IBM®
Skip to main content
    Country/region [select]      Terms of use
 
 
     Home      Products      Services & solutions      Support & downloads      My account     
  IBM Wikis > AIX > ... > nmon > nmon Sample Output
AIX Log In | Sign Up   View a printable version of the current page.
nmon Sample Output
Added by Nigel Griffiths, last edited by Bernd Eckenfels on Apr 21, 2008  (view change) show comment
Labels: 
(None)

nmon Sample Output

This is the start screen with some useful information on it about the machine

This is just the CPU utilization on a X Windows session that supports colour on AIX

This is the CPU Utilization of an Idle machine which is POWER5 and is a Share Processor Logical Partitions (SPLPAR).
This is what marketing calls a Micro Partition.

Here it the same partition but with a workload that takes it near to the entitlement.
Note the utilization is near 100% on the entitlement but this LPAR can go twice as fast as this if there is spared CPU cycles in the Shared Processor Pool

And below is the LPAR stats at the same time:

Here it the same partition but with a workload that has been taken the CPU use is well over the entitlement.

If you workload is peaky it is worth getting an overview of these peaks and so get a feal for the average and peaks.
This is provided by the Longer Term CPU View:

Thee are lots of other stats that are useful

Memory

File-System details

Network initially shows the network troughputs and errors

If there are not errors the error details will disapear until there are errors

If you use NFS this is the only tool to show you the statistics as they happen - here I am running ls -lR /NFS-mountpoint

One key area for tuning is disks there are different ways to view these
Simple disk graphs

The same data but in numbers and some more details like transfers (xter)

Then you might want to know about the types of disk and their connections

Disks are connected by adapters
WARNING: the stats here are simply collected by adding up the disk stats - other devices like tapes are NOT included (there are no tape stats)

Top Processes are clearly important and there are lots of views for these

Top 3 is the default

Top 1 show some more details of the process

Top 2 shows the relationships and collective stats

Top u shows the command name instead of the process name

Top U (upper case) shows the WLM classes too

Related to this is the Async I/O AIX kernel processes - note these are collected with the Top Processes so Top gets switch on at the same time.

Workload Manager (WLM) Classes are also monitored

Powered by Atlassian Confluence, the Enterprise Wiki. (Version: 2.2.10 Build:#528 Nov 29, 2006)
    About IBM Privacy Contact