[Beowulf] A Cluster of Motherboard.
Robert G. Brown
rgb at phy.duke.edu
Fri Nov 11 06:58:01 PST 2005
On Fri, 11 Nov 2005, Jim Lux wrote:
> If you have an unusual packaging concern.. maybe it needs to be in a sealed
> box? maybe it has to fit into that antique grandfather clock?, etc. then
> you've got a case for custom design.
Or if the commercial packaging available is still being sold at a hefty
premium, opening up a potential window of CBA advantage.
Let us recall that all of these arguments might have been applied to the
creation of beowulfs in the first place, when perfectly good big iron
supercomputers that were de facto clusters were available. After all,
it took lots of human time to set up that stack of PC's, buy shelving,
screw around with wiring them together, and then (from a certain amount
of bitter experience) wrestle with Linux (especially e.g. the SMP
kernels and device driver issues) to get the whole thing to "work".
The reasons that it >>was<< economically efficient to build beowulfish
clusters as opposed to buying big iron supercomputers was the factor of
10+ in price-performance of the actual hardware, amortized over a
relatively long lifetime (years of service). The big iron
supercomputers also charged a premium for service and required
significant human management costs to operate, so it wasn't even clear
that there was a win THERE by going that way. This left lots of room
for billed-cost labor to STILL leave a significant price advantage
(order of magnitude ten, not even unity). Opportunity cost labor is
trickier, but in most cases one would argue that it increases the
cost-efficiency of clusters still further, at least in certain ranges of
cluster size and with certain degrees of OC expertise available.
Until very recently, I would have put bladed computers into the same
category -- an easy factor of 2 higher in price in order to achieve
higher (maybe) CPU densities, plus some further pros and cons -- serious
problems with cooling and powering relative to the already nontrivial
problems with ordinary 43U racks, compression of choice for network,
disk, peripherals, video, possibly some improved infrastructure and
management support at the hardware/firmware level. So going with
homemade "blades" in situations that required higher densities than one
can comfortably achieve with racks using COTS micros might well have
been a signficant cost saver even with the most honest CBA.
That appears to be less true as blades are moving more towards the mass
market and some low rent "micro board on cheap plate blades" appear --
pretty much what you'd make yourself but done professionally and at an
efficient scale, sans bells, whistles, and hefty markup.
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
More information about the Beowulf