[Beowulf] RE: Capitalization Rates - How often should you replace a cluster? (resent - 1st sending wasn't posted ).

[Gus Correa]

Huw Lynes wrote:
> On Wed, 2009-01-21 at 12:33 -0800, Greg Lindahl wrote:
>> On Wed, Jan 21, 2009 at 10:57:42AM -0500, Mark Hahn wrote:
>>
>>> I guess our premises are quite different.
>> Yes, they are. Hence your statement about "obviously burning money"
>> was a bit of an over-generalization ;)
>>
>> There are HPC systems that want UPSes, namely ones that do
>> mission-critical things, like weather forecasts, and spooky work.
>> Most university HPC systems don't fall into that category. Most
>> business websites do fall into the UPS category. But other than that,
>> my cluster looks pretty much like an HPC cluster, albeit one with
>> only gbit ethernet and lots of local disk.
>>
> 

> Our new HPC data-centre is completely UPSed. With the UPS sized to
> bridge to the diesel generator that protects the entire building. We may
> be a slightly special case in that the mains in Cardiff is horrifically
> dirty. We've seen very expensive pieces of killed by power spikes on a
> couple of occasions. 
> 
> Our current monitoring shows the UPS cutting over to battery at least
> twice a month. So at least in our instance UPS protection is worth every
> penny.
> 
> 

Hi Huw, list

Regarding bad power, Cardiff is not so special.  :)

You're not alone.
We have aerial power lines serving our suburban/rural campus,
with thunderstorms, fallen tree branches,
or electrocuted squirrels, spiking, shaking,
or cutting power many times a year.
The buildings have diesel generators,
not only because of computer rooms, but also for
expensive/delicate lab equipment,
protected by UPSs of several sizes and types,
from big Lieberts to small VA units.
With the generators, we only need a modest battery time also.
Our UPSs are worth having for the reasons you presented.
Toasting computers or mass spectrometers with electrical
spikes is no fun.

Gus Correa
---------------------------------------------------------------------
Gustavo Correa
Lamont-Doherty Earth Observatory - Columbia University
Palisades, NY, 10964-8000 - USA
---------------------------------------------------------------------