International DB2 Users Group (IDUG) Trip Report - Dave Beulke

Here is my trip report from the annual International DB2 User Group (IDUG) conference held last week. The event was great because of the technical, real world, pragmatic, and easy-to-understand content, especially from the user speakers. The IBMers, consultants, and vendors also provided great material to help make everyone’s DB2 systems and applications perform better with less CPU, I/O, and administration people time costs. The application development sessions showed great techniques for getting DB2 development done more quickly by using its many interfaces, SQL functions, and SQL/PL stored procedures.

Another great aspect of the IDUG conference is the ability to hear from all the users of the latest DB2 versions for both DB2 z/OS and DB2 LUW systems. Technical aspects of the DB2 systems continue their transformation and innovative enhancements through Spark, dashDB, and the new version of IDAA. All the great DB2 family ACID (atomicity, consistency, isolation, and durability) capabilities which have been included in DB2 for decades will be arriving very quickly to open source Spark and dashDB solutions.

Below are some of the other trip report details regarding applications and new capabilities within the DB2 Family that are very exciting.

• DB2 Version 12 Continuous Delivery. DB2 12 for z/OS with IBM’s new continuous delivery model will deliver new DB2 functionality more quickly than waiting for a new DB2 version every three years. To wrap your head around DB2 12 continuous availability, you need to understand the four factors within the DB2 framework that control all your DB2 capabilities. These four factors are the DB2 maintenance level (ML), the DB2 Catalog level (CL), functional level (FL), and the application compatibility (AC) level.
  
  Through these four controlling factors DB2 z/OS is going to be able to manage and advance your DB2 application functionality more rapidly. By applying maintenance (ML) more frequently, you can advance the options available to your business, applications, and programmers. Improve the DB2 Catalog (CL) to support these new functions and embrace the latest functionality (FL) like the new DB2 Version 12 fast traverse blocks (FTB), improved performance of the new Insert algorithm, SQL improvement and many more new DB2 12 features. Then control the application compatibility (AC) and its implementation usage through the APPLCOMPAT BIND/REBIND parameters for managing when new program functionality is put into your environments and/or production.
  
  Understand and start to set up your development practices and your change management facilities now to align on these four factors. Through this alignment your application development can leverage the DB2 Version 12 features as quickly as they come out of the IBM lab or as quickly as you can test and manage them into your production environment.
  
  
• DB2 V12 Insert has tremendous performance improvements. As I said earlier, DB2 has a new Insert algorithm for better Insert performance. IBM testing of this new algorithm showed new DB2 Version 12 applications can experience a tremendous 24x Insert performance improvement while using 1/64 of the CPU. These Insert performance improvements are accomplished through new space search algorithm with UTS tablespaces with MEMBER CLUSTER. By using a more efficient space search, eliminating page contention and false leads, and leveraging the un-clustered Insert MEMBER CLUSTER capabilities, tremendous concurrent performance improvements are achieved. Of course not everyone will get a 24 times performance improvement, but the potential is tremendous and can make a huge difference in your daily processing.
  
  
• DB2 Version 11 of z/OS provides a simple JSON SQL interface. New JSON functions arrive with DB2 Version 12 for z/OS. What is great about these new functions is that they have been retro-fitted back into DB2 Version 11 through PTF PI39003. Every developer can use JSON within their application if needed for a new phone, IoT, or other application interface. DB2 z/OS provides JSON functionality through its binary BLOB tablespace definition support that provides the object data store capabilities. The standard functionality within JSON SQL support is very similar to the XML support, so it is straight forward and easy for the application developers to quickly leverage JSON within Java, C++, or other web support languages.
  
  The following DB2 z/OS JSON functions are great to get your application JSON integration started.
  
  
  • BSON2JSON-converts BSON LOB into JSON string
  • JSON_VAL-retrieve specific value from inside a BSON object (also in V10)
  • SYSTOOLS.JSON_LEN – returns the size of an array holding the JSON data items
  • SYSTOOLS.JSON_TYPE – returns the data type of the data item.
  • SYSTOOLS.JSON_TABLE – returns the JSON data items within an array.
    
    
• DB2 LUW Version 11.1 new features are explained in a new book from George Baklarz. If you have been using DB2 LUW for a while, you know George puts out great in-depth useful examples and information in all of his books and examples. There is another new one that can be added to his many publications for previous versions of DB2. The book can be found here:  https://ibm.ent.box.com/v/DB2v11eBook. What is great about the link is it takes you not only to the new DB2 LUW Version 11 book, but also to a new version of the DB2 DEMO program that contains DB2 Family compatible JSON examples that can be adapted for both DB2 Version 11 z/OS with PTF PI39003 and DB2 LUW Version 11.1.
  
  
• DB2 IDAA success stories continue. During the IDUG conference, there were many great DB2 analytics presentations talking about IDAA. I heard several presentations and success stories using IDAA for a variety of analytics and for a general platform to speed up report processing for a wide variety of operations. There were stories describing how shops are putting their VSAM files into IDAA and getting quick analytical reports. Others are putting Oracle tables and SQL Server Tables into IDAA because it is faster to load and report on them in IDAA then to get a report out of those slow platforms.
  
  Big improvements are coming from IDAA in the next release later this year, including the ability to have no latency for data between the DB2 z/OS environment and IDAA. Real time data for real time analytics over big data within IDAA should help continue IDAA’s and DB2 z/OS success.

IDUG was great again this year, so plan your budget for IDUG North America next year April 30 – May 3, 2018 in downtown Philadelphia or this October in Lisbon, Portugal. Go to IDUG.org for more details.

Dave Beulke is a system strategist, application architect, and performance expert specializing in Big Data, data warehouses, and high performance internet business solutions.  He is an IBM Gold Consultant, Information Champion, President of DAMA-NCR, former President of International DB2 User Group, and frequent speaker at national and international conferences. His architectures, designs, and performance tuning techniques help organization better leverage their information assets, saving millions in processing costs. Follow him on Twitter  or connect through LinkedIn.