How To Build a Homebrew PS3 Cluster Supercomputer 211
eldavojohn writes "UMass Dartmouth Physics Professor Gaurav Khanna and UMass Dartmouth Principal Investigator Chris Poulin have created a step-by-step guide designed to show you how to build your own supercomputer for about $4,000. They are also hoping that by publishing this guide they will bring about a new kind of software development targeting this architecture & grid (I know a few failed NLP projects of my own that could use some new hardware). If this catches on for research institutions it may increase Sony's sales, but they might not be seeing the corresponding sale of games spike (where they make the most profit)."
"super" computer: (Score:1, Interesting)
Last year, Khanna's construction of a small supercomputer using eight Sony-donated Playstation 3 gaming consoles made headlines nationwide in the scientific community.
I'm not trying to be a smartass, but why did he mention in TFA that his supercomputer cost $4000 if the 8 consoles were "Sony-donated"? ALso, like the iPod example at the top of the post, most research use of the technology won't come from actual iPods or consoles but from customized versions of the underlying technology such as the Opteron-Cell hybrid Roadrunner [wikipedia.org] supercomputer. If one wanted to build their own home "super" computer then why not just use CUDA and a few Nvidia cards?
ibm (Score:2, Interesting)
Why PS3s? (Score:2, Interesting)
Re:Why use PS3s? (Score:4, Interesting)
Re:Pretty much useless (Score:5, Interesting)
How is it useless, when the guy who built it, used it already for a month? And it has replaced 200 supercomputer nodes, for his purpose? I'd say that's very fucking useful.
But you know what, maybe you should send him an e-mail and try to convince him how his cluster is useless. Make it a nice, insightful and intelligent e-mail, like your post.
Re:Why use PS3s? (Score:3, Interesting)
Power and maintenance? (Score:3, Interesting)
Re:Power and maintenance? (Score:1, Interesting)
Researchers pay not only for the initial capital outlay required to install a supercomputer, but also for its power, cooling, the building it resides in, and its maintenance.
Someone noted that "AFRL did it already". While doing it, they noted and were able to take advantage of:
1) low power consumption compared to a typical 1U unit (or desktop PC) used in clusters. Very low draw when in idle mode.
2) low heat output, and engineered to be tolerant of warm/hot environments found in typical users' homes.
3) low maintenance, with a consumer system designed for low MTBF, and easily/cheaply swapped out if a unit is a lemon.
The AFRL lead investigator also noted that a PS3 ain't the end all/be all. Given the lack of ram, "his" cluster was set to spawn off specific types of processing that a Cell CPU with middling ram could do a good job on, to get a good FLOP/$ ratio. The AFRL cluster mixes in a number of x86 1U systems as job masters.
Re:"super" computer: (Score:3, Interesting)
basically it comes down to the costs of having your own personal power station in the TCO to run a cluster.
this started (well, really hit it off) a few years back, when the pentium M and centrino tech became widespread. basically, to my knowledge, it was the first time you could actually have more processors with less jiggahertz, that consumed less power in total and still had more flops than the others. it swayed everyone from "more powerful cpus plz" train of thought to the "more cpus, less power-consumption". (also cpus/chips in general will eventually hit an upper barrier, making parallel computing a necessity)
I haven't checked the facts on the ps3, but seeing how much nether-region sucking is going on, ps3s probably fit into this scheme.
Re:"super" computer: (Score:3, Interesting)
Answer: PS3s were used because of the vector processors - they are significantly faster than general purpose CPUs for some of Dr. Khanna's needs and the general vision of the project. These are chips designed for raytracing which makes them perfect for some forms of scientific processing.
Also a rack unit full of PS3s looks way cooler than some crufty old PCs pulled from a dumpster.
Josh - PS3Cluster tester
diy distributed computing (Score:2, Interesting)
My setup for PS3Cluster (Score:3, Interesting)
I helped Chris with the documentation, testing and image capture on this project. I see some "it doesn't do this!" comments above - please remember this is a young project that started out of one researcher's need to solve a specific type of problem. If you want to see this advance, it's all open source so start hacking.
So my setup:
1 40Gb Playstation3 w/ HDMI cable out and keyboard
Hauppauge HDPVR digitizer
PC running Windoze and Photoshop
TV hanging off the HDPVR for reference
Software as described on PS3Cluster.org including Geoff's Cell libraries, boot image on USB and Fedora 8 for PPC.
Plugged everything together, installed Fedora 6 the first time around since we knew that worked, then redid it with Fedora 8. Added the MPI libraries and ran the little Pi test code. Digitized the whole install as video, proofed out the process in terms of instructions. Did frame grabs from the video, cropped etc in Photoshop. Lots of work, totally worth it seeing the project posted here.
Oh, and it runs X - kinda cool having Firefox running on a game deck.
Enjoy,
Josh
Re:Limited use (Score:4, Interesting)
I tell you what: you go ahead and buy $4000 of those Dual core kits, and we'll compare your output from a well-written algorithm versus the Cell system designed by this team.
Some interesting code examples for using the Cell have been demonstrated and it has immense processing power that most people don't recognize immediately. Check out this Dr Dobb's Journal article [uni-erlangen.de] for an example.
Re:Why doesn't Sony make it ? (Score:3, Interesting)
Well the funny thing is they're still losing money on every PS3 sold. At least from the last cost analysis I saw, which was back in July. They are counting on you watching Blu-Ray disks or buying games, and any PS3 in a computing scenario won't be doing any of that (barring someone 'misallocating resources' *koff koff*).
So your idea makes sense. They are partnered with Toshiba to produce the low(er) cost Cell add-ons outside of a PS3. And you should see http://en.wikipedia.org/wiki/Cell_%28microprocessor%29#Possible_applications for some others.
The sad thing is that a Cell is far more suited for supercomputing than it is for writing games on, but Sony seems insistent that the primary purpose is making game devs's lives miserable while IBM and Toshiba seem more focused on using it where it's actually useful.
This is probably a silly question (Score:3, Interesting)
I guess that either the PS3 has a PPC chip as well, or it runs some sort of emulation mode. I can't find either documented.