Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: MythTV: Users

CPU Load vs. CPU Utilization (uptime and top)

 

 

MythTV users RSS feed   Index | Next | Previous | View Threaded


spikeygg.mythbox at gmail

Aug 3, 2012, 8:13 PM

Post #1 of 4 (337 views)
Permalink
CPU Load vs. CPU Utilization (uptime and top)

When I launch top, I see:

top - 21:11:01 up 2:28, 4 users, load average: 1.91, 1.62, 1.36
Tasks: 232 total, 1 running, 231 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.6%us, 0.2%sy, 0.0%ni, 96.8%id, 2.3%wa, 0.0%hi, 0.1%si,
0.0%st
Mem: 132010188k total, 23824268k used, 108185920k free, 180384k buffers
Swap: 0k total, 0k used, 0k free, 21094840k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3223 spikeygg 20 0 2593m 44m 10m S 4 0.0 4:03.10 mythbackend
2918 spikeygg 20 0 516m 22m 10m S 1 0.0 1:21.19 unity-panel-ser
2920 spikeygg 20 0 417m 7680 3464 S 1 0.0 0:58.63 hud-service
5693 spikeygg 20 0 422m 15m 10m S 1 0.0 0:21.20 indicator-multi
2032 spikeygg 20 0 27324 3124 624 S 1 0.0 0:22.39 dbus-daemon
2338 spikeygg 20 0 1371m 92m 36m S 1 0.1 1:47.21 compiz
3239 spikeygg 20 0 1104m 10m 5548 S 1 0.0 1:31.26 mythlogserver
122 root 20 0 0 0 0 S 0 0.0 0:05.24 kworker/1:1
1233 mysql 20 0 2269m 70m 8088 S 0 0.1 0:08.79 mysqld
1288 root 20 0 206m 66m 10m S 0 0.1 2:22.10 Xorg
1568 nobody 20 0 21480 1412 1064 S 0 0.0 0:11.89 LCDd
2934 spikeygg 20 0 412m 5340 3952 S 0 0.0 0:26.34 indicator-appli
5502 spikeygg 20 0 904m 293m 35m S 0 0.2 2:09.67 firefox
6457 spikeygg 20 0 17468 1416 968 R 0 0.0 0:00.48 top
1 root 20 0 24544 2544 1352 S 0 0.0 0:02.63 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root 20 0 0 0 0 S 0 0.0 0:00.25 ksoftirqd/0

What gives, any idea why I have load averages of nearly 2 when the CPU is
basically idle with 1 running job??


blammo.doh at gmail

Aug 3, 2012, 9:04 PM

Post #2 of 4 (321 views)
Permalink
Re: CPU Load vs. CPU Utilization (uptime and top) [In reply to]

On Fri, Aug 3, 2012 at 8:13 PM, Greg Grotsky <spikeygg.mythbox [at] gmail> wrote:
> top - 21:11:01 up 2:28, 4 users, load average: 1.91, 1.62, 1.36
> Tasks: 232 total, 1 running, 231 sleeping, 0 stopped, 0 zombie
> Cpu(s): 0.6%us, 0.2%sy, 0.0%ni, 96.8%id, 2.3%wa, 0.0%hi, 0.1%si,
> 0.0%st
> Mem: 132010188k total, 23824268k used, 108185920k free, 180384k buffers
> Swap: 0k total, 0k used, 0k free, 21094840k cached
> <SNIP>
> What gives, any idea why I have load averages of nearly 2 when the CPU is
> basically idle with 1 running job??

Check VMstat, dstat, but I suspect either disk or network IO.
Something kernel level not assigned to an individual process.
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users


spikeygg.mythbox at gmail

Aug 3, 2012, 9:22 PM

Post #3 of 4 (323 views)
Permalink
Re: CPU Load vs. CPU Utilization (uptime and top) [In reply to]

On Fri, Aug 3, 2012 at 10:04 PM, Blammo <blammo.doh [at] gmail> wrote:

> Check VMstat, dstat, but I suspect either disk or network IO.
> Something kernel level not assigned to an individual process.
>
>
Thanks Blammo, I might have figured this out. I've been reading a little...

