Blogs

Long live the data warehouse appliance

Post Comment
Content Marketing Manager, IBM Analytics Platform, IBM

Upon reading his own obituary in the newspaper, famed author Mark Twain is said to have remarked that reports of his death were greatly exaggerated. I can only imagine that if the data warehouse appliance were a 19th century American novelist, it might say the same thing. For a while now, declarations have been made that the data warehouse is dead, even as organizations continue to use it to good effect.

In this way, the data warehouse joins an elite club. Premature proclamations of death have a long and storied history both in the press and as a comedic device in entertainment. What each example has in common is that the announcement is born out of erroneous information, expedient hyperbole, or a combination of both. And therein lies the problem – these rumors must be built around a grain of truth or at the very least sound plausible. For well-known personalities, a simple visit to the hospital might be enough to spur rumors. In the case of data warehouses, perhaps the amount of historical or current change that has been experienced is the catalyst.

As Claudia Imhoff, Ph.D. acknowledges in her recent paper “Data Warehouse Appliances and the New World Order of Analytics,” analytics architecture has changed throughout its history. Most recently, “the Extended Data Warehouse architecture (XDW) has taken the place of the single repository idea.” However, this does not imply that the data warehouse is obsolete. On the contrary, it implies that the data warehouse is valuable in the context of a larger structure.

In that same paper, Dr. Imhoff listed four components of an Extended Data Warehouse: the Enterprise Data Warehouse (EDW), the Investigative Computing Platform (or Data Lake), the Data Refinery, and a Real-Time Analysis Platform. It is important to note that while the Enterprise Data Warehouse is no longer the sole component, it is a very important part. Moreover, Imhoff goes on to say that data warehouse appliances “are perfectly suited for both the EDW and the investigative computing platform” portions of this architecture. Far from being dead, then, this reveals that the data warehouse plays a significant role within current architectures.

Of course, data warehouse appliances won’t be the right choice for every situation, which is why careful analysis is important. For more information on the Extended Data Warehouse architecture as well as the benefits and considerations related to data warehouse appliances, read the full paper “Data Warehouse Appliances and the New World Order of Analytics” and check the pulse of the data warehouse for yourself.