[Beowulf] commercial clusters

[Vincent Diepeveen]

Orion computer falls in total other line of computing.
It falls in the region of users that's on this list.

Orion computer is interesting for those who wanted 192GB ram
and it runs linux. So for embarrassingly parallel software that is
number crunching it is interesting.

No average joe knows the word 'linux'.

For them an astronaut course is easier than learning linux.

So that line of machine is not for the 'average joe'.

Just based upon computing power it's not worth buying an orion $100k
machine either.

You need to offer a good deal, not something that is very expensive.

I would have bought myself an orion machine would it offer 192 cpu's
for $10k with inside 2 GB of ram.

Then i work very hard for 2 years to get my program to work over
gigabit ethernet with those ugly latencies.

*that* is a good deal.

It released end of 2005 and the latency from node to node is pathetic,
so it is quite hard to run any software at it that the average joe likes in
a faster manner than that you can run it at a fast A64.

Additional it is 1500 watt. Too much for average joe.

With a dual opteron dual core with 680 watt i can already heat this entire 
office room
(ok it isn't winter yet), provided i do not open the window. With 1500 watt 
it produces
more heat than 3 people do in a living room and i bet it makes a huge sound 
too.

The hardest thing to make on the planet is a good toy that works for the 
average joe.

Vincent

----- Original Message ----- 
From: "Jim Lux" <James.P.Lux at jpl.nasa.gov>
To: "Vincent Diepeveen" <diep at xs4all.nl>; "Robert G. Brown" 
<rgb at phy.duke.edu>; "Angel Dimitrov" <stormlaboratory at yahoo.com>
Cc: <beowulf at beowulf.org>
Sent: Friday, September 29, 2006 5:23 PM
Subject: Re: [Beowulf] commercial clusters


> At 08:32 AM 9/29/2006, Vincent Diepeveen wrote:
>
>
>>The above is the biggest problem. That's why you need good software that
>>'hides' the supercomputer. Basically they want with their windows PC start
>>your program (see attachment) and click somewhere and then run on a big
>>supercomputer (for the average guy on the street a big supercomputer is
>>everything that 2 hands cannot lift).
>>
>>If it was possible to build your own cluster in easy manner and then run 
>>for
>>example a chessprogram at it in a user friendly way,
>>there would be 100k+ clusters right now of 64 cpu's and more.
>>
>>Especially now that 1 chip hardly gets faster, we must think in future 
>>more and more
>>in clustered manner.
>>
>>Just doing simple math what i own myself:
>>
>>1996 : 200Mhz p6
>>2006 : 2.4Ghz opteron
>>
>>both are 3 instructions per cycle processors.
>>So hardware guys won exactly factor 12 in raw processing speed.
>>
>>Moore's law: each 18 month doubling in transistors.
>>deduction from that: doubling in speed.
>>
>>Would mean we are faster now a 106 times  ( 2 ^ 6.66 )
>>So somewhere a factor 8 is missing.
>>
>>The average user feels very well that there is no doubling in speed each 
>>18 months.
>>
>>So offering solutions to normal users to get their favourite program 
>>executed faster in an user friendly manner is definitely a big market.
>
> And this is what companies like Orion are targeting.  Specific classes of 
> users who need lots of computation, with a specific program, and with $10K 
> to $100K to spend.  I don't know that there are "consumer" applications 
> like this yet, but in the generalized engineering world, there are a 
> number of finite element codes for electromagnetics, structures, CFD, and 
> thermal analysis that are widely used.  These codes take quite a lot of 
> computation, so something that is a truly "turnkey" speed up by a factor 
> of 10 or 100 is worth it.  Turns a "load and run the model overnight" kind 
> of operation into a "run the model while I get coffee" operation.
>
>
> Jim
>
>