I found this page:
http://blog.scoutapp.com/articles/2009/07/31/understanding-load-averagesthat
explains load average and that lead me to this page:
http://blog.scoutapp.com/articles/2011/02/10/understanding-disk-i-o-when-should-you-be-worriedthat
explains disk i/o. It took me a little while but I finally realized
that they are actually advertisements, albeit very well written. Anyway,
you're right, it seems that most of my problem is likely my slow 5400rpm
disk or my network connection. I'm probably getting a very slow write
performance and this causes the CPU to spend most of it's time waiting for
the data (2.3%wa in my original post). The CPU wait % is defined as load
even though the CPU utilization is idle.

Is my thinking correct?


kpyle at austin

Aug 3, 2012, 9:24 PM

Post #4 of 4 (326 views)
Permalink
Re: CPU Load vs. CPU Utilization (uptime and top) [In reply to]

On 08/03/12 22:13, Greg Grotsky wrote:
> When I launch top, I see:
>
> top - 21:11:01 up 2:28, 4 users, load average: 1.91, 1.62, 1.36
> Tasks: 232 total, 1 running, 231 sleeping, 0 stopped, 0 zombie
> Cpu(s): 0.6%us, 0.2%sy, 0.0%ni, 96.8%id, 2.3%wa, 0.0%hi, 0.1%si,
> 0.0%st
> Mem: 132010188k total, 23824268k used, 108185920k free, 180384k buffers
> Swap: 0k total, 0k used, 0k free, 21094840k cached
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 3223 spikeygg 20 0 2593m 44m 10m S 4 0.0 4:03.10 mythbackend
> 2918 spikeygg 20 0 516m 22m 10m S 1 0.0 1:21.19 unity-panel-ser
> 2920 spikeygg 20 0 417m 7680 3464 S 1 0.0 0:58.63 hud-service
> 5693 spikeygg 20 0 422m 15m 10m S 1 0.0 0:21.20 indicator-multi
> 2032 spikeygg 20 0 27324 3124 624 S 1 0.0 0:22.39 dbus-daemon
> 2338 spikeygg 20 0 1371m 92m 36m S 1 0.1 1:47.21 compiz
> 3239 spikeygg 20 0 1104m 10m 5548 S 1 0.0 1:31.26 mythlogserver
> 122 root 20 0 0 0 0 S 0 0.0 0:05.24 kworker/1:1
> 1233 mysql 20 0 2269m 70m 8088 S 0 0.1 0:08.79 mysqld
> 1288 root 20 0 206m 66m 10m S 0 0.1 2:22.10 Xorg
> 1568 nobody 20 0 21480 1412 1064 S 0 0.0 0:11.89 LCDd
> 2934 spikeygg 20 0 412m 5340 3952 S 0 0.0 0:26.34 indicator-appli
> 5502 spikeygg 20 0 904m 293m 35m S 0 0.2 2:09.67 firefox
> 6457 spikeygg 20 0 17468 1416 968 R 0 0.0 0:00.48 top
> 1 root 20 0 24544 2544 1352 S 0 0.0 0:02.63 init
> 2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd
> 3 root 20 0 0 0 0 S 0 0.0 0:00.25 ksoftirqd/0
>
> What gives, any idea why I have load averages of nearly 2 when the CPU is
> basically idle with 1 running job??
Load is the number of running processes, plus processes that want to run
(in the run queue), plus the number of uninterruptible sleeping
processes. The last group is typically those processes waiting for I/O
operations to complete. You have several processes that could be doing
I/O, like mythbackend, mythlogserver and mysqld. So, a load average of
just less than 2 seems reasonable.

Keep in mind that the load is not determined continuously, but is
calculated from samples taken at intervals and averaged over 1, 5, and
15 minutes. Due to the possibility of sampling error, the load numbers
may not exactly reflect activity you otherwise observe.

Keith

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://www.mythtv.org/mailman/listinfo/mythtv-users

MythTV users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.