[torqueusers] scalability of 2.1.2?

Lloyd Brown somewhere_or_other at byu.edu
Fri Aug 11 14:48:06 MDT 2006

Garrick Staples wrote:
> On Fri, Aug 11, 2006 at 10:19:44AM -0600, Lloyd Brown alleged:
>> were missing.  For us, the server had an MTU of 4004, the clients of
>> 4000, and the threshold happened between 281 nodes and 282 nodes.
> Is that a weird MTU value?  I've only seen 1500 (normal) and 9000
> (jumbo frames).
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers

I agree.  It is a bit strange of a value.  That's the way the cluster
was when I got hired, so I assume there's a reason.  I'm just still
trying to figure what the reason is.  The hosts will do whatever you
tell them, and the switches in question support anything from 594 to
9252, at least according to their documentation.

The problem is that we've had a lot of staff turn over in the last year.
 None of us were here when the system was first built about a year and a
half ago.  We're still figuring a lot of this stuff out.

Seems to work, though, so for now, I'll leave well enough alone.

Lloyd Brown
BYU Supercomputing

More information about the torqueusers mailing list