HP 3000 Manuals

HP GlancePlus User's Manual (for MPE/iX Systems)

Reading the Screen Display 

Although all GlancePlus screens appear similar, they carry different
information about system resources.  Each screen is divided into four
sections--Banner Line, Global Bars, Detail Display, and Function Keys.
The Banner Line and Global Bars present the same information on every
screen, whereas the Detail Display section differs from screen to screen.

Banner Line 

The Banner Line at the top of the screen contains information on the
version of the product, product name, current time (hour:minute:second),
and screen identifier (figure 3-2).  The screen identifier may contain
information pertinent to the particular screen, such as the particular
job number in the Job screen.


          Figure 3-2.  The Banner Line 

Global Bars 

Three Global Bars appear directly below the Banner Line (figure 3-3).
These bars, and the percentages listed to the right of them, show the
system's use of three resources:  CPU, Disc, and Memory.  The bars are
present on every screen to help you keep track of the overall "global"
picture.  The letters superimposed on each bar specify how that portion
of the resource is being used.


          Figure 3-3.  Global Bars 

CPU Bar.   


            

The CPU bar shows how much of the central processing unit's time was
spent on various activities during the last interval.  The current,
average, and highest percentages are displayed numerically to the right
of the bar.  The length of the bar indicates the percentage of CPU time
used or time that could be used if pending disc I/Os were satisfied
during the current interval.  Paused for disc is not included in these
numbers since the CPU is inactive during this time because of pending
disc transfers.  (Scale:  0 to 100 percent.)

The components of the CPU bar are as follows:

M  MPE/iX functions such as memory management,
   interrupt handling, and deciding which process
   uses the CPU

S  System processes such as data communication monitors
   and spoolers

I  Interactive (session) processes

B  Batch (job) processes

P  Paused for disc (no CPU is used but Disc I/O is
   occurring)

*  Average usage of this resource during this execution of
   GlancePlus (not including Paused for Disc).


NOTE  The CPU bar may be configured to show activity by queue, which
      changes the above components.  Chapter 7 contains information on
      customizing the GlancePlus global bars.


Disc Bar.   


            

The Disc bar shows the rate of disc transfers that occurred on the system
during the last interval.  The current, average, and highest number of
transfers per second are shown numerically to the right of the bar.  The
length of the bar indicates the total number of transfers per second
occurring currently.  (Scale:  0 to 100 physical transfers per second.)

The components of the Disc bar are as follows:

M  Memory management transfers (swapping)

S  System process transfers (data communication monitors
   and spoolers)

I  Interactive (session) process transfers

B  Batch (job) process transfers

*  Average usage of this resource during this execution
   of GlancePlus.

Memory Bar.   


            

This bar represents the amount of main Memory in current use as a
percentage.  The current, average, and highest percentages of main Memory
use are displayed to the right of the bar.  The length of the bar
indicates the total amount of main Memory that is currently being used.


NOTE  It is normal for main Memory to be full even when there is little
      activity on the system.


Components of the Memory bar show how much Memory is being used for
different purposes.  The Memory bar components are as follows:

M  Resident MPE (operating system code and data are
   not swapable)

C  Code (both native and compatibility-mode programs
   and libraries)

S  Stacks (both native and compatibility-mode program
   data stacks)

D  Data objects (certain system data structures and
   compatibility-mode data segments)

F  Files (data files such as TurboIMAGE data sets, KSAM
   key and data files, and normal MPE files)

*  Average usage of this resource during this execution
   of GlancePlus.

A particular category will not be displayed if there is insufficient
Memory usage to occupy one character space on the screen.


NOTE  The Memory bar may be configured to show fault rates, which changes
      the above components.  Chapter 7 contains information on
      customizing the GlancePlus global bars.


Percentage Columns.   

The percentage columns at the right of the global bars (figure 3-3) show
current, average, and highest use of each system resource.  Current value
represents use for the current interval.  Average is the mean value of
all usage data collected since the present monitoring session began.
High value shows the single highest usage amount of that resource during
the monitoring session.

If you reset the statistics to zero with the Zero Averages command (Z),
the current value will still represent current usage for each interval,
but subsequent average and high values will represent data collected
since the resetting.


NOTE  Upon entry to GlancePlus, the first global screen will show
      GLANCEXL using a high percentage of CPU. This is normal since the
      program must take two data samples within a few seconds of each
      other in order to display the first Global screen.  At the default
      30-second interval, CPU usage is typically low and should cause no
      problems on your system.  CPU overhead used by GlancePlus is
      proportional to the frequency of screen update and the number of
      processes on the system.  You can cause excessive CPU overhead by
      pressing a key every few seconds.


Detail Display 

The information presented in the detail display portion of the screen
below the global bars depends on the type of screen (figure 3-4).


          Figure 3-4.  An Example of a Detail Display 

Some screens show a summary of processes, such as the Global,
Job/Session, and Filter screens.  Interesting applications are scrolled
down as new entries are added to the top of the listing to keep the
latest information just below the Global bars.  The process information
for the most current interval is displayed at the top of the detail area
immediately below the global bars.  You can use terminal memory to scroll
back to process data from previous intervals.

Other screens show statistics for an individual resource or individual
process.  These screens refresh the detail display at every interval.
Examples of this type of screen include the CPU, Disc, Memory, and
Process screens.

The Filter Select and Set Option Values screens help you set parameters.
Their detail display is a table of value settings followed by a prompt
you can use to change a value.  Every time you change a value, a new,
adjusted table of values is displayed below the previous one until you
terminate the dialogue by pressing Return.

On many systems, it could take several screens full of data to display
all currently-running processes.  Evaluating that much information could
be confusing.  To avoid this, GlancePlus lets you limit the processes
displayed to those that are interesting to you--processes that surpass
predetermined threshold-usage values.  Interesting processes either use a
major portion of the system's resources or must wait exceptionally long
to use them, and are the processes most likely to reflect an existing or
potential problem.  The procedure for setting process thresholds is
described at the end of this chapter.

The Global screen is the starting point for most performance analysis
activity.  Other GlancePlus screens are discussed individually in chapter
5.

MPE/iX 5.0 Documentation