Slashdot Mirror
First 96-Node Desktop Cluster Ships
Panaphonix writes "The Register reports that Orion Multisystems is shipping the first 96-node desktop cluster. 'With the new, larger system, customers get pretty much the most powerful computer around that can plug into a standard electrical socket.' According to the spec sheet, the DS-96 runs Fedora Core 2 and gets 110 GFlops sustained, 230 GFlops peak."
I FAIL IT!
Its not duped, this article has been clustered.
sheep.horse - does not contain information on sheep or horses.
finite bandwidth between processors makes it impossible to sustain anywhere near peak performance for most real-world applications.
Linpack is what is usually used to measure sustained performance on HPC systems.
Nothing to see here; Move along.
Sustained: Riding down the street.
Peak: Taking it off some sweet jumps.
You want to be impressed heh? Well, it is powerfull enough to play solitaire on Longhorn!!!!
10 ?"Hello World" life was simple then
The theoretical max gives a rough estimate of the raw Floating-point power for all of the processors on the system. You pretty much add up the GFlops potental for each node (not exactly, but pretty much). The sustained and demonstrated GFlops of the cluster is based on the Linpack benchmark. The reason there is such a huge difference between the two numbers can be a result of a few factors. 1) The interconnect is GigE and Linpack tends to make use of Message passing comms which are affected adversely by the latencies produced by the GigE connections (myrinet would have been a good choice, but I suppose it was probably impossible to squeeze that into that case) 2) Memory speeds also are a factor as pushing floating point numbers around involves memory. This cluster isn't using anything fancy when it comes to the memory and I suspect this may be another cause for this.
When they say that this line of clusters can "make or break" Orion, I am right now, leaning for broke. For the cost of this machine, one can get a real cluster with a lot more performance. I know this thing is nice because of the power requirements and the fact that you don't need a dedicated server room to store it, but for $100,000, you can get Microway to build you a pimptacular cluster with Dual-Opteron nodes, high-speed memory and a phat interconnect with either myrinet or infiniband. You will get a lot more work done for the same price.
$100,000/96 = ~$1,000.
Not a bad deal.
paintball
I know this thing is nice because of the power requirements and the fact that you don't need a dedicated server room to store it, but for $100,000, you can get Microway to build you a pimptacular cluster with Dual-Opteron nodes, high-speed memory and a phat interconnect with either myrinet or infiniband. You will get a lot more work done for the same price.
You forgot a couple of things:
* HVAC costs
* Realestate costs
Remember, this is a deskside cluster. Try that with your dual-opteron cluster. And try adding up all the costs.
Have you got your LWN subscription yet?
I can't hear you, the guy two cubicles over just fired up his new Opteron cluster. I'm just trying to hold on to my desk!
He's also forgotten about hiring an extra 2 guys at least to maintain the extra machines. A selling point of this box will be that "it just works". Pay for a support contract and wammo, you've got a cheap low maintainence cluster. For people working on top-secret stuff (who else needs clusters? ;-), hiring people is a risk and the vetting process is expensive.
You can tell you're on /. when dividing 1e5 by 1e2 to get 1e3 gets modded up to +5 insightful. :-)
Although having 96 nodes in a single box makes it quite cute, from what I can interpret from the specs, you would get more bang for your $100K by getting what the beowulf crowd like to call MMCOTS (Mass-Market-Common-Off-The-Shelf, i.e. mass produced computers from Dell or the like), hooked toghether with a specialty high-bandwidth low-latency interconnect like Infiniband, Myrinet or SCI. Running a free beowulf cluster OS like for instance ROCKS would mean that a normal linux admin could maintain it quite effectively.
I expect this thing to be marketed towards scientists in small or medium businesses that aren't employing many/any IT staff, who use commercial computer models to do things like theoretical chemistry (Gaussian, ADF etc), bioinformatics (Phase, BLAS, Paralign etc), fluid dynamics, statistics, crypto, you name it. I don't expect to see any of these types of systems used in normal supercomputing sites, where people write their own (parallel) code and skilled staff maintain the cluster.
-- Buzh
Parent poster messed up on their calcs. Current XServes are 18.4 GFlops peak, not 35 eg Virginia Tech currently at #7/500 is 20240 GFlops peak for 1100 XServes. So 7 would only be ~129 GFlops peak, and 33 would be 607 GFlops peak. But not exactly fitting in a single tower case - though 8 would fit nicely one of those mini sound-padded racks which would be almost as good. And at least the last time I saw a price comparison made, the G5s were far cheaper than comparable rack P4s. (The G5 has 2x the FP hardware).
Serious question here. Does production software exist to drive arbitrary computation across a GPU? I've seen articles about software on its way, etc. Does it exist, either as an application or integrated into some OS? Man, if I could push some of my statistical computing off to the GPU...
One of my best friends just bought a tiny little house in downtown Toronto for $377,000. I left Toronto last November and moved to Santiago, Chile and live downtwon where my rent is $260/month, for quite a nice, though small place, in an excellent area.
So, if I spend $100K on the Orion DS-96, that leaves me more than enough for a 250 channel geodesic EEG system which would allow me to compute self-organizing maps of the human mind based on flashing the 1.6 million mindpixels I have collected over the past five years to various volunteers [english teachers], AND still have 56.73 years worth of rent left!
Too bad no bank will loan me $377,000 for a computer and an EEG system and the time to play with it...