Slashdot Mirror
Xgrid Clustering Software and Demo
no_demons writes "Along with a selection of other goodies, Apple also unveiled their Xgrid clustering technology from their advanced computation group today. Xgrid can turn a number of networked Macs into a supercomputer, detects nodes automagically via Rendezvous, and can run in or out of a screensaver mode. You can download a technology demo (including a BLAST test app) here."
Does this work on the small scale as well, like OpenMOSIX? We have a few G4's at school that could benifit from clustering.
You can't judge a book by the way it wears its hair.
I would like a cluster of those...oh wait Rus
CPanel + Root from $35/mo - 10% off with discount code SLASHDOT
It'll be interesting to see if I can make this work with the stacks of old LC520s in my garage. I've been wanting to cluster them for a while. If Xgrid will work on those, Mac just saved me a ton of work. Not that I wasn't going to have fun with it....
Damon,
http://actionPlant.com
What's the cost? I can't find any info on pricing, etc. Sounds like a great product if it works as advertised.
But for me, the model I want is a broker model. I want to sell my processor time to a broker who will resell it on a day to day basis to whoever is the highest bidder. E-bay of grid computing, ya know. I don't want to pick projects, download clients, etc. I just want to pariticipate (i.e. make money) from whoever is willing to pay the most at any given moment.
And when I feel like it, I'll volunteer x% to non-commericial stuff like SETI@home.
"If you want to improve, be content to be thought foolish and stupid." - Epictetus
Basic Local Alignment Search Tool - a pattern matching program for DNA and protein sequences.
This makes me think: "Hmm... Seti@Home with an Aqua theme..."
Xgrid clustering is pretty neat. It doesn;t provide much int he way of an environment for parallel computing ala beowulf as of yet, but it is a great way of easily distributing batch jobs to a number of machines, and has a fancy pants tachometer to boot.
Why limit this to Macs only? Wouldn't this work even better if it were cross-platform -- like many other distributed computing solutions such as SETI, distributed.net, and the UD Cancer projects.
Now every single story posted on slashdot for the next 6 months is going to have a comment of "Imagine an XGrid of these!" The old incarnation was slowly dying too...
What are we going to do tonight Brain?
1 = super computer
1100 = supercomputer
See, with that many more computers you DO run out of space
*rim shot*
Thanks folks, we're here all week. Try the veal; it's delicious. And please, don't forget to tip your waitress.
nt
when the G4 was out the concept of a supercomputer was still the old how many gigaflops it ran, at that point the G4 was considered a supercomputer because it had just broken the old gigaflop barrier.
but with that Apple while advertising it was a supercomputer secretly wanted that concept to change cause it ment they couldnt sell the computer to a LOT of different countries due to export laws and supercomputers (the same thing happpened with Japan and the PS2 I think, or it might have been a urban legend in that case) Intel and AMD where also in on it since their chips ALSO where breaking the gigaflop barrier and where technically classified as "supercomputers" based on old 70's classifications.
Since then the rulling was changed and it will be a long time before any of the makers will reach the current mark with a one or two prosessor machine.
Now in Virginias case, it really IS a supercomputer, actually the third fastest at this point and THE fastest educational supercomputer
before I think the G4 was like 6 or 7 gigaherz slower than the closest "true" supercomputer
"Slashdot, where telling the truth is overrated but lying is insightful."
It's been a while since I played with Beowulf clusters, but I've been thinking of using a Beo-class cluster in an enterprise app. The problem there is that you want the IT people to have to do as little "deep magic" work as possible when it comes to grids. So, auto-detection of new nodes, transfer of jobs out of failing/crashed nodes, etc. are important features. Is there anything close to that for Linux? a few different solutions that integrate well together isn't out of the question either...
Oh good, you mean I can actually talk about XGrid now after signing an NDA over 6 months ago? :)
;)
We had the second installation of XGrid, the only other group using it at the time was NASA. I haven't had much time to play with it personally but we had our coop do some genetic sequence analysis using it and he was quite impressed. Plus the speedometer-like gauge measuring performance just looks soooo cool.
Play a little with it seem realy nice. A batch job scheduler using rendez vous. If only this was open source, would be realy helpfull
I wonder how long it will take someone at MIT or some other tech university to figure out if I put all the Macs located on or around a college campus I could or University could make it's self a mini-Super Computer for very few dollars.
why is this the Xgrid not not the Igrid?
