CPU utlization


 
Thread Tools Search this Thread
Top Forums UNIX for Advanced & Expert Users CPU utlization
# 8  
Old 03-31-2010
Quote:
Originally Posted by pludi
On a 48 core machine any load below 48 isn't high. With a load of 23, one could even go so far as to say it's bored half of the time.
If all the processes are processor and memory load balanced.....

I have no idea how you are saying that with 48 cores, an average below 23 is "it is bored half the time" - where are these numbers coming from?
# 9  
Old 03-31-2010
From the definition of the "load". If nothing is running, the load is 0. Each process running or runnable adds 1 to the load, which then gets averaged over 1/5/15 minute(s). So a 1-minute load of 23.3 means that during the last minute there were on average 23.3 processes running or ready to run. Spread over 48 cores that's about half of them used, with the other half more or less idling away.
# 10  
Old 03-31-2010
That is not the definition of "load" at all. Here are my load balances:
Code:
bash-3.00# w
  2:46am  up 2 day(s), 10:33,  1 user,  load average: 0.00, 0.00, 0.00

So am I to understand nothing is running at all? Here are my running processes:
Code:
bash-3.00# prstat
   PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
  1163 root     3624K 3152K cpu0    49    0   0:00:00 0.2% prstat/1
   288 root     3120K 2624K sleep   49    0   0:00:00 0.0% bash/1
   192 root     2520K 1640K sleep  100    -   0:00:24 0.0% xntpd/1
   610 root       10M 5960K sleep   59    0   0:00:17 0.0% httpd/1
   233 daemon   7016K 6424K sleep   59    0   0:00:12 0.0% nfsmapid/4
   476 root     8160K 2784K sleep   59    0   0:00:09 0.0% sendmail/1
   128 root     7640K 4104K sleep   59    0   0:00:15 0.0% nscd/30
   236 root     2360K 1816K sleep   59    0   0:00:00 0.0% rpc.bootparamd/1
   229 daemon   2760K 2112K sleep   59    0   0:00:00 0.0% rpcbind/1
   253 root     2456K 1912K sleep   59    0   0:00:00 0.0% ttymon/1
   201 root     2864K 1664K sleep   59    0   0:00:00 0.0% cron/1
   109 root     3096K 2224K sleep   59    0   0:00:00 0.0% picld/4
   234 daemon   2768K 2320K sleep   59    0   0:00:00 0.0% statd/1
   122 daemon   4568K 2696K sleep   59    0   0:00:00 0.0% kcfd/3
     9 root       10M 9432K sleep   59    0   0:00:33 0.0% svc.configd/17
Total: 44 processes, 166 lwps, load averages: 0.00, 0.00, 0.00

Please note it is a 1 core machine.

Last edited by pludi; 03-31-2010 at 06:54 AM.. Reason: code tags, please...
# 11  
Old 03-31-2010
Exactly that is what we are experimenting.

We don't want to process jobs in paralel rather we want to finsh off them as soon as we revieved the job.

if i have loaded with 24 jobs will it two times faster in a finishing job compared 48 jobs?

Please help me on this
# 12  
Old 03-31-2010
You should use "prstat" on Solaris instead of "top" if you want accurate numbers.

Also, 1.95% is awfully close to the 2+% that is going to be the max CPU utilization of any single-threaded process on a 48-CPU machine.

Is there a specific issue you're trying to solve?
# 13  
Old 03-31-2010
Kumaran, my opinion is that your load is too high - about time for a reboot! Smilie

---------- Post updated at 06:09 AM ---------- Previous update was at 06:06 AM ----------

Quote:
Originally Posted by achenle
You should use "prstat" on Solaris instead of "top" if you want accurate numbers.

Also, 1.95% is awfully close to the 2+% that is going to be the max CPU utilization of any single-threaded process on a 48-CPU machine.

Is there a specific issue you're trying to solve?
Shhhhh.....you dont want to give it all away now do you? Smilie Keep us apart from those linux admins (I am one myself).....
# 14  
Old 03-31-2010
Okay I will explain my problem here.

Please help me out.

