Date: Mon, 1 Dec 1997 15:52:33 -0600 From: Chad Scott <chad@txdirect.net> To: "'Justen Stepka'" <jstepka@chaos.winternet.com>, Tom <tom@uniserve.com> Cc: Mattias Pantzare <pantzer@ludd.luth.se>, "FreeBSD-stable@freebsd.org" <FreeBSD-stable@freebsd.org> Subject: RE: CPU Load Message-ID: <B5A8F3E0E4E5D0119C1400805F38B38E091E12@enigma.sat.txdirect.net>
next in thread | raw e-mail | index | archive | help
The "Load Average" on a UNIX machine is the average number of *BLOCKED* processes because of unavailable resources. Whether they're blocked from CPU or Disk I/O doesn't matter. They fact that they are blocked is what matters. A load of 4.0 on an NFS server is *not* normal because you have four processes that are blocked from lack of resources. This means you've got too little hardware to handle the requests you're getting. Chad Scott Network Operations Internet Direct, Incorporated -----Original Message----- From: Justen Stepka [SMTP:jstepka@chaos.winternet.com] Sent: Monday, December 01, 1997 3:50 PM To: Tom Cc: Mattias Pantzare; Justen Stepka; FreeBSD-stable@freebsd.org Subject: Re: CPU Load On Sun, 30 Nov 1997, Tom wrote: > > On Mon, 1 Dec 1997, Mattias Pantzare wrote: > > > On Sat, 29 Nov 1997, Justen Stepka wrote: > > > > > Recently I added memory to my NFS server (dx4-100 now w/ 32 megs of RAM), > > > when I did this the overall system preformance increased dramiticly. The > > > problem that I noticed was that when using NFS/NIS the CPU load climbs to > > > about 4.0+, is there a special reason that this might be happening? > > > > The load value is not CPU load. It is the average number of processes > > ready to run or waiting for disk I/O to complete. > > No, it is the average number of processes that are ready-to-run. > Processes waiting for disk io (or any io) are not ready to run. > > > So it is normal for a NFS server to have a high load, as it is often > > waiting for disk I/O. > > It is normal for NFS servers to have a low load, because NFS serving is > not CPU intensive, and io bound. > > A load of 4+ for a NFS server is not normal. I would suggest > determining which processes are using CPU time. the processes that are taking so much CPU time are the nsfd daemons, there are two of them with a 35%+ WCPU and CPU. Justen Stepka
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?B5A8F3E0E4E5D0119C1400805F38B38E091E12>