Slashdot Mirror
World's Fastest Macintosh Cluster
gabeman-o writes: "The Grupo de Lasers e Plasmas has created the fastest Apple G4 cluster. The cluster runs on 16 Dual PowerPC G4/450, 32 processors, 12GB of RAM, .5TB of space, and Mac OS 9. Apparently, they have utilized the AppleSeed technology developed by UCLA. According to the website, the cluster will be used for simulating plasmas. Not too shabby!"
The vector unit is only any good if the application has been compiled to take advantage of it, although granted computational fluid dynamics is vector intensive.
To get the same speed from Intel hardware you need something like 50 dual PentiumUmmm. That seems like a very optimistic estimate of how much the altivec unit will help.
And the price for rackmount computers is almost comparable with the Apple G4 dual 450.For the price of one maxed out dual 450 (~$6-10K), I could get 4 or more (dual proc) linux 1U machines.
Second, gigabit ethernet is standard on those machines.The transport bandwidth matters less in clustering than the transport LATENCY. For that you want something like Myrinet.
If they want to they can replace the 100baseT switch with a gigabit switch.Which would only cost about another $5000 or more. Doubt me? Go look on eBay for "cisco catalyst gigabit". And again, ethernet is not suited for this application.
I'm not saying that this isn't a cool project, but by the same token, stating that Macs are the end all and be all of scientific computing is just wrong.
Apple hyped that optical mouse as if they dreamt up the idea. Too bad Logitech (and I believe MS) had it on the market about a year before Apple. Notice how the new Powerbook has built in ethernet; I'm sure Apple will try to claim that they invented that too. At least they ditched the Toiletseatbook styling and went for the innovative notebook look that Compaq invented in the 80s. I am holding out for the next Mac innovation, the iPokemonitor, which will fold nicely into a little ball and convert to a 2.35:1 40" plasma 3d monitor when summoned.
For desktops, maybe. But clusters have completely different economics. The "training" cost is *zero* after the first (or second, depending how you look at it) machine. And the purchase price is perfectly linear, so it becomes a far more important concern.
> Wouldn't it be cheaper to do this with a bunch of PC's in stead?
> Not as pretty but it's someones tax dollars that pay for these...
Fact of life: hardware is cheap. A $3,000 Mac vs. a $2000 PC is peanuts, even if the total difference is 30-40k for the whole installation.
Re-tooling the development environment, retraining IT and programming personell, porting code to the new platform and the cost to set up and configure the new and cheaper hardware far outweighs the cost of the hardware itself. Then there's the time lost in accomplishing all of the above.
If your admins know Mac, and your coders and QA staff are used to building and testing Mac code with Mac tools (MPW, Codewarrior, Macsbug), you go with Mac. End of story.
SoupIsGood Food
I suppose they could not afford a Giganet switch. Those machines support 1000Base-T out of the box.
Interesting that they did not go with the G4/500MP, which was also available.
It looks like they added RAM to each of the machines, but kept the stock 30 gig hard drive.
- (c) 2018 Hank Zimmerman
Your design acumen is absolutely stunning, though.
Do you have anything, at all, to add to the discussion? Or do you just like to talk?
Why does everyone build clusters using normal desktop boxes? Rackmount clusters make so much more sense. Admittedly, I don't think the G4 comes in a rackmount configuration, but surely that makes other options (Sun Netra, Compaq DS10L or noname Intel boxen) more appropriate?
"The invisible and the non-existent look very much alike." -- Delos B. McKown
Maybe if someone can come up with a cool-sounding name then folks will stop calling them clusters
How about Clumps?
Karma karma karma karma karmeleon: it comes and goes, it comes and goes.
Offtopic?!? WTF?
Karma karma karma karma karmeleon: it comes and goes, it comes and goes.
While I know where you were coming with this, you're pretty wrong...
The G4 might replace the mixing board and tape/ADATs/HD, but to record you need a few other things. Examples:
* Microphones - A good studio will have many, some costing thousands of dollars.
* Facilities - As a rule, good recordings do not come out of the living room. Building a studio is expensive, and outfitting an existing building is not much cheaper.
* (Most Importantly) Knowledgeable Staff - A great engineer/producer team can make a decent record from a Tascam cassette recorder. Why don't they? Because these people already work for big studios, where they get paid what they're worth.
There are many great engineers in many great small studios out there. Mix did a report on about 10 of them a few months ago, and there are surely hundreds more, if not thousands.
My point is not "G4 Studios can't do it well" it's "Buying a G4 does not buy you a studio".
--
What about calling it an orchard?
Actually, I was hoping that Apple, not 3Com, would get to rename Candlestick Park in SF. The "Apple Orchard" sounds a lot better than "3Com Park".
Then again, "Leprosy Field" sounds better than "3Com Park"..
.sig a .sog, .sig out loud, .sig out .strog"
".sig,
".sig,
Take a look at that picture on the right. There are two monitors in the frame. Look at the right monitor. Look closely. If I'm not mistaken, that's OS X.
Default OS X background, Dock on the bottom of the screen. Oh yeah, they're using OS X. In a way, this makes sense. They say they're using Pooch, and Pooch is available on OS X.
Very very cool!
IANAKW (I Am Not A Karma Whore), but I have to say that was damn funny. LMAO. "It's funny 'cause it's true!" Homer Simpson. Disclaimer: Ok, OS X is not -that- slow, and I am using it as my primary OS on my Cube (w/ 320MB RAM), but it/GUI/Aqua does, currently at least, have an annoying molasses-like quality to it, which I hope and pray will be improved soon.
Kindof a valid point, tho it's fairly obvious
that said cluck is just a bitter old VMS user.
K.
-
-- Proud descendant of semi-nomadic cattle-herders.