I don't know much about clustering, but maybe someone can fill me in. Would it be possible to run something like Postgres on a cluster of XServes? Would the clustering be transparent at the application level, so that any program could take advantage of the clustered resources or would specific distributed algorithm apps need to be written?
I'm hoping that I could just get a stack of XServes and run an OSS db on it for free (as in no extra effort required), but I'm guessing that's not the way it works.
I wonder if I could offload some of my compiles to TiVo...
:-)
Share and Enjoy!
How about a linux cluster of old pcs to run a single wine. That could be a cool way to get a linux powered supercharged windows.
We all now know that you got to see XGrid in action first!
.sigs are for post^Hers.
Big Mac Super Computer Movie
If you want to see a behind the scenes vid of the making of the G5 supercluster then go to the link above.
I wonder what, if any, input VT had on XGrid...
Mosix would probably be the best bet although it is still missing migratable sockets and distributed shared memory. The distributable shared memory problem has been at least partially addressed with migshm.
Give it a try.
[RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.
Duh.
This strikes me as weird. The whole point of the ban on export of supercomputers was that they could be used to simulate nuclear explosions - but this was back in the '70s, when a gigaflop really was a supercomputer. Today, therefore, with a competent physicist and a good programmer, any old desktop PC can simulate nuclear explosions at least as well as the technology the US was keen to avoid exporting in the '70s (including, presumably, fusion devices) in a few months.
I can't see the point in retaining a ban on export of supercomputers even if the bar is raised - nuclear simulations are already possible with off-the-shelf hardware that isn't covered by restrictions and, though I suppose biological simulations may well require more CPU time, if the various distributed biocomputation projects are anything to go by they also parallelise well, making them tractable by a cluster of conventional hardware as well as by traditional supercomputers.
"'I pass the test,' she said. 'I will diminish, and go into the West, and remain Galadriel.'"
- JRR Tolkien.
DVD::Rip will utilise a cluster
[RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.
Here is another wonderful example of Apple beating Microsoft on more than one level.
.......... and now Beowolf clustering out of the BOX!!!
1. Over all Coolness AND Geek Factors.
2. Usefulness of coolness factor.
3. Ease of use of Geek factor.
Microsoft hasn't done anything remotely like this in thier existance, rarely do they push the tech envelope beyond the what has become ordinary and benign. And when they do, they end up with over analized and engineered products nobody want (Bob?).
Apple of the otherhand is OFTEN making stuff WAY before the curve. Newton/PDA, USB/Firewire, CDROMs, Floppylessness (is that a word), standard networking
If Apple is smart with this, they could play against the new model of "ON DEMAND" services from companies like IBM.
Imagine a corporation that could automatically timeshare, timeshift computing resources based on such technology using the workstations they currently own/purchase. Peak periods of processing could be syphoned from little or unused desktops "On Demand", back filling any need left over from the DP center.
If this is NOT revolutionary it will be at least evolutionary.
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
From my own inexhaustive observations, Apple gives consumer stuff "i" and geeky stuff "X", hence iPod, iApps, iMac, iLife etc against XServe, XCode, XGrid, etc.
To be frank I've always wondered about Apple's name syntaxes. When the Mac IIx and the SE/30 came out - improvements of the Mac II and the SE with then top-of-the-line 68030 processors - it seems they really should have gone with Mac II/30 or SEx. Mac II/30 sounds like a third grade joke about chinese dentists but the Macintosh SEx would have probably made them billions.
nt
The honest truth is we have a goverment thats clueless when it comes to tech. Sure it might change in the future but right now they really dont know what they are doing and do stupid things like this thinking they "saved the world" when all they really did was make a mess.
"Slashdot, where telling the truth is overrated but lying is insightful."
> Now in Virginias case, it really IS a supercomputer, actually the third fastest at this point and THE fastest educational supercomputer
And how is the ES (and most other supercomputers) not educational? There's no value for the sort of data it crunches other than research...
I'm imagining a nice glass of cider.
Read ClusterWorld and you can figure this out yourself.
