Creating a destination database cluster

You can use existing databases (EDB Postgres AI Hybrid Manager (HM)-managed or self-managed) as destinations for migrations. To use an existing instance as a destination for the migration, ensure the tables you migrate and the load generated on the destination don't interfere with existing workloads.

If you don't have a destination database yet, this page shows you how to create a new database, either on an HM-managed cluster or on a self-managed database.

• Create a new database cluster in HM

OR

• Create a new database on a self-managed instance

Create a new database cluster in HM

If you don't have a destination database cluster on the HM yet, create one that the HM can use as destination for migrations:

1. In the HM Portal, select the project where you want to create the database cluster.

   See Creating a project if you want to create one.

2. Within your project, select Create New and Database Cluster to create an instance that will serve as destination for the EDB Data Migration Service (EDB DMS).

   See Creating a cluster for detailed instructions on how to create a single-node or a primary/standby high availability cluster.

   See Creating a distributed high-availability cluster for detailed instructions on how to create a distributed high availibility cluster.

3. In the Clusters page, select your cluster, and use the Quick Connect option to access your instance from your terminal.

4. Create a new empty database that you will use as a destination for the migration. Alternatively, you can use the default database edb_admin.

Create a new database on a self-managed instance

If you don't have a self-managed destination database yet, create one that the EDB DMS Writer can connect to for HM migrations. This example assumes you have configured your self-managed Postgres instance already.

1. Connect to the destination Postgres server:

   → WrapCopy

   psql -p <PG_PORT> -U <PG_USERNAME>

2. Create an empty destination database that you'll use as a destination for the migration.

   In this example, the destination database is called tr_target:

   → WrapCopy

   CREATE DATABASE tr_target;