As always, with Macs and PCs, the systems can have a similar amount of power but there's always those glaring differences.
.. PC hardware is much cheaper than Mac hardware. Probably from the economies of scale and the fact there's no ONE supplier of PC hardware that can keep its prices (artificially?) high. Cost is a big factor I think.
.. as soon as you get it out of the packaging .. is already obsolete ;>
Like.. Cost?
There's power usage too, like someone mentioned, an Athlon is a bit a of a power hog, PowerPC chips aren't. That is a very good reason to cluster PowerPC chips. What I'd like is to be able to buy PowerPC chips and motherboards cheaply and plentifully like their Intel/AMD brethren, and run Linux on it - without paying the 'Mac tax'.
There's operating system choices too.. and far too many to get into detail about. Linux is common between both platforms, and the various quirks of Linux, Mac OS and Windows may or may not be someones bag.
Best bet is to play with whatever system you feel like buying beforehand.
I wouldn't worry too much about raw speed, as whatever you buy
--
Delphis
Delphis
With Appleseed clusters you don't need some convoluted set of instructions to set it up, the nodes need only TCP/IP and file sharing turned on and you can use POOCH to control them. You can have a mild mannered room full of graphic workstations during the day and turn them into a super computer at night. Who in the holiest of all holy fucks cares if this cluster isn't on the Top500 list? Supercomputing folk will laugh in your face if you mention Beowulf in the same context as Cray you jackasses. By the way for those of you who are complete fucking idiots, they maintained 50 Gflops using AltiVec aware software which can pump out several values per instruction (SIMD in case you're wondering). If you're going to be using a cluster like this there's no reason not to enable every hardware advantage you possibly can on each of the nodes. As for compatibility you can use MacMPI for C/C++ and FORTRAN code or use distributed object coding in Cocoa in OS X. I guess Linux folk don't appriciate things that actually work how the documentation describes.
I'm a loner Dottie, a Rebel.
Check your math. The G4 listed is only 29% faster than the PIII listed, not 75% or 50% in terms of keys/sec.
If you're doing some kind of weird comparison based on keys/Mhz, it's 189% faster, but I'm not sure that's a comparison that makes any sense since you can't build systems that way.
A more interesting comparison would be keys/$ -- ie, how much cracking power can you buy for a given dollar amount. By a very crude measure, a 466Mhz G4 is $1600 and a 1Ghz Dell is closer to $1000.
If you have $5k to spend, the Dell buys you 28,432,490 keys/sec and the Mac buys you 22,166,475 keys/sec. The Dell system would then give you 5,686 keys/$, the Mac 4,618 keys/$.
Since nobody does these projects for $5k or desktop machines, I'd imagine that the performance gap would actually grow substantially in favor of Intel hardware for the same money spent.
You didn't answer the power question, though. All 32 processors in this cluster are using less power than 8 PIII/1GHz. Can 8 PIII/1GHz CPU's do more computation than 32 G4/450MHz CPU's with Altivec? How much cooling does that room have?
Who gives a shit? The marginal costs of power and cooling for Intel hardware are *trivial* next to the increased performance for dollar invested. If I can make $10 for every 5M keys, why would I buy a more expensive computer that makes does fewer keys to save $50 on the monthly power bill when it would cost me thousands of dollars in lost work? That kind of argument works in outer space, the desert or someplace where every watt matters, but in the real world that most people work in it doesn't mean a damn thing.
Check your math. The G4 listed is only 29% faster than the PIII listed, not 75% or 50% in terms of keys/sec.
Jup, a 450 Mhz G4 is about 29 faster than a Pentium
If we compare a theoretical G4 1 Ghz with the same Pentium
A fully loaded G4 733 Mhz. with a DVD/CD burner, 256 Mb of memomy and standard gigabit ethernet is around $ 3499-.
But you can get a dual G4 533 for around $ 2049- which isn't expensive.
A dual Pentium
Wouldn't it be cheaper to do this with a bunch of PC's in stead? Not as pretty but it's someones tax dollars that pay for these...
Well, due to the fact that the PPC7400 has an Altivec unit it is much faster than any comparable Intel processor.
To get the same speed from Intel hardware you need something like 50 dual Pentium
And the price for rackmount computers is almost comparable with the Apple G4 dual 450.
Second, gigabit ethernet is standard on those machines.
If they want to they can replace the 100baseT switch with a gigabit switch.
I know of a G4 cluster with 96 machines. However, it isn't general public knoweledge. I used to administer it, and if I get permission from the owner, I can discuss it. Suffice to say it's used for image processing.
That was the irony in the joke. A beowulf cluster of beowulf clusters. Doesn't anyone see the humor (or perhaps usefullness) of the idea? I guess distributed computing concepts are only valid for finding alien communications among "I Love Lucy" reruns. :P
Speak truth to power.
Maybe things have changed recently....Several years ago I had to support an art department that used Macs (Then on a 7.x version of MacOS, we upgraded to 8.x while I was there). Compared to the 2 artists who used NT Workstation, the Macs bombed all the time. The Macs were definitely faster on the same apps, but the cost was much higher, plus we were constantly wasting time trying to get extensions "just so", contorting our bodies to reset the PROM, etc....
Sure those little shelves look cool - but how do you think those guys felt when the needed opaque mouse pads for their brand new OPTICAL mice?
Dirt doesn't need luck.
I don't think metrics of support time and training directly translate to this kind of cluster. These machines are being used in an unconventional way; you don't really have to worry about somebody screwing up the extensions/dlls/shared libraries on just one of the boxes, for instance.
What's more important to them is the cost of development of software for the cluster. And if they've got enough Mac weenies around to argue for the Apple hardware solution, they probably have the talent and dedication to make it work---so it's probably a pretty cheap solution. This is also why x86 Linux Beowulf clusters work well in some environments...people are excited and knowledgable about a technology that they're recommending.
