Oracle Globally Distributed Database provides the capability to automatically deploy the distributed database, which includes both the shards and the replicas.

The distributed database administrator defines the topology (regions, shard hosts, replication technology) and invokes the DEPLOY command with a declarative specification using the GDSCTL command-line interface.

Before You Begin

There are many different configurations and topologies that can be used for a distributed database. Depending on your application’s particular architecture and system requirements, you may have several choices from which to choose when designing your system. Familiarize yourself with Planning Your Deployment before proceeding with deployment.

Distributed Database Deployment Road map

At a high level, the deployment steps are:

1. Set up the components.
   • Provision and configure the hosts that will be needed for the distributed database configuration and topology selected (see Provision and Configure Hosts and Operating Systems).
   • Install Oracle Database software on the selected catalog and shard nodes (see Install the Oracle Database Software).
   • Install global service manager (GSM) software on the shard director nodes (see Install the Shard Director Software).
2. Create databases needed to store the distributed database metadata and the application data.
   • Create a database that will become the shard catalog along with any desired replicas for disaster recovery (DR) and high availability (HA) (see Create the Shard Catalog Database).
   • Create databases that will become the shards in the configuration including any standby databases needed for DR and HA (see Create the Shard Databases).
3. Specify the distributed database topology using some or all the following commands from the GDSCTL command line utility, among others (see Configure the Distributed Database Topology).
   • CREATE SHARDCATALOG
   • ADD GSM
   • START GSM
   • ADD SHARDGROUP
   • ADD SHARD
   • ADD INVITEDNODE
4. Run DEPLOY to deploy the distributed database topology configuration (see Deploy the Configuration).
5. Add the global services needed to access any shard in the distributed database (see Create and Start Global Database Services).
6. Verify the status of each shard (see Verify Shard Status).

When the distributed database configuration deployment is complete and successful, you can create the sharded schema objects needed for your application. See Schema Objects.

The topics that follow describe each of the deployment tasks in more detail along with specific requirements for various components in the system. These topics can act as a reference for the set up and configuration of each particular step in the process. However, by themselves, they will not produce a fully functional distributed database configuration since they do not implement a complete distributed database scenario, but only provide the requirements for each step.

Example Distributed Database Deployment walks you through a specific deployment scenario of a representative reference configurations. This section provides examples of every command needed to produce a fully functional distributed database once all the steps are completed.