Apple to Award Workgroup Clusters to Scientists

Graff writes "Apple is giving away five Apple Workgroup Clusters for Bioinformatics (each worth approximately $40,000) to four higher education researchers and one non-education researcher. A panel of independent scientists and Apple will choose the lucky researchers."

Comments

Apple and bioinformatics

[Neil Blender]

Apple (as well as other computer companies like IBM) are getting very interested in bioinformatics. They have loaned us a ton of equipment for free even though our product is linux based. Of course, Apple has always had a stronghold in academics.

Re:Apple and bioinformatics

[harvardian]

"Apple has always had a stronghold in academics"

That's a bit broad. K-12 education, sure, but at the bioinformatics lab I worked with, we worked exclusively with IBM. The attractiveness of using consumer-level Macs in a grade school setting most certainly doesn't translate to a high-performance computing environment. That might change as Apple moves into this space, however...we'll see.

Re:Apple and bioinformatics

[the gnat]

When I worked at the University of Washington doing life sciences research, my personal observation saw it to be about 50/50 pc vs mac.

It depends on the branch of life sciences, and more importantly, the biases of the individual professor.

In most computational labs, ALL of the computing is done on Unix, which nowadays means mostly Linux. In crystallography, people still hang on to SGIs but there's a lot of migration towards both Linux and Mac. In fact, my impression is that Apple may be making real progress here, because people explicitly want a computer that can handle both graphics apps, Unix-based physics programs, and PowerPoint/Word.

My previous lab was entirely bioinformatics, and the majority of people used Windows on their desktop, usually a Dell workstation or a Thinkpad. The reason for this had nothing at all to do speed or application availability - it was almost entirely due to laziness or prejudice. When new people joined that lab, they were generally either very computer-savvy and already ran Linux, or they were clueless and needed one of us gurus to set up their computer - so they got Linux. We'd made some progress migrating people, but the boss was such a Windows addict that there's a limit to what we could have done.

There were actually a number of people who would have preferred Macs, but the boss refused to buy them on principle - which principle, I'm not sure. However, most professors are very Mac-friendly; I'd guess that well over half of the ones in my department (mol. bio.) use Macs. Their lab members, however, tend to be very anti-Apple.

This bias is the chief thing Apple will need to overcome if they want their computers to be used for HPC. Most of the apps are custom-coded or open-source to begin with, so there's little or no barrier to using them on OS X versus any other platform. The people using them, however, may be very good biologists or programmers but they know nothing about the computer systems except thirdhand knowledge that hasn't been true since the 1990s. This is why everyone I work with insists on using our old SGIs, while my laptop is several times faster for the same tasks, and they won't touch a Mac unless there's nothing else around.

What I've found really surprising is that of all the comp bio people of various stripes that I've known, those with the most hardcore programming or sysadmin experience have on average been the most Mac-friendly by a long shot.

Re:Apple and bioinformatics

[Mocenigo]

I attend cryptographic conferences on a regular basis, being a young scientist in the field.

Three years ago, I saw NO mac laptops in the audience. 300 people, all mathematicians,computer scientists and electrical engineers, half of them with a laptop, no mac.

Two years ago, some appeared.

Last years, one in twenty had a mac laptop.

This year, one in ten. It's amazing how people are "discovering" powerbooks and ibooks.

And you can tell the owner from the machine!

Firm execs or young academicians backed by a middle-class family which choose not to buy a car (read: me): 17". Others in academia, many from software firs: 15". Women and Ph.D. Students: iBooks :)

And then there are the old-style professors (like my boss) that would NEVER buy a laptop without floppy disc drive, look in disgust at everything that does not run Windows 98, shake their heads when people use USB memory sticks to exchange data...

United Devices

[kyoko21]

Apple should give a set away to United Devices (Profit) or Grid. Both of these ventures specialize in distributed Cancer/Drug simulations. Let's find a cure for breast and prostate cancer!!!! Go Go Go!!!

Re:More trouble than it's worth?

[Erwos]

Um, no.

Market share and install base are definitely linked. If Apple's only selling 3% of computers, their install base is going to trend to 3% over time, holding all other things equal.

Mathematical example: total market is 100 computers, Apple has 3% market share)
Year 0: 6 Macs, 150 PCs (so Apple has about 3.8% of the install base when we start)
Year 1: 9 Macs, 247 PCs (install base is 3.5%)
Year 2: 12 Macs, 344 PCs (install base is 3.3%)
Year 3: 15 Macs, 441 PCs (install base is 3.2%)

I think you get the picture. Market share is not representative of total install base RIGHT NOW, but is certainly a good indicator of what's going to happen in the future. If you disagree, that's too bad, because I've just mathematically shown that you're wrong. Market share and install base are definitely linked.[1]

Apple's profitability really has nothing to do with their install base so much as their margins. If I'm selling stuff with a huge mark-up on actual costs, I could sell 30 pieces of it and still make money, even if the total market is 3 billion pieces.

-Erwos

[1] As for your "PCs don't last as long as Macs": prove it. I've used Macs for years, and Apple's build quality is not as good as people make it out to be. I'm not going to factor in differing "computer decays" without any kind of proof for them.

Re:BFD

[Daniel Dvorkin]

Probably because it's not an outright grant, but a contest which (theoretically) anyone can enter. And there are a fair number of /.ers who might be interested. I'd enter myself, but my chances of winning as a grad student are probably somewhere between 0 and NULL.

computing power is unfairly distributed

[SuperBanana]

For the mad scientist who has everything!

If your definition of "mad scientist" is "person working on weapons of mass destruction", ie, nuclear weapons, most of them already have the world's largest clusters. Pretty sad that we still consider it important to build better nuclear weapons even though we've got thousands of them, and not a single legitimate target for them(the whole deterrence thing is ridiculous- if it's just about deterrence, we only need a dozen or so).