HPC for Primates. Read Cluster Monkey
(the same thing happpened with Japan and the PS2 I think, or it might have been a urban legend in that case)
No, it was not; although Japan wasn't exactly the culprit.
What's more important is what it's clustering, 90 nanometer G5s. Apple and IBM are the first company to bring 90 nm processors to the market. Xserve White Paper
Its fine to talk about supercomputer "based on old 70's classification" But I think that was the last time the word had any meaning. What the heck is a "true" supercomputer? What ruling or "rulling" was changed? The word supercomputer should probably be buried next to "home computer" and "IBM clone." Supercomputers were what Cray made. Since then its just marketing jargon, especially when its from Apple.
The whole point of the ban on export of supercomputers was that they could be used to simulate nuclear explosions
Not exactly, it was also their usage in brute-force decryption. The former point was indeed emphasized in the days of the old Cray supercomputer, but the latter was stressed as the most important factor in mid 1990's, when the aging cold war COCOM was replaced by the Wassenaar Agreement.
Actually, there was an old joke around Apple when I worked there about this exact subject.
Since the SE/30 was essentially a IIx without NuBus slots, the development team wanted to follow syntax and codename the machine the SE/x. This apparently was not cool with Scully or Jean-Louis.
Of course, no machine ever starts development with a final marketing name (at least not back then), so the story is most likely bull.
Why did I get moderator points on the same day as Macworld?
i ment educationally owned, the ES is goverment owned, in this case the goverment of Japan, The V Tech one was built by the school using school money and was not contracted by our goverment. Though your right there is a educational value to just about every supercomputer made, even the ones built for balistics and nuclear testing.
"Slashdot, where telling the truth is overrated but lying is insightful."
It's hard to imagine that this is cost-competitive with Linux clustering. Macs aren't cheap.
Isn't this just a batch queueing system, much like Sun's Grid computing, Platform's LSF and the like.
If it is, it is different in nature than the often joked about beowulf clusters and the mosix systems.
I think its great that Apple has one too, the more ideas thrown into the pot, the better all of them become.
-Tim
-I just work here... how am I supposed to know?
Now that Apple's trying to get into the big league number crunching market, I wonder if we will see a 64 bit version of OS X any time soon?
Personally I'm waiting for the xPod and xPod RAID.
a million songs and four and a half terabytes of data... IN YOUR POCKET!
I thought they had an app that would let you start a compile (or something) on one machine, and if it got busy that machine would go to other Sun boxes on the network and say "Oi! Can I borrow some CPU time?" If the other boxes weren't too busy, they'd share the job.
;)
Maybe this is what Xgrid is doing now, I didn't read the article too well.
To the rest of the bunch when one of the apples goes bad? *rimshot*
Seriously though, I would have seen this as bigger news if it would be possible to afford more than one apple on my wages. Anyway, I figure this would be an appropriate thread to rant about my friend. He's always hounding me about how hot the macs are in the studio, and that they use protools on these mac servers and trying to explain to me what "better quality" audio they produce, and how all the studios are using macs now.....
Ok, original post shoulda been modded funny or troll, but there are cluster solutions for old Macs, so here goes:
NetBSD/68k, supports 68k cpus and various free cluster architectures.
Appleseed, supports OS 8.6 - 9.x on PPC.
Quite a few older PCI Powermacs can be coaxed to run OS X 10.1.x using XPostFacto and some patience. Won't support XGrid for now, but the other free suspects will work.
Luke, help me take this mask off
Breakfast served all day!
Don't forget: this blows ANYTHING available for LInux out of the water too. Basically any large scale computing done on either Linux or Windows will cost a lot more to implement than if it is done on OS X. Once again Apple demonstrates how professional programmers with vision can accomplish technological feats that can leave even the much vaunted (and over hyped) open source developers in the dust. Way to go Apple!
I'm speculating here (I haven't seen XGrid specs yet).
If (and only if) XGrid is one of the Single System Image clustering technologies like opnMosix, then things like MPI jobs are pretty nice to work with on these kinds of clusters.
In openMosix, the load balancing system simply migrates individual MPI processes to nodes that are the least loaded (and which are projected to provide the maximum instantaneous throughput for the job).
