Sample MQ Statistics and Accounting Dashboards

Examples of how various MQ Statistics and MQ Accounting metrics can be collected and organized in customizable dashboards.

More MQ Statistics Videos

1. Overview of MQ Statistics and Health Assessments
2. Assessing MQ Buffer Manager Health
3. Assessing MQ Log Manager Health
4. MQ Message Manager Metrics Supplemented by Accounting Data
5. Log Manager and Buffer Manager Metrics
6. Sample MQ Statistics and Accounting Dashboards

Video Transcript

So now let’s go back and review what we put in the dashboard. When we started out, we saw the correlation between deferred write thresholds and high levels of buffer utilization. Then in the logging area, we saw exceptions relating to waiting for available buffers that were occurring in the early morning hours. When we looked at just the volume of put requests, there was a spike in this particular time of day. And we said, we’re going to come back and research that a bit later. Here we compared the gets versus the puts and saw that there typically were more gets going on. And when we look at the profile for log megabytes written, it’s consistently much higher in those early morning hours. Here we were comparing the log megabytes to the number of write I/Os going on. And then finally, we also saw that there was a high volume of Disk read operations when the messages are not found in the buffer and you have to go to disk. That happens at a high volume very frequently in the early morning hours.

So, as we went through here today, we kind of built a dashboard of views. There was that transaction where the CPU was declining over the interval here. With the work coming from batch, you could see how much CPU by job name or by address space name. Here was the analysis we did on a buffer pool. Here’s how the elapsed time profile compared among different CICS transactions. When we looked at CPU per get across the IMS PSBs there wasn’t much difference between them. When we had that spike and elapsed time on the IMS work, we saw it was that particular PSB. Here was the message length distribution, and we saw there were two particular CICS transactions that were generating the higher volume of the longer messages. And here was, again, that CICS transaction with the declining times. Here was the lengths of messages by CICS transaction, so we see who’s generating the longer ones. This was the Queue Depth and we saw that exception there and we could see which CICS transaction encountered that. And finally, we talked about coupling facility metrics that are available in the accounting data as well.