Replication

What is replication?

Replication is the process of copying data from one database server (primary) to another.

A common use case for replication is to reduce the load on the primary server by having the replicas also accept requests and serve the same data. Other scenarios include:

  • a standby replica taking over when the primary fails or becomes unavailable
  • data warehousing and analytics which don’t require the production/primary database as the source

A read replica is a read-only instance. Read replicas are suitable for offloading read queries from the primary, and for analytics and reporting.

Crunchy Bridge read replicas

You can create read replicas for each Crunchy Bridge instance. The current limit is 5 replicas per primary.

Replication in Crunchy Bridge is done with Write-Ahead Log Shipping. Replicas can be set up to run read-only queries with Hot Standby.

Replication to read-only instances is asynchronous, so you should expect a little bit of lag between the primary and replicas.

Create a replica

Users with permission to provision instances (i.e. administrators) will also be able to create read replicas.

To create a new replica, select a cluster, and navigate to the “replicas” tab. Click “Create read replica”.

Bridge UI with Create read replica button

1. Select a plan

The replica will be created with the same infrastructure provider, in the same region, with the same storage capacity.

You can select the same plan, or move across plans (i.e. downgrade to a lower one or upgrade to a higher one from the primary). The corresponding price per month will be displayed with the list of available plans:

List of available plans and pricing for new replica

You can choose the same plan for the replica, or downgrade or upgrade.

For example, if you are expecting to run intensive analytical queries, you might consider upgrading. If you expect fewer connections or read requests on the replica, you might consider downgrading.

2. Provide a cluster name

The field will prepopulate with <name-of-primary>-replica, but you can rename the cluster as desired.

3. Deploy

After hitting “Deploy,” your cluster should be ready within a few minutes.

Connect to replica

The replica will contain the same data as the primary, so you can connect to the replica in the same way as you would connect to a primary instance.

To get the connection info for the replica, click the icon under “Details”. A “View Replica” window will open that provides the following:

  • Name
  • Provider
  • Region
  • Connection string

Delete replica

To delete a replica, select the primary instance from the dashboard, and select the replica. Click the “Delete” icon. You will be prompted by another window; click “Delete” to confirm.

Billing rates and charges (based on usage) still apply for deleted replicas.