Hybrid Data Pipeline supports JDBC, ODBC, and OData connectivity. To connect to Hybrid Data Pipeline data sources with JDBC and ODBC client applications, the JDBC and ODBC drivers must be used. OData connectivity is handled by an OData layer within Hybrid Data Pipeline, and therefore does not require the use of a separate component.

Note: A Hybrid Data Pipeline data source must be enabled for OData connectivity. When creating or modifying a data source in the Web UI, you may enable and configure OData connectivity via the OData tab. For details, refer to How to create a data source in the Web UI and Configuring data sources for OData connectivity and working with data source groups in the User's Guide.

JDBC and ODBC connectivity

Initial JDBC or ODBC driver configuration takes place during driver installation. Each driver must be installed using the configuration files generated during the deployment of the Hybrid Data Pipeline server. The configuration files provide required configuration information for driver operation such as a truststore for SSL encryption and the hostname of the host machine or load balancer.

OData connectivity

OData applications must be configured to use the hostname of the load balancer or host machine for HTTP or HTTPS requests. Additionally, to optimize performance in cluster deployments, OData applications should be configured to echo cookies for session affinity. OData applications should also be configured to use SSL to encrypt communications between the client and the server. See SSL configuration for details.

When it is not possible to configure an OData client to echo cookies, Hybrid Data Pipeline uses an internal mechanism called the distributed file persistence manager. When a query is executed that requires file persistence, execution results are stored temporarily on the node that initially received the query. The manager associates the query with the node and the execution results stored there. If a request from the same query is routed to a different node in the cluster, the manager obtains the persisted execution results from the original node. The query results are then returned to the client by the node that received the request.

The distributed file persistence manager requires node-to-node communication using the HTTP protocol to achieve session affinity. The Internal API Port specified during Hybrid Data Pipeline deployment is the port used for this node-to-node communication. Data remains secure in the following respects. First, the Internal API Port is not exposed externally to the public facing network. Each node registers itself using this port, and communications are restricted. Second, a UUID is generated during the node registration process. This UUID is passed in as an HTTP header to confirm the validity of node-to-node communications. Third, the Hybrid Data Pipeline service stores persisted files on only a temporary basis. Moreover, these communications can be encrypted by enabling server-side SSL.