[nSLUG] Wither workstations

George N. White III aa056 at chebucto.ns.ca
Sun Feb 13 15:20:18 AST 2005


On Sun, 13 Feb 2005, Dop Ganger wrote:

> On Sat, 12 Feb 2005, George N. White III wrote:
>
>> 1. Have any of you shared an office (not air-conditioned) with a recent
>>     dual-processor G5 or Xeon or AMD workstation?  I suppose it could be
>>     quite cozy in winter.
>
> Not *quite* what you're looking for, but I had an HP DL360 (dual 3.4GHz
> Xeons, I think) on my desk for an afternoon. Cozy isn't the word for it,
> I think "stuffy" would be more apt. Also, I was really glad I had a set of
> earplugs with me.

I get the impression that you can't have an loose papers around these
things or they will get sucked onto the fan grill.  Moshe Bar says his
dual G5 even sounds like a vacuum cleaner.

>> 2. there is lots of info about using rackmount systems to build clusters
>>     or for server farms, but not much about using them to replace
>>     workstations.  Some cluster technology would be useful (central
>>     management of updates, etc.).  I recall reading about some sites (e.g.,
>>     stock exchanges) with very high density of workers that use rackmount
>>     systems with graphics cards and special cabling to connect keyboards
>>     and monitors.  In our case it will be hard enough getting enough
>>     ports on the switch to accomodate doubling the number of
>>     interfaces, so special cabling is not an option.
>
> The problem I can see (with the video cabling, at least) is having
> problems with signal fadeout, or having to buy signal boosters (which have
> always caused problems at high resolutions in the past, but that problem
> may be solved by now). Can't you just export a display back from the
> cluster to run the software on a cheap local desktop? I think that Sun's
> Grid Engine can handle balancing out the load for interactive use, sothat
> may be worth a look.

I think exporting the display will work for most users, but if the
rack mount machines are used as headless workstations then
we will want to bring switched ethernet to every machine (as opposed
to the beowulf approach where you put a switch on the rack for
message passing and the only the master node is connected to the
intranet).

I've been reading up on grid technology.  Our workload is a mix of 
compiles, visualizations, with lots of fairly routine batch processing. In 
principle load balancing tools should help us get some extra batch 
processing done outside working hours, but in practice a lot of effort 
goes into managing disk space.  A typical batch processing run reads a 
terabyte and generates 100GB of output -- some people just fire off 
jobs until they run out of space, clean up the mess, and start over with 
the first job that bombed.  That's OK on a "personal" workstation, but I'm 
not sure the existing tools are robust enough to let those jobs loose on a 
grid. Looking at some of the published experiences with real grids, 20-30% 
of jobs fail due to running out of disk.

-- 
George N. White III  <aa056 at chebucto.ns.ca>


!DSPAM:420fa88d58493158213597!




More information about the nSLUG mailing list