- Started off by defining BI and DW. I like the Wikipedia definitions myself. Also discussed what Master Data Management is; i.e., managing data that has become aged and may or may not be needed as much as current data. (The company I currently work for doesn't really think of what I do as BI and DW, but developing Decision Support Systems (DSS) is really at the crux of what I do here.)
- Two polls on where participants are in implementing BI and DW in their organizations. Most are either currently using or evaluating, versus implementing.
- Discussed scope of data definition and data modeling that should be done before getting into BI/DW. One caller will be building theirs from scratch versus buying something that was developed for someone else in their industry. (I wonder if different BI solutions have been developed for different industries.) Data cleansing is so vital when doing modeling. (Amen to that!)
- Discussed risks in software selection and implementation. Licensing brought up as part of risk. Making sure that appropriate number of licenses (user vs site) important. How solution builds report cubes (pre-calculated or built at run-time) can affect performance and user experience. sBase(?) mentioned.
- Practical support from vendors also brought up as consideration. Everything works great at the demo, but on the implementation there could be issues. Ask for references. NOREX mentioned that's part of the service they provide for members.
- Discussed tools that are used, and pros and cons of each. One member mentioned Sybase IQ. Others are using Business Objects, SQL Server Reporting Services, and Oracle tools. Crystal Reports also mentioned.
- Someone asked for experience with the Obvient strategies tool. No one on the call mentioned having any experience.
- Any local government entities using data mining tools? No one on the call had any comments. Topic to be passed on to the government members outside the call.
- SQL Server 2005 Analysis Services - how is it being used, other considerations. Has a lot of capability and interacts well with Visual Studio. Another member, however, saw a demo at Microsoft- some problems with creating a cube and error messages hard to decipher.
- Discussed how to implement SQL Server Reporting Services through Sharepoint. (We use Windows Sharepoint Services as our company's intranet and are looking at how to do this.) No one on the call had any expertise to offer.
- Discussed whether anyone uses Excel 2007 as a BI client. Some of the new functionality of Excel lends itself to an entry-level BI implementation. Some accountants in one organization using Excel to manipulate data from their other BI system. No one taking advantage of the BI capability yet. Concerns raised about quality of data, security, and versions.
- Poll on whether anyone using SAS to access Star Schema data: dimensional modeling vs entity-relationship modeling. (Not really applicable to us in our organization.)
- Release management best practices from Dev and Test to Production environments sought. How to manage things that are not necessarily files that can be version-controlled? Really speaking about reports and dashboard components.
- One member mentioned his process is such that a developer will create a report and work from it at his desktop. Then when it's ready for deployment, he will publish it.
- Other members mentioned a more rigid change management structure with development, QA, and production environments. (I guess it depends on the specific tools and systems being used.)
Check out my other blogs:
Daniel Johnson, Jr.
Get That Job!
Journey Inside My Mind Blog
Journey Inside My Mind Podcast
Connect on LinkedIn
Interesting Things I've Read
Related tags: daniel+johnson+jr