I had lunch with an old SQL Server friend of mine today. I say "old" because not only are we getting a bit long in the tooth, but also our SQL Server days started back in the Sybase and OS/2 world. We've seen a lot of changes in the SQL Server space. We were chatting about old-school DBAs and how the DBA role has changed, and I stated that extraction, transformation, and loading (ETL) has always been, and always will be, a skill set that's valuable in both the transaction and analytical database worlds. But my more recent observation is that many of the "best of the best" SQL Server Integration Services (SSIS) experts that I know seem to be more aligned with the OLAP side of the database world than the OLTP side.
Certainly a lot of Microsoft's marketing material and product positioning tends to have SSIS aligned with the business intelligence (BI) stack—not that Microsoft would come out and say that SSIS is a BI-centric tool. Perhaps it's just the circle that I run in these days, but the folks that I tend to think of as "relational" generally don't seem to have the same breadth and depth of SSIS expertise as the people that I think of as, and who probably define themselves as, BI professionals.
Am I right or wrong? Does the current crop of SSIS junkies tend to have an affinity for BI? If so, why don't more old-school relational DBAs use SSIS in elegant and interesting ways?
I did a really simple Google search using the keywords SSIS and OLTP and got just 11,500 hits compared with the 96,200 hits I got when I used the keywords SSIS and OLAP. No, that's not a scientific study of any kind, but it's an interesting number. Does that mean there's almost nine times as much content out there related to SSIS for OLAP people than SSIS for OLTP people? If that's true, does it somehow correlate to the depth and breadth of SSIS expertise in the OLAP and OLTP communities?
Old-school DBAs, BI jocks, and SSIS junkies speak your minds. Will the real SSIS expert stand up? What do you think?