As far as the TCO numbers go, I'd like to see a recent citation. Yes, it's my intuition that MacOS is cheaper to support than Windows, but most of the numbers people usually quote are traceable back to a study done in 1996, comparing, what, MacOS 7.5 to Windows 95.
BTW, my sources in IT departments say that upgrading boxes from W9x to W2K has cut trouble ticket counts dramatically.
(I'm nominally a Linux weenie, but I gave in and ordered a used iMac because I miss the NeXT. Anybody else in the same boat?)
Sorry, I got ranty on this. To summarize, the parent article makes some good points, the previous replies are too extreme in their criticism. I supply better anecdotal evidence than them. :-)
For CFD the altivec units will be heavily used, it is correct to use them for comparing speed to PIIIs. There may be a case where it matches 50 dual PIIs, but that sounds extreme.
I just bought three lowend Gateway 2U rackmount boxes. They come in at ~$2200 for a single 800MHz PIII which will be slower than the dual 533MHz mac at ~$2500. That certainly meets `almost comparable'.
Neither of these machines are very cheap computers. There definately is a quality problem when you get into very cheap computers. There are applications for them, but if your diagnostic time is worth much, then I believe it pays to buy better hardware. (Plus you don't bleed when you have to open them up and add memory or drives. Once you get above the nasty low end machines they take time to deburr the stampings. My brother runs a metal working factory and their worst injury this year has been an IT guy that stumbled while holding a cheap PC and sliced several fingers to the bone.)
The new Macs do come with gigabit ethernet (although I don't think those old 450s had gigabit) and an OS. That gateway price is a bare machine. I put Linux on `for free', but the 2.4 kernel series (which I needed for iptables) had bugs in the interrupt routing for the chipset and it took me weeks of effort to get it all worked out into a stable configuration (manifested as a AIC-7xxx problem, took a while to find the interrupt controller problem). I knew I should have bought a 4th scratch box.
Gigabit ethernet hubs are still $250/port, but that will come down quickly. I remember when I bought a Powermac and thought it was just plain silly to put a 10/100 adapter in it, only servers could afford to have 100mbit ports. In 12 months all my new hubs where 100mbit.
I've got loads of Apples and loads of PCs. The apple hardware failure rate is less than half our PC hardware failure rate. Depending on the cost of a failure (in terms of ruined work, lost work, diagnosis, and repair) I do lean toward Macs because of their better reliability. (And before we get into a flame war on your reliable PC... I'm sure there are reliable models of PCs, but I need a vendor I can count on to make every model I might buy reliable. I can't look at historical data and say "look, here is a model that was made two years ago that was a good one" because I can't buy that anymore. (And yes, I own a PB5300, I know the counter argument, but Apple fixes it for free whenever it breaks so thats not so bad.))
--
Although Mac OS9 doesn't natively support dual processors, application support for duals can be added via a system extension.
Several Mac OS9 apps are specifically coded to use the second processor... Adobe AfterEffects, Adobe Photoshop, Cinema 4D, and other graphically intensive apps. Users of those applications appreciate all the cycles they can get.
Since the second processor is dedicated to the single application using it at the time, you usually see pretty high effeciencies for a kludge-- 190% speed increases in those specific apps.
Apple has flirted with duals since 1996's 9600 MP, but it wasn't until OSX that they really made sense.
Marc Siry || interactive media professional, motorcycle enthusiast ||
Noone would ever defend that bastard mouse.
Adam
A massively parallel distributed "BUSHEL" of Apples.
In its day a dual processor 450MHz G4 went for $2500 or less. Even maxing out the RAM to 1.5GB and adding SCSI drives (which they didn't do) you would still be well under $4000 not the $6K-$10 you quoted. That's just FUD. For $2500 now you can get dual 533MHz with 133MHz bus and $1000 with get you 3 good quality 512MB PC133 DIMMS.
Don't lead me into temptation... I can find it myself.
I'm sorry but you're not a very good mac admin then. And neither was the admin at the job I am at now. But all our machines now are running spot on with almost no crashes for the last year. All it takes is a little understanding of how MacOS works and making sure that new applications you install are compatible with other apps and the OS. At that point you either need to install the necessary patches or update the offending apps. Also don't let people install just anything on their Macs. Use Multiple users to create restricted user accounts to prevent that. And why in good God's earth are you always reseting the PRAM? Excercising your fingers? I find it useful every once in a while to rebuild the desktop, but very rarely. Restoring the PRAM is a last ditch effort when you are having problems with Monitors and conflicting resolutions or Appletalk troubleshooting. Other than that you are wasting your time.
Don't lead me into temptation... I can find it myself.
See Marathon Computer. They have a pretty wide range of rack-a-Mac equipment.
-Mark
Classic Mac OS can use multiple processors, it's just that the applications have to be specifically written to use them, which the "clustering" software here apparently is.
I guarnette any admin that is worth his weight can keep 16 X86/Sparc/Alpha boxen running as stable as 16 Mac boxen...
Sure, but if it only takes the admin half the time to manages the Macs, that means he has more time to do other work, or that you can double the number of Macs without hiring another admin.
How to solve most of our problems: 1.Lots of nuclear plants. 2.Cure aging.
I doubt that. But even if your premise is correct you'd have to add the time for rewriting all their software for Windows or Linux. Let's not even start thinking about if some of that is commercial software.
...since you can't buy these machines in any other casings that tower or cube. I doubt that it pays off for 64 machines.
How much extra does the case cost? It's just a few pieces of cheap plastic. And the extra features (handles & door) makes it a lot easier to work with than the standard beige box.
I see a lot of posts complaining about the cases, of all things. I don't think this crowd is really upset about the cost. I think it's that they just can't stand functional things being beautiful. And that's very sad.