Assuming that XGrid has a similar architecture (a big assumption), having a commercial implementation working across Apple's relatively stable hardware platforms (with AltiVec doing vectorized floating point behind the scenes on each node) sounds like a pretty decent cluster implementation to me. Now, as always, Apple must work on cost.
You are correct that only certain kinds of problems "work" well on LAN clusters. But experience shows that a sufficiently motivated computational scientist can adapt many more "big grunt" problems to various weird-and-wonderful architectures than might be apparant at first glance.
Adapting codes to fly on "true" supercomputers (e.g. Big Iron vector machines) is just as much an art form as adapting them to fly on MPI.
YMMV.
what a phenomenal idea. to take advantage of the idle processing power of machines that sit unused at least 50% of their time. virginia tech is just the beginning.
any college with a several ~25-machine labs can use this app to do supercomputer stuff, AND get the return on investment from normal users being able to utilize these machines during the day.
www.pixelectric.com
A code snippet from here(!) reads:
/* disable conditional is constant warning */
#ifdef _WIN32
#pragma warning( disable : 4127 4706 )
#endif
Why would developers at Apple ifdef for Win32?!
... is to make this available to *any* program that does intensive rendering. Using Toast's iMovie extension to export a 44 minute movie as an MPG for VCD took 2 hours 12 minutes at the best quality on a dual-G4/1250. A few other Macs laying around (like the other 22 in my department) could have cut that down quite a bit. Sure, it'd take time to send out 9 GB of DV, but less time than it took to encode it.
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
Clustering databases has different issues/concerns than clustering computational problems. I wrote an article about database clustering a while back, available here, if you're interested.
Breakfast served all day!
Apple's Xcode already does this.
(From the Xcode website)
With theRendezvous-enabled distributed build feature it's easy to simply farm out your build by distributing compile workload across idle desktop machines or, better, deploy a dedicated Xserve build farm to do in minutes what would take hours on any single machine.
Launch Den Mother and Launch Den Puppy are a distributed computing application pair for older Macs. You can download a 68k version despite their claims of requiring a Power Mac.
Its not as automagical as XGrid and you will have to write your own multiprocessing code, but it is a way to do distributed processing on old hardware. Clustering a bunch of LC520s may not be the best use of time and electricity, but then who said that a hack must be cost-effective.
Two wrongs don't make a right, but three lefts do.
Anyone here seen POOCH? Been around for, what, the last year and offering the same thing? Now XGrid is big news?
Imagine a beowolf cluster of ...
Oh! Never mind.
[Insert pithy quote here]
It depends on what you're paying for:r formance?
Service?
Support?
Design?
Usability?
Pe
Functionality?
Real estate?
Cooling?
Electricity?
VIrginia Tech found Macs were cheaper than the competition, and that's with hulking huge G5 PowerMacs. Imagine the same cluster, but with G5 XServes:
1/4 the size
1/4 the real estate costs
better servicability
better management tools
greater diversity of possible nodes (XServe RAIDs, XServe Compute Node, and XServe)
Same price
So for the same reason that the G5 cluster was cheaper than a similar Linux cluster, a G5 XServe cluster will be cheaper than a similar Linux cluster; however, if you target *specific* costs, some things may be cheaper, and others will be more expensive.
For example, how much will BLAST + XGrid cost you?
GPL Deconstructed
From Apples Website: http://www.apple.com/acg/ Xgrid Xgrid clustering software is intended primarily for scientific researchers who have a set of networked computers that are not yet being used to full collective CPU potential. The software provides a remote execution environment and file staging abilities that coordinate the running of tasks on distributed computing resources and ensure that each computer has access to all of the files necessary to execute the tasks, thus freeing both user and developer of such resources from this tedious work. Xgrid does not solve all clustering problems. It does not replace clustering software such as MPI or clustering hardware such as InfiniBand. Xgrid does not accelerate or "grid-enable" existing applications on your computer. For an application to take advantage of the Xgrid technology, you must update it to use the Xgrid APIs. However, if you currently run long computations in your Terminal windows using, say, an already-compiled executable, you should be able to use Xgrid out of the box to run batch jobs for this executable on your clusters. In this way Xgrid offers high-throughput computing with minimal development work on your part.
I haven't seen a DNS problem in close to a year.
Veteran, Bermuda Triangle Expeditionary Force, 1992-1951
You might be thinking of Sun Grid Engine, an open-source project sponsored by Sun which does cluster-level load balancing and other tricks.
