Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
?
Solved

AIX 5.1 - Please help me understand "runnable processes"/Load Average

Posted on 2010-09-24
2
Medium Priority
?
771 Views
Last Modified: 2013-11-17
In this thread wmp answered the difference between load average and CPU idle time.
http://www.experts-exchange.com/OS/Unix/AIX/Q_26444059.html

I am a bit confused by what this means "load average is the number of runnable processes"

Here is the situation.  We do telephone interviews.  Each of 250 or so interviewers has a terminal session open.  (some telnet, some dumb terminals) and in that session they are running the process for our software that handle displaying questions/recording answers.  This process is called "survent".  And for each interviewer a single survent process runs for the entire shift.  

usually each survent process takes 0.1-0.3% of CPU -- and load average is relatively low.  (3-8)
Every once in a while the load average jumps up to 20-30, and lots of these survent process's start using much more CPU.   I see a lot of them dance around 1-5% each in topas.  

Obviously I need to work with the vendor who makes the survent software -- but all the explanations they have given so far for why it would use more CPU, dont really fit with what is happening.

So I am hoping if I can understand better what "runnable process" means, maybe that can help me point the vender in the right direction of where to look.

NOTE: that there is not really any change in what the interviewers are doing, when all this happens.  They are reading screens and entering codes for answers.  They are doing this all the time.  So I dont see why runnable processes would suddenly increase.  

Any input/ideas would be greatly helpful.  

Thanks wmp!  Since you are surely the one who is going to be answering this.   :)
0
Comment
Question by:Xetroximyn
2 Comments
 
LVL 68

Accepted Solution

by:
woolmilkporc earned 2000 total points
ID: 33755544
Hi again,

as I explained, the load consists of the number of processes running on a CPU plus the number of processes (or better threads) waiting for a CPU.

Now why would the load go up?

Obviously, one reason could be that there is simply more demand - more interviewers are starting processes (hitting <Enter> to store their data or to run queries) in a smaller time window.

The other, very common reason could be that a given process cannot be finished in the usual time, so others must wait.

This can be due to I/O wait - there are processes demanding much disk I/O thus slowing down the whole system - which leads to an increase of processes in the wait queue.

Or it could be due to paging - processes whose working set is paged out must wait until it is paged in again - thus occupying the CPU with a non-payload related task (the page-in processing). This leads to a growing wait queue as well.

If a database is involved a third reason could be the database not responding fast enough - maybe due to index splits, checkpoint processing or the like.

When such a situation arises you should carefully examine the values for "Wait" ( = I/O Wait) and " PgspIn"/" PgspOut" (= Paging from/to disk) in the upper left area of topas, as well as the overall disk and network utilization (miggle left) - not only the CPU utilization of single processes in the top lis (lower left).

wmp
0
 

Author Closing Comment

by:Xetroximyn
ID: 33862866
Thanks!
0

Featured Post

What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Attention: This article will no longer be maintained. If you have any questions, please feel free to mail me. jgh@FreeBSD.org Please see http://www.freebsd.org/doc/en_US.ISO8859-1/articles/freebsd-update-server/ for the updated article. It is avail…
Every server (virtual or physical) needs a console: and the console can be provided through hardware directly connected, software for remote connections, local connections, through a KVM, etc. This document explains the different types of consol…
Learn how to find files with the shell using the find and locate commands. Use locate to find a needle in a haystack.: With locate, check if the file still exists.: Use find to get the actual location of the file.:
This video shows how to set up a shell script to accept a positional parameter when called, pass that to a SQL script, accept the output from the statement back and then manipulate it in the Shell.
Suggested Courses

578 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question