When I competed in a sceince fair back in March, amoung other awards I won the "Princeton Plasma Physics Award", an award sponsored by the Princeton Plasma Physics Laboratory (PPPL) here in New Jersey, and as a winner I was given a free tour of the facilities. For those of you who don't know, PPPL is "The hottest place in the Universe", lying at the forefront of Nuclear Power and Plasma research...very cool.
.... MACS? The engineer giving me the tour explained that it was in the personal interest of most of the researchers. Yes, there were Sun's and other UNIX boxes scattered on the control room floor, but I would look closely, and sure enough, amounst the three or so monitors at each workstation, one of them was hooked up to a mac. There were g4's and g3's scattered all across the floor. Wack.
So I was going around the facilites, visiting their $125,000,000 tokamac's and torsotrons and all this crazy equipment (very cool science plamsa physics is, too much to elaborate on here), and I get to the control room, from which they run all their Data Acquisition (DaQ) and such to monitor the expirements, and the room is filled with
So yea, Mac's are playing a key role in plasma research, helping achieve effecient fusion, one step at a time.
For another cool plamsa physics project (unrelated to mac's), check out Garrett Young's ISEF project Quasi-Elliptical Torsatron - A Study of Induced Radial Electric Fields and Plasma Turbulence. He is a senior in high school and on the cutting edge of plasma physics research. Quite the talented individual.
my company has several g4 systems an order of magnitude more powerful (320 CPUs) both in house for testing and in customers hands. and they only eat up 9u in a 19" rack.
and as far as ppc architecture clusters go, just about any IBM RS-6000-SP2 system is more powerful, and probably takes up less space too.
The difference between Theory and Practice is greater in Practice than in Theory.
I'm very proud of the way a G4 looks, and show it to people quite often, despite it's being shoved under my desk, thankyouverymuch. It's called (say it with me now) A-E-S-T-H-E-T-I-C ... !
---
Powerbook G4/1.5GHz 12", Toshiba Satellite 1135-S1554
WTF? Are you implying that OS X takes a long time to boot up? I have a G4/350, and both OS 9.1 and OS X boot up rather quickly. It takes my windoze machine at work MUCH longer to boot up. You want to talk about bloated OSes and software, look to Redmond, WA.
---
Powerbook G4/1.5GHz 12", Toshiba Satellite 1135-S1554
Pics from the new Apple Retail Store!!
---
Powerbook G4/1.5GHz 12", Toshiba Satellite 1135-S1554
I guarnette any admin that is worth his weight can keep 16 X86/Sparc/Alpha boxen running as stable as 16 Mac boxen
I work for an all Mac company with over 16 Macs and we don't have any admins. Everything from imacs to ibooks to G3's and G4's and even a few 8500's and not one tech guy, Hmmm...
'Yes, but a 6 year old with no computing experience, with OS X, can compute circles around a 6 year old using your OS' but if said 6 yearold were useing bsd they would be learning to use a computer, while if they were useing your osX they would be learning to point and click. I mean where would i be if I wasnt messin around with dos when I was 6.
lynx
"Wooo hoo!! Look at how fast our Mac cluster is!! Now we can run... Oh wait... what the hell are we going to do with this thing?" /. users hate Windows or think Microsoft is out to get them!
____________________
Remember, not all
Prevent linux based DDOS's!
http://linux.denialofservice.org/
You mean now that some new piece of whizzy kit gets released we've got to speculate what a Beowulf "distributed parallel computing array" of them would run like? Maybe if someone can come up with a cool-sounding name then folks will stop calling them clusters : )
Bingo! Photoshop supports multiple processors under OS9 but it must be coded at the application level.
Insanity is the last line of defence for the master diplomat. But you have to lay the groundwork early.
I suppose one that developed AI would be an Apple Tree of the Knowledge of Good and Evil?
Insanity is the last line of defence for the master diplomat. But you have to lay the groundwork early.
6 CPU Cray T3E= $630,000
I can't even find a quote on a 32 proc T3E much less a 100(128). Nothing realistic price-wise here.
Insanity is the last line of defence for the master diplomat. But you have to lay the groundwork early.
Gosh! I've been waiting since NeXTStep 1.0 for them to remove all that debugging code.
blessings,
"Only in their dreams can men truly be free 'twas always thus, and always thus will be."
--Tom Schulman
I'm drunk and you're all a bunch of fucking assholes. Why don't you all be who you are, instead of trying to say something "intelligent" or "insightful". Everything that gets the +5 or +4 scores can be imagined by the average slashdot user, but simply is not stated because it's so obvious. Why are all you fucking assholes so condescending? The worst is all you l33t k5 people who think you're smarter than slashdot. Well, listen to this: The trolls here are the smartest around. They realize that you're all a bunch of fucking wannabe's who want your droll opinion (not really your own) heard so you can get fucking karma points. Why do you care about karma? I'll tell you why. Because you don't have the balls enough to believe that your real opinion is worth anything. You say what is expected and what you think people want to hear so they will mod you up. Mother Fuckers! Be yourselves, forget about fucking karma. I learn a lot more from browsing at -1 than at 2 or higher. You're all Karma whores. The trolls only realize that, while boring and reduntant (even if not said before), it will get modded up because you people are a bunch of cattle, following the herd. Grow up, speak your mind, and forget about what the others think and what they want to hear. I was going to post this anonymously, but I guess I'll grow some balls and post it logged in. What do the rest of you think? Goodnight,
Mike.
-
They set up the "cluster" in one hour. That lowers the TCO substantially.
That sounds like a bit of an exaggeration. I just turned on my G4 400 w/ OSX last night and it's almost ready for me to use it already.
No thanks. I don't smoke anymore.