Cheers,
David
Could you imagine a Beowulf cluster of those!?
XPostFactoallows installation of OS X on *unsupported* Macs...not quite as old as your LC520, but it *should* allow for a certain degree of recycling...
We apologise for the fault in this post. Those responsible have been sacked. -- Signed RICHARD M. NIXON
And what fun all those newbies will have thinking they now own a 900MHz Cube :)
I think that Virginia Tech would disagree with you on that last point there. There are not many applications that you could or would want to use 64-bit computing anyway so I don't understand your point. These days it's either DBs or number crunching. For the latter, having distributed system helps it even more.
But if you're not happy with your G5, you could send it to me. I'm sure I could use it to play Pong or something.
Well, there's spam egg sausage and spam, that's not got much spam in it.
right now, each folding@home client runs on each CPU of a dual CPU mac - becuase the client app isnt set up to take advantage of threading. I can see where Xgrid would give clusters of macs some huge folding advantages if a suitable client was coded. Instead of each segment taking 10 hours, you could do one in 1 hour given enough macs and move on to the next. this opens up some some neato distributed client advantages for macs if the proper client software is written to use it. I think I'll go over to the folding forum and see if anyone else has this idea.
hahhahahaha...he made a pong reference...HAAAHAAHAHAHHAHAHHAHAHAHAHHAHAHAH
He forgot the EX at the end - as in Exabyte
From excellent karma to terible karma with a single +5 funny post...
Can I use Xgrid with other UNIX-based computers?
Motorola 68030/25MHz with a 68882 FPU.
g e= gallery&model=lc520
I had a Performa 550 which was basicly the same thing but with a 33MHz 69030.
http://www.apple-history.com/frames/body.php?pa
1.
Confirm that your hardware can run Mac OS X Version 10.3 Panther Mac OS X Version 10.3 requires a Macintosh with a PowerPC G3, G4, or G5 processor, built-in USB; at least 128MB of physical RAM and a built-in display or a display connected to an Apple-supplied video card supported by your computer. Mac OS X does not support processor upgrade cards. Verify your hardware is supported from the list below.
2.
Verify you have enough hard drive space. While the amount of disk space required depends on your computer and the way you are installing Mac OS X, you are recommended to have at least 2.0 GB of available space on your hard drive, or 3.5GB of disk space if you install developer tools.
-- ac at home
On an LC520, you can run the following:
u s/listm achine.php?id=56
t num=112 213
e t. html
- A/UX *
- NetBSD (all hardware supported) **
- Linux ***
- OpenBSD *
* Requires you to install an 68882 FPU chip on the LC520's motherboard. Last time I bought one, they were $17(!).
** Requires an 68882 FPU, or else uses FPU emulation, which may be buggy and/or slow (maybe no "may be" about it)
*** The status of Linux on the LC520 is here:
http://maclinuxstatus.sourceforge.net/stat
The specs for the LC520 are here:
http://docs.info.apple.com/article.html?ar
ALSO, you'll need a supported LC PDS ethernet card for networking, since the LC520 has no built-in ethernet or Nubus slots. OpenBSD doesn't support the PDS slot cards at all. NetBSD and Linux probably do.
Here's a list of supported cards for NetBSD from _1997_:
http://www.macbsd.com/macbsd/macbsd-docs/ethern
And lastly, you'll want a 3 button ADB mouse, or else I'd advise using the LC for any purpose other than running an X display server. The damn things only use 40W of power according to Apple, about as much as a modern powerbook, which actually sounds about right.
Democracy. Whiskey. Sexy. Pick any two.
But we *nix users did it first. As usual Apple is publishing things which they think they did first, but we in the research community have been doing for years.
This is a test. This is a test of the emergency sig system. This has been only a test.
I think charitable uses of CPU time give the most bang for the, er, buck.
Donate background CPU time to fight cancer.
or i could be wrong, heres hoping im not!
I want 2D games back.
Now, with the Mac zealots insisting that it's OS TEN - not OS ECKS - how is XGrid pronounced? TEN Grid? Or does this not bolster my theory that calling it OS TEN is retarded?
