RDS. Now they can.
- Active-Active cross-region deployments
This use case was for active-active deployments that span across regions. For example, an e-commerce site might want infrastructure in USA and in UK simultaneously to serve clients locally, although the product database could have been the same. This requires database the sync across regions and again RDS was not an option. Now at the very least both locations can have local read-replica copies to read data from. The writes will still have to go to a central location. This setup would work out for for read-heavy applications like e-commerce.
- Data migration
There could be use cases where even though applications and database could be in many regions, there might be periodic need to copy data over to single region for all the other regions. This could be for data aggregation in some central region for analysis. Imagine a big IT company having infrastructure in all the regions, but daily reports that need all the data are generated by aggregation data from all the regions. RDS was not an option until now for such use cases. But now there could be multiple different databases with each having their primary different RDS masters in their own regions and a read replica in a single region where data from all other regions and can processed locally.
There are many more use cases that this setup will allow. What more use cases do you think this feature can enable?? Please share your thoughts!! If you liked the article please do share..