The third day of the Gold Briefing, and opening day for IOD, is in the history books and there is no shortage of new announcements and interesting tit-bits to share. The opening session was the usual mix of high level overviews and interesting technical information accompanied by a little light entertainment (courtesy of Terry Fator, the very gifted winner of “America’s Got Talent”). Overall attendance here looks pretty good, and I’d say the numbers are pretty much the same as last year but with more users/customers and less IBM employees.
Yesterday I talked about IBM’s exciting new pureScale technology for the DB2 for LUW platform. Before I move on to talk about some interesting mainframe related technologies as I promised, I just want to briefly mention a couple of extra pureScale items that became apparent yesterday when IBM issued the formal announcement for the technology (see announcement letter ENUSZP09-0487). The first is a firm availability date: December 11th 2009 for electronic media delivery. There has also been some confusion over the version of DB2 for LUW that pureScale will work with: it requires a DB2 for LUW 9.7 Server Edition licence, and when installed the database will report its version as being 9.8.
And now a quick look at what’s been happening on the mainframe side. More and more details are being made available regarding the next release of DB2 for z/OS (cunningly codenamed “DB2 X” as a way of concealing the real version number from those that don’t understand Roman numerals). I’ll cover the contents of this release in more depth in future blog entries and will also produce a new version of the “Business Value” white paper series to assist with upgrade justification and planning. In the meantime, the headline features that most people seem to be most excited about here at IOD are the anticipated CPU savings (potentially helping to claw back some or all of the performance regression that may have been experienced in the move to 64-bit at Version 8) and the new temporal data support (which allows DB2 to store historical versions of a row and enables the developer to select data as of a given timestamp and see what the row looked like at that time).
There’s a major push on Analytics here, as part of the Information-Led Transformation strategy I talked about a couple of days ago. The mainframe is not immune to this, and in addition to some nice extensions to the core database engine in DB2 X to support warehousing queries, IBM has also begun to talk of a much more fundamental and exciting development in the shape of its “Smart Analytics Optimizer”. This is essentially a BI/analytics appliance aimed at boosting the database query performance of the servers that it’s attached to. The big news is that this will be available to connect to System z servers very soon, allowing DB2 for z/OS to offload suitable workload to the SAO for processing (with all of the attendant performance and cost advantages). I hope to be in a position to talk more about this exciting new development as more details are announced.
That’s it for today. Tomorrow is the last day of the Gold Briefing and I’ll be heading home, but I’ll try to find time for a last Gold/IOD post to cover the day’s news and draw a few conclusions on the whole event.