WTF? What ever happened to the word "automatically?" Originally, "automagically" meant the same thing as "automatically" except that it implied awe-inspiring magic took place to actually implement the described automation. And I'm not talking lame, I-bought-a-book-about-illusions-and-I-want-to-perf orm-at-children's-birthday-parties-because-I'm-a-c hild-molester magic. I mean serious magic. The kind of magic that would have brought Merlin, Gandalf, and Harry Potter all to their knees begging for mercy.
Now, however, the only word I ever see written is "automagically." I mean, come on, that word was clever about a year ago (maybe not even that recent) and now every spelling-challenged, self-deluded, double-posting idiot incapable of independent thought uses it. I think less of any person who continues to use "automagically" when the so-called magic implied does not seem outside of the realm of mere mortals.
[Generic comment about Slashdot editors and submitters perverting the English language, fight back, holy Jihad, morons, etc.]
True story.
Apple gives consumer stuff "i" and geeky stuff "X", hence iPod, iApps, iMac, iLife etc against XServe, XCode, XGrid, etc.
They couldn't have called it iGrid.. there is no I in team.
ZERO CONFIGURATION.
Let me say it more plainly:
ZERO CONFIGURATION.
Nobody gives a rat's ass about loose clusters or distributed computing. Seriously. It's old hat. It's not exciting.
What's exciting is taking a network of Macs that are used for other tasks, installing (remotely, possibly) a tiny program on each, DOING NOTHING ELSE, and running jobs across the whole set of machines in parallel.
ZERO CONFIGURATION.
Apple did it first.
I don't think you guys really understand how big of a deal GarageBand is (and I mean to paid musicians whom already own pro audio software). I produce downtempo electro-acoustic psychedelic (bassy breaks stuff) tracks for a local label with Reason, Ableton Live and Logic, but none are as sweet (or should I say organic) looking as GarageBand for recording and editing tracks (Reason actually looks quite good for a synth/sampler/effects rack : yet visually lacks when one is editing within the track mode). Why should I even mention to you how good looking GarageBand is? Why you say? Simple. If it excites me to work with a good looking peice of software - which I admit - it does very much (appealing to my eye - with incredible ease of use and superior workflow), it will inevitably inspire my work. And if it inspires my work, in any sort of creative way, I would gladly pay far more than the small price of $49 for it! Now granted, it does not do all the things that Live, Reason or Logic does, but by the looks of it.... I think it will do what it does better than anything else I've seen. GarageBand should be able to listen to (and record in stereo, I believe) a firewire enabled piece of hardware (like Yamaha's brand new o1x) with the knobs controlling any enabled AudioUnit plugins (in real-time of course). Will the AudioUnit capability within GarageBand allow me to use mastering plugins (not to mention 5.1 mixing)? How many AudioUnit plugins can run in real time while simply monitoring (or recording) my 12 String & vocals? Can you pre-record midi to the AudioUnit plugin effects while monitoring guitar or vocals while jamming to a pre-recorded piece (to provide dub-like capabilities)? Will GarageBand listen to more than one midi device at any given time (keyboards, mixer & envelope pedals)? Obviously I'm not sure on some of these paticular details, but I'll find out soon enough. I must remind myself this is Apple's first release of this product, and updates will inevitably follow! The fact it will seemlessly integrate with Soundtrack (when scoring within FCP 4), plays and records at 24bit 96khz (I think - haven't found the pdf on it yet - Soundtrack does - so GarageBand should), uses AudioUnits, comes with a large amount of high quality (better than CD quality) samples (some of the best I've ever heard by the way) and comes with 100 software instruments... well... it's a steal (to say the least) at $49. Comparable software goes for hundreds of dollars more (not to mention the fact you get to upgrade your iDVD/ iMovie/ iPhoto as well). Not to mention the interface of those other apps are half as clean (re: slick) as GarageBand. Just check out the detailed "wood" side panels on the main mixer window. Don't you think it adds a warm touch? I like details like that. Thank you Apple! I'll be buying it... I'll also be upgrading to the "Jampack" available for it as well. GarageBand is the start of something incredible, for professional and amateur musicians alike... End of story.
Now, if you were using XGrid to create a beowulf cluster of John Mayers demoing GarageBand, you might have something there.
