October 2008 Archives

Kamal Hathi from Analysis Services/Gemini and Mark Theissen from DATAllegro, visited the BBBT from the Microsoft SQL Server group. They covered the SQL Server roadmap, Project Madison, and Project Gemini. We first unscrambled the name cloud around the Kilimanjaro release of SQL Server, the next normal release (vNext), Project Gemini, and Project Madison. See previous blogs for details.

We discussed six referenceable customers (Premier Bankcard, USDA, HMC, Crossmark, CLALIT, US Veterans Affairs) who have operational data warehouses in the 2 to 15 TB range using the current SMP product. With the DATAllegro technology, the target is to move SQL Server from SMP to MMP architecture and from 10's of terabytes to petabytes. Mark tied the motivation for supporting mixed workloads to cost reduction by reducing surplus processing capacity. Goal is to offer an appliance-like experience for DW customers via reference platforms from major hardware vendors -- one SKU, one pallet, pre-configured, and hopefully one-throat-to-choke. The Microsoft Data Warehousing Product Unit was created to house the DATAllegro folks. The significance of this is that they are assisting in extending all of the SQL Server components into best DW practices.

Unfortunately I missed the discussion on Project Gemini by Kamal to catch a flight. But see the previous blogs.

As a division of TIBCO based in Boston, Spotfire focuses on data analytics with advanced visualizations and complex statistics. I first played with Spotfire in early 1990s when it was a research project at Human-Computer Interaction Lab at the University of Maryland lead by Ben Shneiderman. A colleague of Shneiderman, Christopher Ahlberg continues to lead Spotfire. Presenting are Mark Lorion, Senior Director of Marketing, and Tim Wormus, Analytics Evangelist. In the open spirit of BBBT, we quickly jumped into discussion of market positioning with business users versus IT management for data analytics tools in the enterprise. See us in action.

A traditional characteristic of Spotfire has been their slice/dice interaction of the data. For example, the right panel displays major variables with check boxes for low-cardinality values or sliders with high-cardinality values. Even after twenty years, the viz still have that oooohhhh effect.

Most of our discussion was under NDA, so it is not appropriate to elaborate. However, a major challenge is the balancing business users exploring business data in an open and agile style and IT professionals managing the reliability and integrity of enterprise systems. I highlighted this challenge in blogs on the Microsoft BI Conference.

We then focused on the intersection of BI and CEP (complex event processing) leveraging their TIBCO connection. An example is their Operations Analytics offering announced in April. Exciting!

Operational BI is seeking advanced analytics that operate upon event streams. The gaps are quite apparent between mainstream BI sitting on top of the enterprise business data versus CEP (like TIBCO) sitting on top of the enterprise business processes. Spotfire can act as an integrating component that bridges those gaps. If Spotfire moves beyond the pixels-on-the-screen, its integration value will be based upon consuming data from and generating data to the BI infrastructure. In the long term, my feeling is that the penetration of Spotfire into the enterprise will depend on its synergism with traditional BI tools like Cognos and Business Objects and with the enterprise data warehouses like Teradata and IBM.