I use a G4 400 w/ 192MB Ram at work. It's hard to make accurate/fair comparisons, but the closest thing I've got is SETI. Prior to the 3.x series of SETI clients, my G4 was approximately 50% faster than my PII 400 w/ 384MB Ram at home. It took ~6 hours to do a unit while my P2 average around 9. I'm pretty sure that with the 3.x clients it remained with that comfortable lead. I'm now running OSX at work, and with it's overhead, my PII running Linux feels snappier. I think the SETI advantage is still there for the Mac, but I'll be glad when every bit of debug code is removed from OSX to help remove its sluggishness. I haven't used any apps to keep track of SETI wu's now on the mac side, but I've got a buddy that runs a dual G4 533 that processes two units every ~12 hours (Approx. 6 hours per unit.) I have a Dell P3 1000 w/ 256MB Ram running Linux that averages 7h 20m per unit at the moment.
No thanks. I don't smoke anymore.
Have a look here. It's not running AppleSeed, but it is the largest collection of racked Macs.
Glad to see someone can run Linux on a K6-2. I cannot. Don't know why, don't care, whole box will be trash soon.
However, back when I used it, with a 333MHz K6-2, to compile a pretty big project (the game Total Annihilation) with VC++ under Win98, and the same code plus porting shims with Metrowerks CodeWarrior under MacOS 8.x on a 300MHz G3, the Mac was some 40% faster.
I wasn't expecting it to be such a drastic difference. Yeah, it could be that CW is just a much better compiler than VC++ (in each and every way), but I was expecting the 100MHz FSB and UDMA HD to more than make up the difference (my 7600 has 40-50MHz FSB and SCSI HDs).
I'd expect this G4 cluster to totally whoop the asses of our K6-2 boxen! :-)
-toddhisattva
Proposed terms of "cluster of apple computers"
AppleSauce
Bushle
Basket of apples
Apple Pie
Apple Juice
Rainbow color array
box of apple jacks
transparent wall
apple tree
Cowboy Neal
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
Are you referecing the fact that MacOS 9 doesn't have Dual proc support... do they realize the second cpu (and the cash used to get that extra cpu) is just sitting doing nothing but burning up (the cash) and causing heat (the proc)?
It's like running a dual proc P3 and using Windows 95/98. Atleast go with something that has dual proc support like MacOS X, Windows NT or Linux... if your going to play with 2 procs.
No, this isn't a flame/troll... just wondering what there logic is behind this... maybe you can code an application by hand under a OS that using more than one proc even if the host OS doesn't support it?
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
Good point and all... but how exactly do you save money by buy a SECOND CPU that the OS (Mac OS 9) doesn't even see... won't single proc mac's or an OS that could use them be cheaper?
Last time I looked apple didn't have a "buy one get one free" deal on their CPUs
Also we are talking Professinal Admins here, any decent admin should be able to handle a 16 boxen, no matter want the hardware/software is as long as that is the admin's "grove"...
I guarnette any admin that is worth his weight can keep 16 X86/Sparc/Alpha boxen running as stable as 16 Mac boxen...
"`Ford, you're turning into a penguin. Stop it.'" -THHGTTG
a) When G3s came out, they were marketed as mini-supercomputer.
b) When Mac OS X came out, they said 'your old G4 isn't fast enough to run OS X'
Meanwhile, my 486/33 sub-notebook (which cost me exactly $20) is running smoothly on emBSD (32meg tiny OpenBSD derivitive) on a 100 Meg hard drive, and 12 Megs of RAM. Surfing the web over my cablemodem with Links, and laughing at all you Mac fanatics. And don't even get me started on my Psion 5mx (www.psionusa.com)
---=-=-=-=-=-=---
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
---=-=-=-=-=-=---
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Researchers at the Grupo de Lasers e Plasmas (GoLP) achieved the first milestone of the GoLP simulation program on Extreme Plasma Physics: the installation of the first Macintosh G4 cluster in Europe, called epp (or ep2), which is based on the AppleSeed paradigm developed at UCLA by Viktor Decyk et al. The epp cluster is capable of delivering over 50 GFlops of peak power, and it is based on 16 Dual PowerPC G4/450, 32 processors, 12 Gigabytes of RAM, 0.5 Terabyte of hard disk space, running Mac OS 9, over 100 Mb/s Fast Ethernet, switched by one Asanté Intracore 8000. This is the fastest Macintosh-based cluster in the World. The installation and set up of this cluster took less than 1 day (including moving the machines to the computer room, unpacking the machines, and making all the cables!), and it did not require previous knowledge of networking: a one-page recipe for Mac OS clusters can be found here (AppleSeed website) (Portuguese translation coming soon). This "supercomputer for the rest of us" will be used for the numerical simulation of plasmas, novel plasma particle acceleration schemes using ultra intense lasers, and relativistic shocks in astrophysics. This work supported by the Fundação para a Ciência e Tecnologia. More info available soon (also in Portuguese), as well as science using epp. For more information also contact Luís O.Silva(+351 21 8419 336).
---=-=-=-=-=-=---
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
if that optical mouse works on the glass table :p
Thats pretty neat. Would a Linux cluster on no name x386 harware be cheaper (in terms of hardware). Would it take longer to set up for users of a similar ilk. Would this time cancel out any benefit in the cheaper hardware?
Special Relativity: The person in the other queue thinks yours is moving faster.
Yes, there are many reasons as to why x86 processors do almost everything faster.
Just to name two:
1. Memory Bandwidth.
2. Bus Design
Plus, the athlon really is kind of a math monster
http://www.wpi.edu/~cmorgan/cpu_compare.html
Check this out. There also are some good ars technica articles on the matter. But they do not provide benchmarks.
