IBM: Volleying data at the U.S. Open

As IBM closes out its 16th U.S. Open on Sunday it'll take away a few lessons in service oriented architecture and virtualization.

On Thursday, I toured IBM's U.S. Open operations in Flushing Meadow, Queens. I've been on a few of these tours and what's striking about this version is that the footprint IBM takes up is smaller than previous years. One reason: Virtualization has cut the number of servers needed from 60 to 9.

Other takeaways from the U.S. Open (more detail is in the gallery at right):

• SOA is playing a bigger role. IBM runs the technology for all four tennis grand slams so reusable code is critical. Mashups are also easier with SOA.
• Data splicing is more important than the data itself. IBM collects a massive amount of data but it has to be packaged in a format that's useful.
• IBM is using Tivoli software for virtualization and systems management. If you asked the average IT guy about virtualization the first word out of his mouth would be VMware. Tivoli may be (or become) a big virtualization player, but it was a bit of a surprise to me. My hunch is it's a surprise to others too.