you have to prove your case first by upgrading the memory of one of the
low end node by using the memory of the other node (that is if you don't
have any spare). then insert the upgraded node back to the cluster, and
then let it handle a test job and observe.
if the problem persist, you have to re
On Sun, 2007-10-14 at 07:24 +0800, Michael Calizo wrote:
> Its a HIGH Performance Computing (HPC). To elaborate the setup, we
> have a thinanywhere setup for user to use and to submit jobs to those
> computing nodes. The management node will then push the those jobs to
> available computing node
Glennix,
You're probably referring to an internal modem. Try linuxant.org have tried
once on the "so-called winmodem" it gave a lot of blood hours trying to make
it work with different flavor of linux but i got it going til the modem
died. I forgot man, i know linuxant drivers are not free but th
http://news.zdnet.com/2100-3513_22-6213270.html -- and here's a
raspberry for Ballmer.
--
Daniel O. Escasa
independent IT consultant and writer
contributor, Free Software Magazine (http://www.freesoftwaremagazine.com)
personal blog at http://descasa.i.ph
If we choose being kind over being right,
Its a HIGH Performance Computing (HPC). To elaborate the setup, we have a
thinanywhere setup for user to use and to submit jobs to those computing
nodes. The management node will then push the those jobs to available
computing nodes.
Cron job is not an option because we can not just kill/restar
can you be more specific to your setup? is it an HPC or HTC?
can you also elaborate on your problem? does the job stays
idle on the low-end node? the way you deal with the problem
is the typical way of responding to such but be done automatically
via the job scheduler. and since you've already iden
Hi Guys,
A newbie here needs an expert opinion regarding Linux HPC.
In my current company we have a Linux(Redhat) cluster implementation, say
100 nodes per cluster.
I notice that on the problematic cluster, some nodes are low end server say
2GB memory while the
other nodes have 4GB memory. This p
nagios is the best. pero pagmadalian pero mo ng graphical output I suggest
cacti :)
On 10/9/07, thad <[EMAIL PROTECTED]> wrote:
>
> nmon,
>
> http://www-941.haw.ibm.com/collaboration/wiki/display/WikiPtype/nmon
>
> hth,
>
> On 10/1/07, jon robles <[EMAIL PROTECTED]> wrote:
> > Hi Guys,
> >
> > Wha
Hi, I run into that question before and after goggling I found this useful
site. Actually you need a combination of commands and a little bash skills.
But an example provided is enough for me :) I hope this helps.
http://www.performancewiki.com/diskio-tuning.html
On 10/14/07, Gerald Timothy Quimp
hi all,
Occasionally I'd like to know which application(s) is/are using the most
disk bandwidth. The reason is, on my desktop and my laptop, I often
have background jobs which do a lot of IO and I'd like to know which
program it is exactly so I can take appropriate action if IO is slowing
down in
To all who replied on my posted problem, thank you very much. I
finally fixed the issue last week by doing up2date upgrade of the
kernel version. I used the RHEL 4.0 CD to install and it has this
issue, Red Hat fixed it with 2.6.9.55 kernel version.
thad
On 9/22/07, Eduardo Tongson <[EMAIL PROTEC
11 matches
Mail list logo
