Continue in 2 seconds

An operational data store (ODS) is different from data warehouse, but it is one part of data warehouse. Do you agree?

By
  • Ross Armstrong, Chuck Kelley, Clay Rehm
Published
  • September 01 2000, 1:00am EDT

Q:  

An operational data store (ODS) is different from data warehouse, but it is one part of data warehouse. Do you agree? ODS is normally "home grown" - do you have any idea about the cost comparing with the off-the-shelf solution? What's the future of ODS.

A:  

Ross Armstrong’s Answer: Ah, the rascal of the data warehouse architecture. The famous ODS, where does it fit, where does one buy it and what can it do?

Well, I like the ODS. It is part of the data warehouse architecture. It is the first stop for the data on its way to the warehouse. It is here where I can collect and integrate the data. I can ensure its accuracy and completeness. In many implementations, all transformations can't be completed until a full set of data is available. If the data rate is high, I can capture it without constantly changing the data in the warehouse. I can even supply some analysis of near current data to those impatient business analysts. The ODS has a life in many implementations, although Ralph Kimball may not agree in WebHousing.

Clay Rehm’s Answer: I agree that an ODS is a component of the data warehouse environment. However, not every DW needs an ODS. Typically the ODS is a normalized structure that integrates data based on a subject area, not a specific application or applications. For example, one client site I worked at had 60+ premium applications. A "premium" subject area ODS was created which collected data using a feed from each application to provide near real- time premium data across the enterprise. This ODS was refreshed to stay current, and the history was sent to the data warehouse. The ODS satisfies the tactical/operational reporting needs. I recommend you read the book by Bill Inmon and Claudia Imhoff called "Building the Operational Data Store."

Chuck Kelley’s Answer: ODS is yet another operational system. I believe that it will be a great source system for the data warehouse. I do not know of any off-the-shelf solutions, since they are mostly used to tie multiple source systems into a single system for tactical analysis for short-term, volatile data. For example, if you were running a promotion, you might want to keep track of how a particular item is selling compared to last year, or the last two years. This could influence future purchases for the current promotion. Thus, I think there is a great future for the ODS.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access