[Simh] VAX: CPU Usage and Network Timeout
Bishop, Lewis
lewis.bishop at eds.com
Thu Jun 9 17:03:51 EDT 2005
Sure - but bear in mind that this was for Windows servers. I think I posted
this a while back but I tried a piece of software called Relevos and whilst
it worked well it was a commercial product and had a near US$400 price tag.
After looking around I found threadmaster (http://threadmaster.tripod.com/)
which does pretty much the same - the only thing I didn't like about it was
that it limits everything except if where you exclude it - whereas relevos
limits what you define... Still I can live with that. ;-)
Lewis Bishop
---
EDS Ltd, 8 Gilmer Terrace, P.O. Box 3647, Wellington, New Zealand
NZ Applications Delivery Unit Infrastructure Specialist - Oracle Database
Administrator
DDI: +64 (0)4 474 5423 (internal x5423)
Mobile: +64 (0)27 246 2787
Fax: + 64 (0)4 474 5565
Email: lewis.bishop at eds.com
-----Original Message-----
From: Aaron Walsh [mailto:walsham at jmusa.com]
Sent: Thursday, 9 June 2005 10:54 p.m.
To: Bishop, Lewis
Subject: RE: [Simh] VAX: CPU Usage and Network Timeout
Would you be willing to share what you did to create the "CPU limiter on
the emulator"? That sounds exactly like what I need.
Aaron
>>> "Bishop, Lewis" <lewis.bishop at eds.com> 6/8/2005 10:24:12 PM >>>
Yes, I saw a similar effect with a windows server that was running
exchange
server - was getting all kinds of connection problems over the network
whilst the emulator was running and maxing out cpu - although the cpu
was
dropped when required by other processes it seemed to have an adverse
affect. I solved this by using a cpu limiter on the emulator. So yes,
I
could almost definitively say that in my case the high cpu usage was
causing
network related issues.
Lewis Bishop
-----Original Message-----
From: simh-bounces at trailing-edge.com
[mailto:simh-bounces at trailing-edge.com]
On Behalf Of Aaron Walsh
Sent: Thursday, 9 June 2005 1:59 p.m.
To: newsham at lava.net
Cc: simh at trailing-edge.com
Subject: Re: [Simh] VAX: CPU Usage and Network Timeout
I did pay for the whole CPU, but I also want the CPU to last without
overheating ;) My understanding is that nice and renice only control
the priority of the process. But if nothing else is running on the
box
I get the feeling that nice wont stop SIMH from using the whole
processor. Overall, I am not too concerned with the CPU usage, I am
more concerned with the network drop outs I am seeing. I did some
more
research by pinging both the linux ip address and the emulated vax ip
address from a remote machine. While the linux IP address always
responds back within 1-5 milliseconds the emulated ip takes longer.
Which I am still okay with. The main problem is that every minute
(and
I mean on the minute, for example at 8:30.02, 8:31.02, 8:32.02, etc)
for
one to three seconds the emulated IP doesnt reply to any ping or any
other network communication. I have tried all kinds of variations of
tcpdump to see if I can track down what the issue might be, but I have
had no luck.
Has anyone else see this, or have any ideas how I can track it down.
Thanks
Aaron
>>> Tim Newsham <newsham at lava.net> 6/8/2005 4:12:14 PM >>>
> 2)CPU usage is about 100% on the Linux box. I have searched google
and
> found that is to be expected. I tried using the code below in the
> sim_console.c file but regardless of the value of timeout.tv_usec is
> used the emulated VAX was VERY slow.
[...]
> Anyone else have any ways to have simh use less CPU?
Why would you want your machine to use less than 100% of its cpu?
I mean, you paid for the whole cpu, might as well use it. If
you're concerned about other processes getting more share of
the cpu, see the nice(1) and renice(p) man pages.
> Aaron
Tim Newsham
http://www.lava.net/~newsham/
This message has been scanned for viruses by MailControl -
www.mailcontrol.com
_______________________________________________
Simh mailing list
Simh at trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh
More information about the Simh
mailing list