[Beowulf] Clusters and Distro Lifespans

[Gerald Davies]

On 7/19/06, Joe Landman <landman at scalableinformatics.com> wrote:
>
> Ok.  FC as a distro is supposed to be experimental.  Some releases of it
> are actually quite good.  FC4 has been great on one of boxes (FC5 has
> been a disaster on the same box, which has followed a pattern we have
> observed of FXe being good, and FCo being garbage, where e is an even
> number, o being an odd number).
>

--snip--

Apologies, I should have given more background to the post.

The software we use is centered around FC so switching to other
distros is not really an option I want to consider.  I put forward the
questions because of the FC release cycles and just a general interest
in what other people are doing! :)

Some of the clusters I manage just run Condor and the users can make
do with gcc, but one cluster in particular is split in terms of
schedulers (Condor and PBS), hardware (ranging from duals to 8-way
opterons), a few extra compilers.  This coupled with some proprietary
mechanism for management leaves me slightly concerned about upgrades.
It's a lesson learned when buying new clusters!

>
> Cluster upgrades are rare unless you are missing functionality or
> something is broken.  That is of course one opinion, some here do
> upgrades nightly.  From a purely production oriented viewpoint, where
> downtime == lost money for our customers, we usually advise against that.
>

Thanks for the reply! It's great to get insight into what other people
use and recommend!

Many thanks,

-- 
Gerald Davies
---------------------------------------------
w: http://www.geralddavies.com