It'd be nice to see some computing horsepower, if only a small piece, go to those trying to do something other than make better nuclear bombs or look for little green men...ie something (gasp) productive.

Re:The award should be for PCs

[strook]

Most bioinformatics software, Windows versions are barely supported. Blast, the SAM toolkit, Clustal_, belvu, BioPerl and BioPython, all of these work perfectly on Mac OS X. Among the more computational bioinformaticists it's very handy to be able to recompile the publicly available software for your needs. Also, it's very common in bioinformatics to have questions you need to answer that don't exactly fit the parameters of the software, so it's important to have an environment where it's easy to write scripts to analyze text files and control the (possibly distributed) running of algorithms. In short, the field of bioinformatics is a perfect fit for Unix-based OSes and a fairly godawful fit for Windows. I don't think this is pure slash-bias; I think most all bioinformatics researchers would agree.

This part slightly OT, but this reminds me how much better bioinformatics tools would be if there were more people who could contribute to the open source tools in the field. Often times a widely used program is released open source, but there are so few people who can code well and also take notice of bioinformatics tools that bugs don't get solved like they could. Somebody please make belvu stop crashing all the damn time, make phylip accept alternate data formats, et cetera... I've already got my advisor's software to debug.

Re:Apple clusters?

[Ffakr]

I do realize your comment is tongue in cheek.. but Bioinformatic tools (some of them like BLAST) run multiples faster on Apple hardware than on x86 hardware. Apparently apps like BLAST really run great on Altivec.
I haven't seen anything recently, but at one time BLAST ran up to 16x faster on a G4 than a P4.

Re:Calculating the payoff

[Anonymous Coward]

Don't be silly, Apple is a very small fish in a very big pond compared to real supercomputer companies like SGI.

What the big companies do is let you login to their machines free of charge, over the internet, and take their machines for a spin. See e.g. http://www.testdrive.compaq.com

On top of that, what the companies with a *serious interest* in high performance computing (like SGI and Cray) do is let you email your program to them: they will spend days or weeks (up to you) tuning it, telling you how it works on their hardware, and even telling you that you're better off with the competition, if that ends up being the case. What these companies understand very well, that Apple doesn't, is that people who spend $10M+ on computing equipment, typically with public money, need to show due diligence in their choice of hardware. This means THOROUGH evaluation: benchmarks, benchmarks, benchmarks - and not _standard_ benchmarks, but the benchmarks that matter: the user's own software.

Take a look here to see a real-life snapshot of this kind of process.

Any company that tried to "fudge the numbers" would be caught out, and that looks VERY bad. So the companies instead do all they can to help with the evaluation process, and hope that they get chosen. If not, there's always the next sale. A big supercomputer is sold somewhere every day or two, after all.

Virginia Tech's "X" doesn't come into this category: for example, no actual scientific work has been performed on the machine so far apart from benchmarks/system development, etc. Dr Varadarajan is probably going to get a pretty nasty grilling in a couple of years, when the University asks "so, what did we get for our money?" Some P.R., and not a whole lot more it would seem. I wish them luck...

Apple is not really a supercomputer company: and really, who cares? Apple has been, is, and will probably remain a great computer company for a good long while yet. But don't expect Apple to help you out a great deal if you pop them an email going "say, how does the G5 run on the ASCI Purple benchmarks?". That's just not what Apple are about. (example: they are happy to sell you a box with a 64-bit CPU, but they could care less that they don't have a 64-bit OS to go with it..)

Re:Apple

[nettdata]

You're right...

My cousin does cancer research at Harvard, and I design/write software for a living. He found that there was a huge empty space for software that would help him do his job (cancer research), so 2 years ago we started a software company that specializes in reagent management (cryogenic storage, dna plasmids, oligos, antibodies, protocols, animal experiments, etc., all cross-referenced), and made sure that it was 100% Java and cross-platform.

While we really have no direct competition (yet), it is very interesting to see the platform requirement limitations (mostly DOS/Windows) that a lot of the other software companies have. There really is a huge shortage of cross-platform software.

Our experience has shown that most commercial labs tend to be Windows based, while most academic labs are Mac based. It is also not uncommon to have the Academic labs have 1 or 2 Win32 boxes that are there just to run a particular program they're using. It also appears that the IT departments in academia tend to use Linux back-end servers, with an interestingly high occurance of Yellow Dog. (That's Linux on PPC, for those of you unfamiliar with YD). Usually, we've found that the YD servers are older G3 and G4 towers that have been repurposed.

Now, these are the environments that we've been exposed to, and may not accurately represent the group as a whole, but regardless, it's been an interesting and enlightening experience seeing what/how different labs are currently (and used to) using by way of software.

Good to see It just works computing in the Lab

[gsdali]

Think of the time saved by being able to buy a preconfigured cluster. Just open the box plug in, fire up and go. Now Beowulf may be cheaper, butthink of the time saved by just being able to buy a cluster off the shelf. Think of the time saved by not having to set up and configure. I don't really mind who's selling this kind of solution but I'm glad someone someone is.

Roll on clustered Shake and FCP renndering though.

Ah Yup...

[eyegor]

I'm dragging around a five year-old Apple laptop and it's still a very usable system (unlike my two year old Dell laptop). It's stable, light, and "just works". I recently handed out an old G3 laptop as a testing system to a co-worker. It's a bit slow, but it still gets the job done.

And no. I'm not some mindless Mac drone. I've been using and admining x86 and SPARC boxes since forever.

I use Solaris, Linux, OS X and Windows on a daily basis, and I'm totally blown away with the stability of OS X and how well the OS is integrated (ah, the joys of a propriarity system).

That being said, there are a few "Mac-isms" that drive me nuts (like command line user management, for instance).