Basically, Athlon 600MHZ and above is faster than G4 450MHZ. My T-bird 1Gig is faster than G4 whatever, I am fairly confident ;)
There goes my karma ;(
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Check out some benchmarks before you display your ignorance for all to see.
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
So what he says is correct, me thinks
Don't just mark this as flamebait, why not argue the (somewhat) salient point.
Get a clue before you hide his words
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Just because someone says something less than positive does not make them flamebait (ahhh...forgive my poor English).
Get real, folks. In the real world, Box A is sometimes superior to Box B.
Dreamcast is superior to Super Nintendo.
The amount of personal ego-based pride on Slashdot astounds me sometimes.
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Ewwww.....It gives me the creeps.
WhiteWolf666 an exBush supporter. All you new-school,compassionate,save the children Republicans can rot in hell
Actually, for this particular application (plasma physics) Ethernet is fine and well. The code is not so communication intensive at all. Appleseed could benefit from a better TCP/IP stack though.
I actually had a chance to run Aplleseed on an occasion. The Fortran application written for Cray was ported using Absoft F90 Fortran. Unoptimized, its performance was 1/3 that of an 21164 Alpha per processor per node. Those were G3 350 machines at the time. It was fun.
You don't need a lot of bandwidth for plasma simulations. Just a lot of FPU horsepower.
I'd love to see a Beowulf of... oh, wait. Never mind.
/Brian
No one ever did, and Mac users are a loyal bunch. The hockey puck was one of the worst mouse designs Apple ever cooked up, just as the current optical mouse is one of the best (I still won't use one, though -- I need two buttons).
/Brian
Geez, give somebody a bridge to live under...
First off, there was a good reason to hype the optical mouse: everyone was laughing about the hockey puck and Apple wanted to show the world they'd not only fixed the problem, but they'd found a more elegant way to do it. And frankly, I hate the Intellimouse anyway -- a mouse does not need to be that big. (And fwiw I think Sun and/or Mouse Systems was doing optical a long time ago -- they just needed special mouse pads at the time.)
Built-in Ethernet first showed up on the Blackbird (5xx) series in 1993 or so. It disappeared from the 5300s and was brought back in the 1400s and has been standard equipment in every Powerbook since. I don't know if they invented it, but they were doing it a long time ago.
The rest of your statement is too silly for a response.
/Brian
And 'they' can even do that on OS X, of course...
Actually, I don't believe Sherlock does regexps (though it's not half bad as a search engine), but a lot of editing programs on the Mac (especially BBEdit) do. It's not quite the same, but hey...
And as for burning CDs... You might want to look into, say, Toast (3d party, but...) or iTunes 1.1, because I have a Yamaha CRW8824 that might take issue with that...
/Brian
programs in mac os9 can be written to take advantage of multiple processors. The 2nd processor isnt just sitting there. most commercial applications just were not written this way because dual processors were rare at the time.
Thanks for playing though!
Time for some tasty Shiner Bock!
On the Appleseed page, there are graphs in which they usually include the Cray T3E 900...including "single processor performance." I'm sorry, but...huh? When have you ever heard of a T3E with just ONE processor? Or 4 or 8 for that matter?
Okay, I get that they're trying to make it look better by comparing it with something completely out of it's league, to turn people who would just be using a bit of time on some big iron to having their own personal power playground...but lets be more fair, and see some stats comparing 100 Mac clusters to 100 CPU Crays. More realistic.
Poster A:Well, due to the fact that the PPC7400 has an Alt ivec unit it is much faster than any comparable Intel processor. To get the same speed from Intel hardware you need something like 50 dual Pentium /// computers.
.1% of all computer users (because they like to make their art "pretty")? Everything the guy said is wrong. And obviously wrong.
PosterB:Who moderates these comments up when they are completely false and incorrect babble only believed by
I would like to point out the following FACTS:
For certain well written applications where there are good developers the AltiVEC unit of the PPC 7400 (G4) shit canes all over the PIII. Allow me to post some results from the distributed.net live stats database: http://n0cgi.distributed.net/speed/
Disclaimer: I have selected Only a FEW of the RELEVANT results on the site, please feel free to verrify what i have posted.
Power PC 7400 G4- 450- 7,388,825 keys/sec.
Intel Pentium III- 1000- 5,686,498 keys/sec
AMD K7 Athlon Thunderbird- 1600- 5,662,963 keys/sec [NOTE: this is a single proc machine!]
This tells me that at RC5-64 cracking the G4 is faster than the PIII (not 50x though, approx 50%) The other thing is that the G4 draws what ~7 Watts? the PIII ~35-55 Watts? (i honestly dont know, but i have a feeling that this is ball park). Dont forget that power costs money, and in South Australia, it looks like depending on who you are you could be facing 300% price hikes on it!
How every version of MICROS~1 Windows(TM) comes to exist.
Do the following really mean anything? SCSA MCP CCSA CCNA
--I'm not actually after an answer!
I've read an article on the Register containing some speculation regarding Apple rackmounts and OS X Server 2.0. The source of the Reg's article came in turn from a similar article at ThinkSecret.
should be roughly 8 mkeys/sec per machine (a little less, usually). multiply by 16 and you have your answer. 128megakeys/sec. not too shabby. my single athlon 1Ghz only gets about 4.
Be ot or bot ne ot, taht is the nestquoi.
I usually consider a G4 to be 2x as fast as a PIII or Duron, clock for clock. Details in what you're doing and who coded your app make a huge difference, however. So do things like having enough RAM to never, ever hit the HD.
The G3 is essentially just as fast as the G4 for nonAlitVec ops - although it was actually MORE expensive when you could buy both at comparable speeds.
Tbirds and PIVs are faster than PIIIs, duh. How much? anywhere between nothing and 50%, probably.
