Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 03 Jul 2008 08:54:12 -0700
From:      Peter Giessel <pgiessel@mac.com>
To:        John Almberg <jalmberg@identry.com>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: Does 'top' work on multi-processor systems?
Message-ID:  <3F8B80E5-011A-1000-94A3-A1AA554CE4A1-Webmail-10016@mac.com>
In-Reply-To: <46194DB4-E56D-4F9B-ABDF-2D76CBBF6C41@identry.com>
References:  <46194DB4-E56D-4F9B-ABDF-2D76CBBF6C41@identry.com>

next in thread | previous in thread | raw e-mail | index | archive | help
If I recall correctly, on dual processor systems, 200% is full processor utilization,
so on an essentially 8 processor system, 800% would be full processor utilization.

157% in top would actually amount to about 20% of your full processor power.
 
On Thursday, July 03, 2008, at 05:41AM, "John Almberg" <jalmberg@identry.com> wrote:
>I have a 3 month old server with two quad-core processors, 8G of RAM,  
>and an array of fast hard drives. The two main applications are web  
>server and mail server. There are only about 20 small-business  
>websites and approx. 40 email accounts on the server. i.e., not much.
>
>In terms of actual usage, performance is great. Web pages load fast,  
>and email is processed quickly. And the 92 days of up time says that  
>this server has been up 100% since it's been installed in colo.
>
>However, 'top' shows a fairly high load (see below). If when I leave  
>top running for a while, I see the load average spike up to  7 or 8  
>occasionally. However, this doesn't translate into slow  
>performance... pages still load quickly.
>
>Also, what's up with that 157% WCPU for the mysql process? That just  
>seems wrong. The WCPU number for mysql has been stuck up above 100%  
>for a few weeks... it seems like something is broken there.
>
>On my previous single processor system, top was a good rough  
>indicator of how the system was doing. But it doesn't seem to work  
>very well on this 8 core system.
>
>My best guess is that the bogus mysql number is also throwing off the  
>load averages, making them higher than they really are, but that's  
>just a guess.
>
>Any thoughts? Is there a better tool for measuring load?
>
>-- John
>
>
>last pid: 43730;  load averages:  1.93,  2.64,   
>2.22                                           up 92+19:45:54  09:26:27
>238 processes: 3 running, 235 sleeping
>CPU states:  8.1% user,  0.0% nice, 17.3% system,  0.2% interrupt,  
>74.4% idle
>Mem: 1384M Active, 3753M Inact, 373M Wired, 884K Cache, 214M Buf,  
>2150M Free
>Swap: 16G Total, 88K Used, 16G Free
>
>   PID USERNAME  THR PRI NICE   SIZE    RES STATE  C   TIME   WCPU  
>COMMAND
>  1393 mysql      63  20    0   400M   221M kserel 0 191.5H 157.13%  
>mysqld
>43698 www         1   4    0   169M 29888K sbwait 5   0:00  2.63% httpd
>43697 www         1  20    0   169M 29804K lockf  1   0:00  1.18% httpd
>23376 vpopmail    1   4    0 81468K 55772K select 7   0:28  1.17%  
>perl5.8.8
>43729 root        1  96    0  7228K  2676K select 5   0:00  1.00%  
>couriertls
>43695 www         1   4    0   169M 29768K sbwait 5   0:00  0.67% httpd
>43417 www         1   4    0   170M 31340K sbwait 7   0:00  0.20% httpd
>85622 root        1   4    0 98588K 68764K select 5   7:54  0.20% ruby
>43325 www         1  20    0   170M 30412K lockf  7   0:00  0.15% httpd
>  6352 root        1   4    0 97660K 67784K select 3   1:04  0.10% ruby
>42848 www         1   4    0   169M 30004K sbwait 4   0:00  0.10% httpd
>43111 www         1  20    0   170M 30336K lockf  2   0:00  0.05% httpd
>
>_______________________________________________
>freebsd-questions@freebsd.org mailing list
>http://lists.freebsd.org/mailman/listinfo/freebsd-questions
>To unsubscribe, send any mail to "freebsd-questions-unsubscribe@freebsd.org"
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F8B80E5-011A-1000-94A3-A1AA554CE4A1-Webmail-10016>