IBM information management

Comment This is the third of my articles derived from IBM’s Software Group analyst conference, in this case focused on the Information Management part of the IBM software portfolio, which is my main interest.

The Information Management group is the only part of the Software Group that is not known by its brand name. It is, of course, most well known as the group based around DB2 but it also includes WebSphere branded products in its Data Integration Suite (previously Ascential) and Information Integrator. Other products within the group include content management, master data management (MDM), AlphaBlox, Intelligent Miner and an increasing emphasis on providing vertical solutions (that is, industry specific data models and MDM solutions) as well as traditional horizontal products.

I have to say that I was somewhat disappointed by the lack of emphasis on Information Management within the conference. There was only passing mention of the Hawk release of the Data Integration Suite and the Serano release of Information Integrator, and there was very little discussion of MDM and its relevance to SOA (service-oriented architecture), which was particularly surprising given that this was the main theme of the conference.

As far as DB2 was concerned the main interest was on the Viper release, which is the next version of the product, which will provide native XML storage. I was disappointed to discover that even some other analysts have not yet appreciated the extent to which this is revolutionary. The product is now in open beta and it will probably be generally released during the first half of next year. Initial users report significant performance benefits compared to conventional approaches to storing XML and it will be interesting to see what new applications emerge as a result.

Also interesting was the reports of initial users of the DB2 for SAP product, which was released earlier this year. This has special features for improving performance for SAP users (both in transactional and data warehousing environments) and in making implementation easier. Early adopters are now starting to report their experiences with this product and, in particular, some of these have converted from Oracle environments. Comparisons suggest a performance improvement of between two and five times across a range of functionality, which should make Oracle sit up and take notice.

However, the most interesting piece of information was with regard to the Information Server product that is due to be released next year. While I have not had a detailed briefing on the product, it has clear advantages in two areas. First, it can sit above MDM, the Data Integration Suite, Information Integrator and other products and allows their services, such as a DataStage process, to be published as a service operation. This is very easy to do: you simply drag a box around the process that you want to wrap and then use the product’s wizard to create the relevant service.

Finally, the second thing that Information Server does is that it will act as a bridge between, say, IBM’s MDM capability and environments such as those of SAP. One of the problems with IBM’s approach to MDM is that, taken in isolation, it would mean that application vendors would need to completely rewrite their applications to take advantage of the MDM layer. This isn’t going to happen, not least because vendors like SAP have their own MDM capabilities. The Information Server resolves (or will resolve—it was not clear whether this will be in the first release) this issue by acting as an interface between the two environments. As I said, I have not been fully briefed on the product.

Copyright © 2005, IT-Analysis.com