When to Replicate Data with Oracle Streams

If connectivity is not an issue, then organizations might prefer to replicate data in a near-real-time manner. Doing so insures that the data is up to date at all locations as soon as possible. Oracle Streams supports near-real-time data replication in a variety of configurations, depending on an organization's specific requirements. In an Oracle Streams replication environment, databases push changes to each other automatically.

Common uses for Oracle Streams replication include:

• Creating a reporting site to offload processing from a primary online transaction processing (OLTP) site.

• Providing load balancing and improved scalability and availability for a call center or similar application.

• Providing site autonomy between locations to satisfy certain common business requirements.

• Transforming and consolidating data from multiple locations, such as regional offices.

• Replicating data between different platforms and Oracle Database releases, and across a wide area network (WAN).

There are two common types of Oracle Streams replication configurations: n-way and hub-and-spoke. Specifically, a multimaster (or n-way) configuration is frequently used by organizations that must provide scalability and availability of data. Often, these applications use a "follow the sun" model, with replicas located around the globe. For example, an organization might have call centers in the United States, Europe, and Asia, each with a complete copy of the customer data. Customer calls can be routed to the appropriate call center depending on the time of day. Each call center has fast, local access to the data. If a site becomes unavailable for any reason, then transactions can be routed to one of the surviving locations. This type of configuration can also be used to provide load balancing between multiple locations.

Another common configuration is hub-and-spoke. For example, an insurance company might use this configuration to share customer data between its headquarters and local sales offices. A networked version of this configuration can be especially useful in cases of limited connectivity between the end spokes and the hub. Suppose local sales offices have direct connectivity to regional offices, which in turn connect to headquarters, but the local offices have no direct connectivity to headquarters. This type of networked routing can eliminate some of the complexity that results when there are direct connections between all locations. The hub-and-spoke configuration is also useful in data warehousing environments, where detailed data is maintained at each store or spoke, and higher-level data can be shared with the data warehouse or hub.

In both n-way and hub-and-spoke configurations, organizations can configure Oracle Streams replication to allow updates to the replicated data at multiple locations. In such replication environments, data conflicts are possible. Oracle Streams provides conflict resolution methods that can resolve these conflicts automatically.

Because Oracle Streams provides a flexible infrastructure for all information sharing requirements, including messaging and replication, it is easy for an organization to change its configuration as its needs change.