Understanding a window's time and duration fields


You may find that the time and duration fields on the window information line do not always contain the data you expect. These fields reflect the actual data displayed, which may or may not be the same as you requested with the TIME command.

For example, let's say it's 9:00 A.M. and you want to look at APOVER to review workload performance between 5:00 A.M and 8:00 A.M. this morning. You display your historical APOVER view by entering this command:

TIME * 8:00 3h

After entering this form of the TIME command, you expect the window information line to look like this:

 H1=APOVER==========SYSB=====*====07NOV2000====8:00:00=180M==MVVP=======24==>

8:00 is the last interval in the duration you requested; the three-hour period between 5:00 A.M and 8:00 A.M. is equivalent to 180 minutes. However, the window information line actually looks like this:

H1=APOVER=====SYSB=====*====07NOV2000===7:15:00==120M==MVVP=======21==>

Why does the time field display 7:15 instead of 8:00? And why is there only 120 minutes worth of data instead of 180?

Here is what is happening. Data is not always available for the intervals you request. Sometimes the PAS is shut down in the middle of a recording interval, or data-collection monitors become inactive. This creates gaps in the data recorded to the historical data set. What appears on the window information line represents the cumulative time of the intervals that actually have recorded data in the duration you specified.

In the example shown above, 7:15 was the last interval within the time frame in which data was recorded. There simply was not any data recorded at 7:30, 7:45, and 8:00. The window information line displays 7:15 instead of 8:00 because 7:15 is the last recording interval that provided data to the historical data set.

But if there was not any data for three of the intervals, why is there only data for 8 intervals (120 minutes), rather than 9 intervals (135 minutes)? The answer is that BBI-3 uses the time between the first and last available interval. In this case, there must have been an interval's worth of data missing from 5:00 to 5:15 and, as a result, the interval count is reduced by an additional interval—or another 15 minutes.

Although there may have been some other gaps in the record between 5:15 and 7:15, they are not relevant; BBI-3 tries to normalize things so that you get an accurate picture of the data actually displayed in the view.

Tip

The time field always contains the end of the last interval for which data was available and the number of intervals for which data was actually available, (normalized over the time frame you that requested).

A Word about Summarization

VistaPoint views can become crowded when you use the TIME command DURATION parameter. A separate view line is created for every entity in each interval that is part of the duration. Historical views that show data from durations that extend over many intervals may require you to scroll through many windows to see all the data.

To make it easier to manage your historical views, you should use the view customization summarization option Group by to summarize the data. Summarization allows you to compress several lines of data into a single row, based on criteria you specify.

Tip

Summarization is an extremely effective tool for managing data from multiple intervals. However, it should be used carefully because some view data becomes meaningless when it is summarized.


 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*