DATA WAREHOUSING BASICS

Some of you probably are wondering where the operational data store (ODS) fits in our warehouse components diagram. Since there’s no single universal definition for the ODS, if and where it belongs depend on your situation. ODSs are frequently updated, somewhat integrated copies of operational data. The frequency of update and degree of integration of an ODS vary based on the specific requirements. In any case, the O is the operative letter in the ODS acronym.

Most commonly, an ODS is implemented to deliver operational reporting, especially when neither the legacy nor more modern on-line transaction processing (OLTP) systems provide adequate operational reports. These reports are characterized by a limited set of fixed queries that can be hard-wired in a reporting application. The reports address the organization’s more tactical decision-making requirements. Performance-enhancing aggregations, significant historical time series, and extensive descriptive attribution are specifically excluded from the ODS. The ODS as a reporting instance may be a steppingstone to feed operational data into the warehouse.


In other cases, ODSs are built to support real-time interactions, especially in customer relationship management (CRM) applications such as accessing your travel itinerary on a Web site or your service history when you call into customer support. The traditional data warehouse typically is not in a position to support the demand for near-real-time data or immediate response times. Similar to the operational reporting scenario, data inquiries to support these real-time interactions have a fixed structure. Interestingly, this type of ODS sometimes leverages information from the data warehouse, such as a customer service call center application that uses customer behavioral information from the data warehouse to precalculate propensity scores and store them in the ODS.


In either scenario, the ODS can be either a third physical system sitting between the operational systems and the data warehouse or a specially administered hot partition of the data warehouse itself. Every organization obviously needs operational systems. Likewise, every organization would benefit from a data warehouse. The same cannot be said about a physically distinct ODS unless the other two systems cannot answer your immediate operational questions. Clearly, you shouldn’t allocate resources to construct a third physical system unless your business needs cannot be supported by either the operational datacollection system or the data warehouse. For these reasons, we believe that the trend in data warehouse design is to deliver the ODS as a specially administered portion of the conventional data warehouse.

Finally, before we leave this topic, some have defined the ODS to mean the place in the data warehouse where we store granular atomic data. We believe that this detailed data should be considered a natural part of the data warehouse’s presentation area and not a separate entity.