AFAIK, the G3/G4 already has a 64bit pipe - and imo that accounts for most of the big differences. Most of the Apples are now PC standard. The best thing Apple did was let you buy PC100 RAM for cheap, and have it work.
Looking for freelance Actionscript (Flash/Flex) or ColdFusion work and/or freelance developers. Email me, put Slashdot
So what you are saying is that for a cluster, a better TCO equation would compare the amount of performance per dollar.
Considering that this type of installation requires accuracy and resiliency, a home built type system does not fulfill their needs. Then you get into pricing of machines made by the pros (Compaq, HP, etc.) and Apple does not look so expensive anymore, especially when you look at the performance of a Mac vs. a PC.
>Wouldn't it be cheaper to do this with a bunch of
>PC's in stead? Not as pretty but it's someones
>tax dollars that pay for these...
This myth has gone on enough. Every TCO study around shows that Macs are cheaper than PCs. Most of the TCO of a computer is the support time and training, NOT the purchase price of the hardware/software.
Macs require less support staff, for more computers. If I recall correctly, Intel has one IT person for every 30 computers. Except in their graphics dept, where they have 1 support person for all 300 Macs.
Because the Mac's interface is more standard across most applications, when you train on one app, that training can be utilized in other apps. This means that you get better training than if you have to train for every app.
It was recently figured that the difference averages between $400-1000 per machine, per year.
If you want to save your company $, you'd switch to Macs, but noone will do that, because it would cost (at least) half of your IT support their jobs.
Well, i'm not a native english speaker, thus i have to look up the word cluster in the dictionary, to get its real meaning. And this tells me, a cluster is nothing more then an amassment of something, and in this case of apples :)
..." use foreign words always in the wrong sense ... - 'cos everybody's doing so.
OTOH there is some saying here at my place: "Fremdwörter immer falsch verwenden
:-)
How can they say that this is the worlds fastest Mac cluster when they haven't benchmarked my cluster of 2,000 derelict Mac 68020s strung together with Appletalk and Duct Tape??
"Glory is fleeting, but obscurity is forever." - Napoleon Bonaparte
Well the programs they're running are probably AltiVec and MP optimized.
---------------------------
Pooty tweet
I guess some moron had to bring up the one button thing. You can use multi-button mice if you so choose under OS9. You can also burn CD's under OS9 very easily.
Do your research before opening yer trap
And they say Mac users are zealots and bigots! Just look at all the Linux zealots and bigots on this board! Sheesh
---------------------------
Pooty tweet
Some sad, old PC "Fragger" will be trying to justify one of these to play the latest 3D Cardboard effects game from ID. Quake XIII:
"Wow! Now you can actually see the pimples on the characters in REAL, ULTRA, MEGA, GENUINE 3D, as supposed to fake ass crap that GeForce II produces now."
"You cannot play a decent game nowadays which does not run at 2000 FPS, with full ray-tracing."
Rod.
Keeping
Why the hell they're using Fast Ethernet is beyond me. If you're going to shell out the money for 16 dual G4s, why wouldn't you spend the extra money and install gigabit networking? Or does 100Mbps suit the purpose fine.
- Ando
You are the weakest link, goodbye.
What about calling it an orchard?
I donate all spillover Karma to the charity of my choice... Ada was still a babe despite what people may say...
Wouldn't it be cheaper to do this with a bunch of PC's in stead? Not as pretty but it's someones tax dollars that pay for these...
Well, as the other replies said:
It took them considerably less than a day.
That is actually an achievement. As far as I can tell, the costs of personnel dwarf the costs of hardware, where an engineer costs upwards of $800 a day, and engineers sitting idle costs as much as engineers busy doing work, so the cost deltas of the Macs is offset by the speed and effectiveness of setting up and maintaining the array.
In the long run, it's not the fixed, sunk costs that a PC represents that is meaningful, it's the daily ongoing performance, productivity, and maintanence costs that mean anything, especially in a place that runs plasma tokamaks at $125k a day...
Geek dating!
GPL Deconstructed
What about using Sun Blades? They can be clustered, can't they? You can pick them up starting at a grand.
Need a Linux consultant in New Orleans?
I wonder if someone should do some work with the MPI API to add those features? A layer on top of MPI that does the things we do through programming would be really nice. :)
Oh BTW, I was talking about a cluster intended for parallel computing... in reference to the article, which is also about a parallel computing (Appleseed) machine/cluster/array. I was trying to demonstrate that the points the poster made about this NOT being a "cluster" in the sense of the word can be solved in the parallel world with APIs... because a Beowulf's programming IS its behavior, for the most part. And through programming you can make it behave in the desired ways when it's running your program! (And wouldn't an API on top of an API be great for that!)
- Distributed Lock Manager: locks can be simulated with certain types of MPI messages.
- Cluster-wide File System: use MPI to pass data back and forth between nodes, including instructions on where to write the data. Not only is there a cluster-wide file system, it's a customizeable cluster-wide file system!
- Process control: Scyld's bproc allows all processes to appear as if they are running on the master node. You can also move processes between nodes transparently with this.
- Connection manager: Scyld provides this to some degree. You can do remote shutdown/startup of nodes or groups of nodes.
- Shared System Disk: well, nodes bootstrap from thier own drives, but download a new kernal image from the master node on bootup. They also pull down libraries from the masternode on bootup.
- Single security and management domain: permissions are the same on the slave nodes as they are on the master. But the slave nodes are truely compute nodes, and permissions there matter little, except for data files.
- Cluster wide process control: you said it. Beowulfs do it.
- Mixed Architecture: what you can do depends on your cluster. For a mixed Mac/Alpha/x86 cluster, you'd have to have different executables for each node. I'm sure there's Beowulf software that lets you do this, but for ours we don't need it (all Athlons.)
- Rolling Upgrade Support: yup. Acually, with Scyld, if you reboot a node, it will come back up with the newest configuration, imaged off the master node.
