Quite some time ago, I wrote a blog on Understanding Disk Space Usage. Near the end of that article, I mentioned that Collection Services collects storage usage data in the QAPMJOBMI file.
Late last year, there were additional enhancements to Navigator for i, which included updates to the Performance Data Investigator (PDI). With the Fall 2013 PDI enhancements, you can now visualize these Collection Services storage usage metrics. You will need to install the required PTFs to enable this support, as documented in the “Fall 2013” section of the Performance on the Web information.
You will now find some new charts that will show the storage usage on your system over time.
In PDI, open Collection Services then expand the Disk folder. Here you will find a new folder named Storage Allocation. In this folder are two new charts to display storage allocation information.
Storage Allocation/Deallocation Overview
This graph displays an overview of the storage allocated and deallocated for the partition overall; you can drill into storage allocation/deallocation by job from this overview.
Here’s an example of the graph:
Storage Allocation/Deallocation by Thread or Task
This graph displays the storage allocated or dealloacted by job or thread. You can then view the details for one job or thread over time.
Here is an example of this graph:
From here, we could even go look at the details of one of the jobs, where we can see how much storage the specific job allocated and deallocated over time.
Here’s an example:
If you think you have a storage leak in your system, the Collection Services data, along with the new support in PDI to visualize this information, is another useful diagnostic tool.
The wonderful thing about looking at Collection Services data is that you can see what has been happening over time. With interfaces such as Work with System Activity (WRKSYSACT), you see what the values are right now. Collection Services can help you understand what was happening in the past to help you understand how you got to where you are now.
This blog post was originally published on IBMSystemsMag.com and is reproduced here by permission of IBM Systems Media.