Getting the Skinny in Philly: What’s New in DB2 for z/OS?

This week I had the opportunity to attend a DB2 Gold Consultant’s Briefing, which ran back-to-back with the IDUG North America Conference in Philadelphia: a grand total of 6 days of in-depth technical presentations and a great excuse to catch up with many friends and colleagues in the DB2 world.

So, what did I learn in the city that prides itself on being the cradle of  US democracy? Way too much to fully cover in one blog post, but here are some highlights from a great week:

• On Monday the legendary Roger Miller briefly put his well-earned retirement on hold to attend the conference and help celebrate DB2’s 35th birthday as part of a DB2 for z/OS Spotlight session alongside new DB2 Labs Director Mo Townsend and DB2 Distinguished Engineer Jeff Josten. An entertaining and informative tour of DB2 past, present and future that really put into perspective how far DB2 has come (and how much is there is yet to do!)
• I attended several interesting sessions on IBM’s Analytics Accelerator technology (previously referred to as IDAA). This started out as a highly optimised hardware solution for efficiently running DB2 for z/OS analytics queries against large amounts of data (based on the technology obtained as part of the Netezza acquisition back in 2010). In October 2017 IBM announced Analytics Accelerator V7.1, but don’t be fooled by the standard version numbering because this is more than just an incremental upgrade: the product has been fundamentally re-architected and is now based on the Blu column-store engine that was originally developed for DB2 for LUW. The move from hardware to software-based acceleration opens up some interesting new possibilities, so from V7.1 onwards the Analytics Accelerator can continue to run on a dedicated hardware appliance (now known as the IBM Integrated Analytics System, or IIAS) or directly on an IBM z14 (using a dedicated Secure Service Container (SSC) LPAR and Intergrated Facility for Linux (IFL) processors). The move to the Blu engine has some interesting implications that I’ll cover in a future blog entry.
• Machine Learning was also a common theme at the conference. In addition to some interesting sessions covering the internals and use cases of IBM’s Machine Learning for z/OS product, it’s apparent that IBM is “eating its own dogfood” and moving to exploit the advantages of ML within its own products. From operational analytics to building a smarter DB2 optimiser, there are some fascinating new use cases for this approach that should start to benefit us very soon.
• It’s difficult to read any IT-related articles currently without seeing at least a mention of DevOps. While it’s generally accepted that it’s somewhat more difficult to integrate typical DBA tasks into the kind of Continuous Delivery/Continuous Integration processes that are used for mainline application development, it’s not completely impossible and several presentations covered some of the progress that many customers are making in this area. The Triton team have also been doing a lot of work in this area recently (particularly on the mainframe platform) and again I’ll be covering this in more detail in future blog posts.
• Of course, there was also the usual round of great presentations covering core DB2 developments from stalwarts such as John Campbell, Terry Purcell and Florence Dubois. Despite all the fancy/trendy stuff above, it’s good to know that the engine that drives so many critical workloads around the world continues to be developed and enhanced.

Philadelphia is Greek for “city of brotherly love”, a place that embraces diversity and tolerance. A glance at the topics above shows that technical diversity is alive and well in the DB2 world as well.