View Printable Version

C09 - Find the treasures of DB2 monitor data with deltas

Session Number: 1096
Track: DB2 for LUW - I
Session Type: Podium Presentation
Primary Presenter: Paavo Tukia [Database administrator - Tieto]
Co-Presenter 1: {no name provided}
Co-Presenter 2: {no name provided}
Co-Presenter 3: {no name provided}
Co-Presenter 4: {no name provided}

Paradise Valley => Thu, May 15, 2014 (08:00 AM - 09:00 AM)

Session Code: 1319
Speaker Bio: Paavo Tukia started working with mainframe DB2 in 1986, then as a DB2 pioneer in Finland.
He started working with DB2 LUW in the nineties. First just occasionally, but then more and more and since 2005 DB2 LUW has been his platform for DB2. Now he works as a team leader of a DB2 administrator group in the company Tieto in Finland.
Audience experience level: Intermediate
Presentation Category: Managing Performance
Presentation Platform: DB2 for Linux, UNIX, Windows
Audiences this presentation will apply to: Application Developers, Data Architects, Database Administrators, Systems Programmers, New Users
Technical areas this presentation will apply to: Database Performance (DB2 for LUW)
Objective 1: To understand deltas is vital for doing db2 monitoring, but the issue is not much covered. Here you learn the delta theory and praxis and hopefully learn to love deltas.
Objective 2: Steve Rees(IBM) and Martin Hubel have read a pre version of the presentation and think this to be a good material about the important  issue.
Objective 3: You learn the SQL principles for getting deltas and a stored procedure is given to get deltas automatically for any table having cumulative values.
Objective 4: You get advice on how to store cumulative historic information of monitor data and to analyze it without disturbing the production databases.
Objective 5: You learn clever, but easy reporting methods using  delta values.

Abstract:  From DB2 you can get all the monitor data you ever need and store it easily into tables. The problem is that the numbers are cumulative. You often want to know how much the cumulative numbers have changed between points of time t1 and t2 – that is: you want to know the deltas.
Here's a universal method for getting the deltas. The method applies to any table with cumulative data and it does not matter how often you collect monitor data (by every n minutes or hours etc). You also get a stored procedure to automate the getting of deltas.

For questions or concerns about your event registration, please contact