CICS Response Time Analysis Scenario 1 Values Differing by System ID

CICS response time analysis scenario where insights into the causes of variances across z/OS systems can be quickly derived from isolating the components driving the differences.

More CICS Transactions Videos

1. CICS Timing Metrics – Overview
2. Transaction Response Time Analysis
3. Timing Analysis by High-Level Components
4. Response Time Analysis Scenario 1 Values Differing by System ID
5. Integrating WLM Performance Index and CICS Transaction Analysis
6. CICS Web Services Metrics
7. Overview of Non-timing Transaction Metrics

Video Transcript

All right. So let’s see what we might be able to learn by exploring the components of response time data. So I’m going to focus here, this is the Top 20 report, it’s kind of busy, so I’m going to focus here on the Top four transactions by volume. And so let’s say that we’re tasked with analyzing the day shift response time for this particular transaction. So let me go ahead and change that to day shift.

So, we might begin by determining if response times differ across the zOS systems where that transaction is executing. So, when we do that, we see that indeed, we kind of have two groupings of three systems for that. So let’s go ahead capture that in our dashboard. All right. So now let’s drill into the underlying components. So let’s take one of the high response time systems, I/O Wait Times, a major driver, 26 or so milliseconds, and the component driving that is Wait for Control at the end of MRO, almost entirely.

Okay, so now let’s go ahead and take one of the systems that had the lower response times, do the same analysis again. It’s total I/O Wait Time, but you can see the number is about half of what it was up above 12 or 13 milliseconds. And again, if we look at that in detail, it is Wait for Control at the end of MRO. Okay. So it appears, we’ve got very different profiles for this MRO metric across those two sets of regions. So let’s zero in on that particular component of the I/O Wait Time and remove all the other ones.

And once we do that, then let’s look at this at the System ID level and across all the systems here. And again, we’re just looking at weight or control at the end of MRO. There we go. Okay. And when we look at that, we see indeed that there’s a very different pattern in that weight metric between those three systems and the three systems with the lower response times. So again, let’s go ahead and capture this in our dashboard and as you may have guessed in this environment, this set of three regions runs on a different CEC from this set of three regions. And so that indicates that there’s longer inner region communication time going on for the regions on the one CEC versus the other. So that’s a good example of the kind of interesting insights that can be uncovered from exploring the timing metrics in the transaction data.