Hybrid Data Pipeline can be deployed on a single-node without a load balancer, or on one or more nodes with a load balancer.

Important: There is no migration path from a deployment that does not use a load balancer to one that does. Therefore, a non-load balancer deployment is not recommended for scenarios where scaling up the service may be desired. In contrast, a load balancer deployment allows you to scale up the service by adding nodes behind the load balancer.

If you are deploying Hybrid Data Pipeline on more than one node behind a load balancer, you must install Hybrid Data Pipeline on each node in the cluster. During the initial installation of the server, the Hybrid Data Pipeline installer writes properties files, encryption keys, and system information to the key location. When performing subsequent installations on additional nodes, you will need to specify this key location. The installer uses the files written to the key location during the initial installation to install the server on additional nodes. Hence, for installations on additional nodes, the installer bypasses prompts for configurations such as load balancing, SSL, external database, and On-Premises Connector.

If you are using the silent installation process to deploy Hybrid Data Pipeline on more than one node behind a load balancer, the response file generated during the initial installation must be modified to install the server on any additional nodes. For a GUI generated response file, you will need to designate the name of an additional node with the D2C_HOSTNAME option. For a console generated response file, you will need to designate the name of an additional node with the D2C_HOSTNAME_CONSOLE option. Also, note that environment variables are supported for specifying system database and server credentials. See Silent installation process for details.

Note: Before proceeding with an installation of the Hybrid Data Pipeline server, you must copy the product file to a temporary directory, for example, /tmp.