Hi, I’m trying to understand the concept of Read Replicas in Cloud SQL. In fact, in the video course, the instructor said that the read replicas limited to the same region as the master. But I found that was not the case because when I create a Read Replicas, Cloud SQL allows me to choose the different region for it. So how can we explain that, please? Many thank.
1 Answers
Hi there,
The purpose of a read replica is to scale the use of a database without affecting its performance or needing to resize it. A read replica can handle additional reads from clients, thereby distributing the load. Note that you can’t send a write to a read-replica server.
I’m not sure but I believe cross-region read replicas (ie. replicas in different regions) may have been introduced since that lab was created. These are useful for cross-region redundancy, or for migrating data between regions.
You can still just have an in-region replica (ie. in the same region) to assist with performance and scale.
Hope this helps!
Hi Tim, yes, I’ve checked that too, the cross-region read replicas was introduced in GCP on June 3, 2020, so that explains why we can do this. Thank you !!!