When the xserve lacked ECC it was "Completely worthless due to lack of ECC". Now that it has it, the same people "Well, ECC isn't all that".
Sheesh
it is a carbon app
You mean Cocoa, I guess? Anyway, thanks for your informative post.
Anyone got any benchmarks on how fast (and how many) these puppies can burn through a SETI@Home packet? I have read that one G5 is very comparable to AthlonXP/P4 in regards to SETI@Home, but would love to see the results of a G5 cluster doing SETI.
"Apple Corporation is proud to announce new version of XGrid technology, extended for use on iPod digital media player and Apple Pro Mouse."
:)
I wonder, what computational power will cluster of iPods have. Cluster of mice, anyone?
-- grmbl woz heer
Would a cluster of eMacs be referred to as eGads! ?
I normally don't respond to ACs but you are truely a Mac moron.
This is a test. This is a test of the emergency sig system. This has been only a test.
http://eshop.macsales.com/OSXCenter/XPostFacto/
Doesn't support your model though.
Why port from Obj C? gcc compiles Obj C. Obj C is faster than C++, uses less memory, and once you get used to the smalltalk way of message passing and model, view, controller... It is, imho, a better environment than C++.
Toast v6 performance for VCD or DVD video is not that great. I have seen it take hours for conversions, and that's with a Dual 2Ghz G5....
Wow. You really don't get it. You have no idea what "zero configuration" means.
/etc/antsd.hosts, which tells the antsd which hosts are allowed to join the ANTS network (the JobNet).
Let's talk about ANTS. (http://unthought.net/antsd/sample.html)
On both nodes, you should now create the file
That doesn't sound like zero configuration to me. That sounds like configuration, in fact.
With Xgrid, you don't need to do anything like this. Xgrid automatically discovers other nodes on the local network using Rendezvous.
Now let's talk about distcc.
3. Put the names of the servers in your environment:
export DISTCC_HOSTS='localhost red green blue'
Sounds like configuration again. If you've got two machines, it's no big deal. If you've got two hundred... well, you know.
Interestingly enough, Apple modified distcc to be Rendezvous-aware when they included it in Xcode. So Xcode has ZERO CONFIGURATION distributed compilation, another Apple first.
Yeah, you heard me. Apple first. Because of those pesky little details like ZERO CONFIGURATION.
"Mac moron?" Sounds like the moron here is some asshole who goes by the name "biginsa500."
(And the snobby note--"I normally don't respond to ACs"--just made it extra-delicious.)
i realize it might be fun to try, and this is slashdot, but consider how much effort it would be to get Xgrid on a slew of old old old macs (that now live in a garage). i wonder how many of them would it take to equal one LCD iMac G4 (including the loss to cables and whatnot that must exist?)? then consider the electricity used? even an older CRT iMac. would be better i would think, let the screen shut down and the lack of fan makes it happy. yes you may have those machines (i have my mac collection too) but there is a cost to actually running them all at the same time and that kills baby seals and gets you bad karma.
now if you take a university computer cluster that's closed for 6+ hours a day, or something like that...... i could see some interesting uses of current technology being used on downtime. almost like the idea behind SETE@home or folding but for your specific needs. i'm sure this won't be like rackmounting and fibre linking the beasties in one windy room, but if the machines are already there.......... it's kind of free? i wonder if you could take, say, all the secretaries iMacs and use them in the 17:01-08:59 time period to run this, let alone weekends and holidays!
Not exactly off the shelf. More like first off the assembly line. In the keynote speech yesterday, Jobs told the audience that VT got the first 1100 dual 2ghz G5's made and apologized to those that got bumped and had to wait. A brief audience chuckle followed but Jobs sounded like he meant it.
Is this thing on? (crickets chirping) : )
Again, any beginner *nix user has no trouble setting these up. I consider these zero configuration. But anyway, what's your gripe?
This is a test. This is a test of the emergency sig system. This has been only a test.
You're making an artificial distinction between Apple and Unix which I feel is unwarranted. OS X is basically a variant of *nix. The distributed compiling in XCode *IS* distcc:
marke$ uname -a
Darwin Lugh.local 7.2.0 Darwin Kernel Version 7.2.0: Thu Dec 11 16:20:23 PST 2003; root:xnu/xnu-517.3.7.obj~1/RELEASE_PPC Power Macintosh powerpc
marke$ distcc --version
distcc 2.0.1-zeroconf powerpc-apple-darwin7.0 (protocol 1) (default port 3632)
built Dec 9 2003 20:54:55
Of course this slashdot discussion is a day old already and its unlikely anyone will see this...
