Selected MQ Accounting Data SMF Metrics – Part 2

Examples of various SMF metrics captured in MQ Accounting data that provide insights into detailed MQ activity.

More MQ Accounting Videos

1. Overview of MQ Accounting Data
2. Viewing Accounting Data by Queue Level
3. Viewing Accounting Data by Connection Type
4. Selected Accounting Data Metrics – Part 1
5. Selected Accounting Data Metrics – Part 2
6. Sample MQ Statistics and Accounting Dashboards

Video Transcript

So we mentioned earlier when we were talking about logging that persistent messages drive more logging activity or most of the logging activity. So let’s look at the volume of persistent messages here. And in this environment, they are almost all coming from channel initiator messages. And so, for those, we can look at them by queue manager and we can see that it’s these two queue managers that are generating the great majority of the persistent messages. And if we want to look at it and see what queue they’re going to, we can see they’re almost all going to that particular queue. And again, there’s a real low number of CICS messages here, but, again, we could look at it by transaction, if we wanted to and see which transactions are generating persistent messages, particularly, you know, if we’ve got an issue where that’s driving some unacceptable or challenges, at least with the logging. The accounting data also captures the maximum encountered queue depth, which is here. So this initial view is by connection type. So again, we can look at the CICS work if we want to look at the CICS work by queue.

So the CICS messages, when there’s a big, long queue depth, all are involved with this particular delete queue. And if we looked underneath, we’d see that it’s widely used by many, many transactions. I’ll grab onto that one and you can see there was a spike here in the queue depth for one other queue. And so if we drill into that, we’ll see which transaction it was that encountered that queue depth.

So that was maximum queue depth. The accounting data also has time spent on the queue, maximum, minimum, and total elapsed time from which you can calculate an average. So here, in terms of time on queue, the messages involved with the channel initiator workload are on the queue far longer and kind of squash the scale for everything else. So let’s just go and remove those from this view so we can look at the others.

And when we do that, we see that there were kind of 10:00 AM and 2:00 PM from batch, there were a couple of times when there were messages that were spending a really long time on the queue. So again, we can drill into that by job and see that actually there’s kind of dots on top of each other. There’s a set of six jobs here that were dealing with those messages that were on the queue for a really long time.

So a final area of accounting data that I’ll just touch on is coupling facility metrics that are involved with supporting shared queues. The use of shared queues across multiple z/OS systems leverages the coupling facility to manage that. So as you would expect, based on what we’ve seen so far, we can view this data by queue name, and we can also view it by connection type.

But in addition, the coupling facility-specific metrics are also provided, including the volume of requests that are broken out by synchronous and asynchronous for the coupling facility. So here, those are viewed by queue name. Let’s go ahead and grab that one, and then also view by structure name. There they are synchronous and asynchronous. And if we’re interested in finding out which queue is involved with persistent messages, again, we can look here, and we can see that it’s that queue, or we can also look and see which workloads were generating those persistent messages. The accounting data can answer those questions.