Articles & Content

Archives

E16 - Analyzing DB2 Application Performance using Big Data

Topic: Cross Platform DB2 for z/OS & LUW

Subtopic: 2016



First, we’ll look at extracting performance data from your DB2 monitors. Thread data from SMF is useful for recent performance problems, but to really understand application performance and trending SQL performance data is more useful.
Next, we’ll discuss building a performance warehouse. What are the factors you must consider in order to pick the best place to store this data? The data will be large and to be successful you will want to keep history going back over a year or more. What are good, low cost options? We’ll consider various Big Data offerings, including Hadoop, MongoDB and Cassandra.
Then, how do we analyze the data by smoothing out the spikes. We’re looking for big trends, not individual anomalies. How do we smooth out the data while keeping usable metrics?
Finally, we’ll see how to create baseline thresholds once “normal” behavior is determined. Then identify growth over time. Once we’ve see where you’ve been and we can predict where you’re going. We’ll see how to save the analyzed data to be able to repeat this process over again in the future.



Click Here to Download

NOTE: These are only open to members of IDUG. If you are not a member, please CLICK HERE for more information.