AC(s): 2
biginsa500: 0
Wanna come back for Final Jeopardy, where the scores can really change?
Coincidentally, I was working on setting up Sun Grid Engine on a couple of G5s here at work yesterday when I heard the news about XGrid. I dropped my work on SGE immediately.
It was a little confusing at first to set up, but I eventually got 3 machines total configured as part of my grid - 2x2GHz G5s + 1x867MHz G4. Perhaps if I read the instructions, I would've better understood some of the terminology (agent v. controller v. client, etc.).
The tachometer is sort of flaky. Sometimes it's stuck at zero on one machine while it it is actively moving around on others. Other times it's stuck at some non-zero position. Opening up 2 tachs on the same machine (XGridBlast has its own tach) will show different speeds. Though I should have in theory 8.867 GHz total speed, I could never get it to go over the default 8 "red line"- I was curious if it would rescale once it exceeded 8.
The XGrid client (where you submit jobs) has some default demo type things (I've mostly been testing the Mandelbrot one as it runs in a continuous loop), but it also has a way to build "custom plug-ins" which allow you to submit arbitrary jobs. In other words, executables don't have to be modified per se. Of course for any kind of parallel execution, they do, but if you need to to run 1000 iterations of the same command with slightly different arguments, then it should be able to distribute that "run" pretty well. The GUI for building up such a run is pretty easy to use.
One potentially big issue I've noticed is that once you submit a job, you have to keep the XGrid app open until it is done. For a run that may take days or weeks, I think this could be a problem. I'd like for you to be able to submit the job, quit, and log out (or shutdown the client even) then come back later to check your results.
Also, there doesn't seem to be a queue manager where you can see a list of jobs and their states.
I think that for any file-dependent commands, you need local copies of the files on each Agent node. At least that's how it appeared from perusing the documents for XGrid Blast - each node needs a copy of the database.
FWIW, we're using SGE successfully on a bunch of RedHat based servers, but though OSX maybe *nix, the installation and config was turning to be a pain.
Anyway, those are my initial impressions. I'm sure some of these things will change in the "release" version. In the meantime, I'll have to get working on getting some real work stuff to try out (I work in a university bio department).
-h3
Actually, this probably isn't as difficult as it seems. GCC compiles ObjC (in point of fact, Apple uses GCC to compile OS X), and Cocoa is just Apple's implementation of the OpenSTEP API which was originally laid out back in the NeXT days (it's actually pretty remarkable how similar it is, just move the application's menus from the corner of the screen to a bar across the top and pretty things up and you've got yourself a brand new NeXT). Anyway, GnuSTEP is another OpenSTEP compatible environment, and I suspect that the Cocoa interface wouldn't take too much work to port over.
Anonymous Luddite: "What do you think of the dehumanizing effects of the Internet?"
Andy Grove: "Not Much."
Status: Job failed (task: failed with status 255). I suspect that was because I was trying to run it off an xgrid consisting of one G3 iBook, and it needs the velocity engine (though nothing of the sort was documented). However if that is the case it shouldn't have failed, it should have waited to run off of other computers and just ignored the G3.
I'm potentially seeing a situation where some well meaning non-compatible agent joins an open XGrid and flubs someone's whole job. There needs to be a way to not fail just because one agent can't handle a particular program.
Also I'd like to know how well the xgrid processes behave themselves when an agent is "Always available" but is working on other stuff. So far things seem to work OK.
---If you can't trust a nerd, who can you trust?
Someone feel free to correct me if I'm wrong, but it doesn't seem that XGrid offers anywhere near the flexibility and power of Sun Grid Engine, which is open source and currently supports OS X among many other Unix platforms. SGE allows for tight integration of MPI processes, works for heterogeneous and distributed clusters, has lots of configurable options for remote monitoring, prioritization, and submission of jobs, and works well from the shell. The main selling point of XGrid appears to be the ease of configuration, which makes a difference particularly if your cluster of machines is "fluid" rather than fixed.