[Beowulf] Repeated Dell SC1435 crash / hang. How to get the vendor to resolve the issue when 20% of the servers fail in first year?

[Jan Heichler]

Hallo Rahul,

Montag, 6. April 2009, meintest Du:

RN> On Mon, Apr 6, 2009 at 7:46 AM, Chris Samuel <csamuel at vpac.org> wrote:
>> Even though we could reproduce it on 64-bit Debian
>> and 32-bit CentOS they wouldn't escalate the issue
>> until we could reproduce it on RHEL5 - which we did
>> today.


RN> Thanks for sharing the anecdote Chris. I wonder if there is any clause
RN> in the contracts restricting us to run certain OS's.

RN> So long as we are using a "reputable" well-tested OS I find it unfair
RN> that the vendors engage in so much arm-twisting. Is there any
RN> scientific evidence that the core kernels of Debian or Fedora or CenOS
RN> (that *is* essentially RHEL isn't it?) are any less reliable then
RN> RHEL? What is / are the distros of choice on the Beowulf community?
RN> Just getting a feel.

RN> I have tried out cutting edge distros meant for scientific
RN> applications like ScientificLinux or ComputeNodeLinux but I've found

ScientificLinux is - in fact - RHEL. So it is not more "cutting edge" than CentOS... 

RN> it more practical to stick to a larger, well used distro. No doubt I
RN> might take some performance cuts on the benchmarks but the simple
RN> reality of a larger user community out there makes it easier to debug
RN> stuff and get well-tested apps and code that will run on my Distro
RN> out-of-the box.

I See that ScientificLinux has a huge community using it - at least in Europe. Some big research institutes are using it (CERN for example) - some projects set it as the default distro (D-Grid - a german GRID project). 

So i don't see you point here ;-)

Jan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.scyld.com/pipermail/beowulf/attachments/20090406/7089ccbd/attachment.html