June 08, 2010

Ten Reasons to Upgrade to DB2 10

When I attended IDUG NA 2010, there was, naturally, considerable buzz about the upcoming DB2 release. While there's always hype surrounding these key releases, I think many folks at IDUG were genuinely excited about DB2 10.

The general theme for DB2 10 is improved performance, scalability, availability, security and productivity. Of course I'll be dwelling on specifics in the coming weeks and months, but to give you a taste now, I've come up with a list of 10 reasons to upgrade to DB2 10. Why 10? I know it's not original, but hey, this is DB2 10, so it only seems appropriate. Besides, companies generally don't upgrade for any one reason -- usually a multitude of factors come into play. And DB2 10 does give you many good reasons to step up. So here's my list:

1. CPU reductions for transactions, queries and batch jobs: While improvements to the DB2 10 code itself should significantly reduce CPU usage, IBM has an entire box full of new features and functions to help with performance. More on those later in the list.
2. Scalability in the number of concurrent connections on a subsystem: DB2 10 now supports 10 times more users (up to 20,000 concurrent threads). There's that number 10 again.
3. Improved availability through improved concurrency for the catalog, utilities and SQL.
4. Improved availability through enhancements to online schema changes: This means you'll be able to make changes to indexes and table spaces without having to unload and reload the data. With the catalog redesign, links are removed in the catalog and directory. In addition, new lock avoidance techniques reduce the time a lock is held and prevent writers from blocking readers of data.
5. Security is enhanced with better granularity for administrative privileges, data masking and new audit capabilities.
6. Improved portability brought about by SQL enhancements.
7. pureXML: With these enhancements, companies will be better equipped to adapt and respond to changing business needs.
8. Temporal or versioned data.
9. Hash, index include columns, access path stability, skip migration are all designed to reduce CPU consumption.
10. Productivity improvements for database and systems administrators, and application programmers.

As mentioned, I'll dig into these and many other new features in future posts. If there's something you'd like to see covered, contact me or leave your request in Comments.