- Parallel IO support: simulated/managed through MPI pretty easily. Set up "IO" nodes and let them handle it.
- Interconnect failover: networking on a Beowulf is up to you. We use a high-performance switch and some channel bonding. It can be done.
- High-end scaling: Beowulf? "OK?" Have you ever heard of ASCI-Red?
- Load Balancing: ours does round robin scheduling of jobs, but it usually doesn't utilize the higher number nodes unless you run a job that requests a large number of nodes. We wish for better control, but this works pretty well.
- Cluster Alias: yeah, what you said.
:)
Anyway, the URL for our machine is http://www.cs.uidaho.edu/thecollective. It's up to 64 nodes right now, even though I don't think the site mentions that."While I think this is a sweet application of technology, it continuously galls me to see the misuse of the term cluster when applied to distributed parallel computing arrays such as AppleSeed or Beowulf. These are not clusters! There is no distributed lock manager present, no hardware-level device sharing (as opposed to software-level file sharing a la AFT, SMB, or NFS), no means of transparently starting, controlling, and stopping any process (not just those written to a custom API) on any node from any other node. While an excellant and usefull technology, this has a long way to go before it could be considered a true cluster..."
Blue skies... Barthie burgers... girls.
If you were seriously deciding whether you're going to purchase dozens of PPC or X68 boxes for some networked processing, you'd think 250 pages of intro material might be worth reading.
Besides, the AppleSeed one page manual casually mentions that you'll naturally be running custom C and Fortran code on your machines. So, the --look ma --no hands-- effect kinda lose some of its punch.
If you're going to write custom apps for it, then you may eventually even want more documentation for when things don't turn out like you hope.
I think Mac has inspired some good ideas in software, but I imagine many small time users of parallel processing would probably be into 3D rendering and they'd probably be a price conscious lot. I guess some people were implying encryption key ehem. . . testing might be another application. But again, we're talking about a price sensitive crowd who would probably just go ahead and spend some time with a book before obtaining those PCs.
After all, what about if you can get yer finger on a source some of those ultra-dense server cases --the rack they used in the article blew-- that fit with some new dual AMD boards. Ooh ahh.
Besides, let me get lateral to the topic for a second. Have you heard that Via of Taiwan is doing CPUs already and they're shooting below radar in the price point penny arcade. they claim they're gonna take over the Mainland China market by trailing AMD and Intel in performance, but doing that price limbo dance like we haven't seen yet. I heard they were talking 1 GhZ for less than fifty bucks by the end of the year and Via claims in the local press that this is the price point they want to be at as they continue to plod along following the big guys and eating at the bottom edge.
Now, if these things have good chipsets and boards, then maybe it would be time for joe six-pack nerd to start stacking them up with a cool rack-mount configuration in every home. Perhaps we'll see generic rack mount systems with modular PC units and matching stackable stereo power amps. They should share some kind of heat soultion. Hell, if your amps and CPUs were hot enough --as they ought to be-- you could scavenge the process heat to run a heat pump to assist in cooling and heating your house. Wow, rad! I've got to talk to some of these Taiwan case moguls and see if we can get something going here.
It could be the fun new DIY consumer category --the power stack. It makes a house a home.
The combined 14.4ghz of processing power is cool and all, but what I want to know is how they managed to get optical mouse to work on that plexiglass shelf. Now that's amazing.
Sorry, dude. Whilst your comment may be true ( I don't know if a G4 comes off the same line as a 7400 ) the instructions and performance are the same for the G4 and the 7400 at similar clock rates. The guy knows what he's talking about. http://e-www.motorola.com/news_center/press_releas es/PR990831A.html
This is gonna sound anti-geek, but this just sounds like technology for the sake of technology. I think it's cool and fun that a bushel'o'Macs have been clustered to run real fast, but I have to wonder 'why'?
Is there some benefit in using this solution vs. the umpteen other (cheaper, maybe better?) clustering solutions that exist today? Just sounds like a misuse of time/effort/money to me.
Rebuttals anyone?
CrazyLegs
"Pork!!" said the Fish, and we all laughed.
If you used P3/ P4/ Athlon machines, then you'd have a much cheaper cluster.
Even Slashdot wants to hide some things
I spent a year in Iraq looking for WMD and all I found was this lousy sig.
1) I didn't explore much the site and related links, but I wonder if there is MPI available on this cluster. Without it, they cut themselves of a large community, and cannot even give the crudest evaluation of supercomputing performance, this is, Scalapack Gflops. 2) concerning human costs, I think that in a research institution with a lot of permanent staff, human costs are indeed zero. It is a better option to spend the tiny budgets available for research in Europe on cheap hardware, and spend some of the almost free and unlimited time of the permanent staff on hacking some Linux/Beowulf stuff. And don't tell me this time would be better used doing fundamental research ! The benefits of understanding the inner working of a system are huge, as well for education as for research.
Google passes Turing test : see my journal
Wroot
Sweeeeeeeet.... I was published on slashdot... /me runs to tell his friends
I just would like to know, how they can use multiple CPUs if the OS (os9) doesn't support them?
===> An eye for an eye makes everyone blind - MG
Can I rack mount my SE/30s? (they run NetBSD)
Amoeba ??
I believe apple plans to release Mac OS X Server 2.0 soon.
Anyone hoping some PowerPC Rackmounts will be released at the same time?
I have exactly the opposite experience with SETI. My Athlon 750 box running Win2k blows my brother's G4/400 (with MacOS 9.01) out of the water. I can process a wu in about 8.5 hrs with numerous other processes running (playing DVDs, AIM, etc), but he's taking 11.5 hrs running SETI only when the rest of the system is idle. This sort of performance is sad given how much faster the G4 is in general.