Barry Lustig wrote:
On Wed, 15 Nov 2000, Terry Lambert wrote:
I'm always tempted to set up a company where the main
engineers have a centralized batch compile server, so as to
not slow down developement, but requiring that they run no
better than a 386SX/16 on their desktop. If
I guess it might be useful to see the difference between
"true" idle time and time the system couldn't do anything
useful because it was blocked on the disk (but /should/
have done something useful...).
You mean because the programmer didn't interleave their I/O,
and wrote to a
On Thu, 16 Nov 2000, Terry Lambert wrote:
Modern bloat-ware really pisses me off; I built the bind
library the other day: the frigging thing was 4M, unstripped.
How does this affect the (non?-)usefullness of the
%iowait statistic?
When you are waiting for I/O in a well written
On Thu, 16 Nov 2000, Terry Lambert wrote:
Ummm, how about a situation where you have a steadily
increasing work load (more customers?) and want to have
decent statistics of your servers to determine exactly
what parts to upgrade and/or if you need to put extra
machines into service?
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell me how to tell what percentage of
processor time is being spent waiting for disk I/O to complete?
Uh, none?
If there is disk I/O pending, the processor just runs a
On Wed, 15 Nov 2000, Terry Lambert wrote:
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell me how to tell what percentage of
processor time is being spent waiting for disk I/O to complete?
Uh, none?
If there is
On Wed, 15 Nov 2000, Terry Lambert wrote:
I'm always tempted to set up a company where the main
engineers have a centralized batch compile server, so as to
not slow down developement, but requiring that they run no
better than a 386SX/16 on their desktop. If they are good,
I'd give
On Fri, 10 Nov 2000, Terry Lambert wrote:
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell me how to tell what percentage of
processor time is being spent waiting for disk I/O to complete?
Uh, none?
If there is disk I/O
void [EMAIL PROTECTED] writes:
Is there any reason top couldn't add these up and report a %iowait
like Solaris'?
Yes. It would conceal valuable information. Do the adding up in your
head.
DES
--
Dag-Erling Smorgrav - [EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with
In article [EMAIL PROTECTED] you write:
not how busy the disks are. I want relative data, not absolute.
systat -vmstat?
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell me how to tell what percentage of
That was something I was
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell me how to tell what percentage of
processor time is being spent waiting for disk I/O to complete?
Uh, none?
If there is disk I/O pending, the processor just runs a different
On Tue, Nov 07, 2000 at 04:13:30PM +0100, Dag-Erling Smorgrav wrote:
void [EMAIL PROTECTED] writes:
I've been using Solaris a lot lately, and I've noticed that in e.g.
top's output, it has a distinct CPU state called "iowait", which seems
to be a pretty good indicator of how I/O-bound a
On Fri, Nov 10, 2000 at 12:33:31PM +1000, [EMAIL PROTECTED] wrote:
On Thu, 9 Nov 2000, void wrote:
not how busy the disks are. I want relative data, not absolute.
systat -vmstat?
Thank you! This gets the me disk %busy, which is one of the things I
was looking for. Now, can anyone tell
void [EMAIL PROTECTED] writes:
I've been using Solaris a lot lately, and I've noticed that in e.g.
top's output, it has a distinct CPU state called "iowait", which seems
to be a pretty good indicator of how I/O-bound a system is. Is there
any reason that FreeBSD doesn't have such a state?
I've been using Solaris a lot lately, and I've noticed that in e.g.
top's output, it has a distinct CPU state called "iowait", which seems
to be a pretty good indicator of how I/O-bound a system is. Is there
any reason that FreeBSD doesn't have such a state? "iostat" also seems
a lot less
On Tue, Nov 07, 2000 at 05:44:13AM +, void wrote:
I've been using Solaris a lot lately, and I've noticed that in e.g.
top's output, it has a distinct CPU state called "iowait", which seems
to be a pretty good indicator of how I/O-bound a system is. Is there
any reason that FreeBSD
16 matches
Mail list logo