My server has to keep on processing the files as and when it arrives to the server. And the queue of pending shouldn't be long.

That is i have to finish processing them as soon as possible than processing many files at a time.

To achieve this which is preffered, loading until i reach load avg 48 or keeping it less than 48, so multiple cpus can perform a task to complete it easy.

I have tried keeping it higher than 48, it does parallel processing but it takes lot of to complete a single task. that is not i wanted.

Please suggest me.

I have SUNOs and 48 cpus.

Thanks
kumaran
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. UNIX for Beginners Questions & Answers

Why Entitlement CPU can't be set to same as Virtual CPU?

I read that Entitlement CPU should be set to max 75% compare to Virtual CPU. May I know the reason. I have set the Entitlement CPU = Virtual CPU on AIX . It works fine . Can you help to understand. (1 Reply)
Discussion started by: gabhanes
1 Replies

2. AIX

Entitled CPU lesser Than Used CPU !!

Hi All, It may be a n00b question, but i really want to know , How Entitled Capacity is less and Used CPU is more when there is no Free CPU is available in the managed system. I have 5LPARs in a MS with Dual VIO. Managed System CPU details Available: 0.20 Assigned to partitions: 15.80... (11 Replies)
Discussion started by: Thala
11 Replies

3. UNIX for Dummies Questions & Answers

Is it possible to combine multiple CPU to act as a single CPU on the same server?

We have a single threaded application which is restricted by CPU usage even though there are multiple CPUs on the server, hence leading to significant performance issues. Is it possible to merge / combine multiple CPUs at OS level so it appear as a single CPU for the application? (6 Replies)
Discussion started by: Dissa
6 Replies

4. Solaris

Find why CPU utlization WAS high yesterday

I want to investigate why CPU utilization was high yesterday on my solaris 10 machine. I have been reported this problem but how can I investigate why CPU utilization went high at that time? :confused: (4 Replies)
Discussion started by: kashif_islam
4 Replies

5. Solaris

Difference between CPU CHIP and CPU ID

Hi All, What is the difference between CPU CHIP and CUP ID on SUN/ Oracle M5000 servers.. Thanks in advance (3 Replies)
Discussion started by: kumarmani
3 Replies

6. Solaris

CPU and CPU-Utilization

Hi, I am a weblogic Admin I our env there are 5 servers running in one Solaris 10 Sparc machine. But when i tried for process status using <top> command I got the following output load averages: 1.75, 2.18, 2.12; up 134+08:28:49 22:24:21 79 processes: 77 sleeping, 1 running,... (1 Reply)
Discussion started by: Joseph Antoine
1 Replies

7. Solaris

cpu-shares vs cpu-cap in solaris

Can anyone tell me difference between cpu-shares vs cpu-cap in solaris & how FSS will work with cpu-caps ? (9 Replies)
Discussion started by: fugitive
9 Replies

8. Solaris

Multi CPU Solaris system shows 100% CPU usage.

Hello Friends, On one of my Solaris 10 box, CPU usage shows 100% using "sar", "vmstat". However, it has 4 CPUs and prstat and glance are not showing enough processes to justify high CPU utilization. ========================================================================= $ prstat -a ... (4 Replies)
Discussion started by: mahive
4 Replies

9. UNIX for Dummies Questions & Answers

how to get persistant cpu utilization values per process per cpu in linux (! top,ps)

hi, i want to know cpu utilizatiion per process per cpu..for single processor also if multicore in linux ..to use these values in shell script to kill processes exceeding cpu utilization.ps (pcpu) command does not give exact values..top does not give persistant values..psstat,vmstat..does njot... (3 Replies)
Discussion started by: pankajd
3 Replies

10. Solaris

How to down a particular CPU in a server, witch is having 16 CPU's

Hi.., my dout is a solaris server is having 16 cpu's. in tht one cpu running some error process, accupaying more space. I wanna down tht particular CPU only with out interrupting the other 15 CPU's. how can i do this. is there any command for this ?? (5 Replies)
Discussion started by: b.janardhanguru
5 Replies
Login or Register to Ask a Question