Re: [Beowulf] A start in Parallel Programming?

On Wed, Mar 14, 2007 at 07:39:03PM -0400, Mark Hahn wrote: > >When curmudgeonly addressing everyone, you want the super-plural, > >"all y'all". > > I thought it was "all yunz" - curmudgeons don't alliterate ;) Robert is in the South, "all youse guys" is a Northeast super-plural. -- greg, raised

Re: [Beowulf] A start in Parallel Programming?

When curmudgeonly addressing everyone, you want the super-plural, "all y'all". I thought it was "all yunz" - curmudgeons don't alliterate ;) ___ Beowulf mailing list, Beowulf@beowulf.org To change your subscription (digest mode or unsubscribe) visit h

Re: [Beowulf] What is a "proper" machine count for a cluster

I want to initiate a project involving a beowulf cluster. I have two engineers who are VERY interested. what kinds of things do they do? Anyway. I was wondering what would be a "decent" amount of equipment to start off and the preferred linux if any. I've run full-scale cluster environme

Re: [Beowulf] Is Beowulf a standard?

As as a newbie I have a question, "Is Beowulf a clustering standard?" if yes I would call it a defacto standard, but I'm fairly liberal ;) "What makes Beowulf a standard" As I read about Beowulf, it appeared to me as a method for starting Linux clustering, but some people call it a standard, I

Re: [Beowulf] A start in Parallel Programming?

exceptions - g95 is pretty good, and intel's ifort is free for academic use.) unless this has changed, Intel compilers are free for "non-commercial" use, which Intel defines somewhat unusually. an academic cannot use it for research, such research is probably compensated-for. that would even a

Re: [Beowulf] Re: A start in Parallel Programming?

On Wed, Mar 14, 2007 at 11:29:51AM -0700, David Mathog wrote: > > Conversely, the CS departments like to teach with idealized didactic > computing languages. What that language is changes from era to > era, but they are in any case notable for rarely being used to > accomplish anything significan

RE: [Beowulf] Weird problem with mpp-dyna

Joshua, I'm willing to bet this is an issue with the mpp solver and good luck trying to get any help from LSTC. We have ran into so many quirks with the mpp solver. Most of our work are FE models. Mark McCardell Computer Systems Engineer Center for Applied Biomechanics - University of Virginia w

Re: [Beowulf] Benchmark between Dell Poweredge 1950 And 1435

On Mon, Mar 12, 2007 at 01:40:42PM -0700, Bill Broadley wrote: > Subject: Re: [Beowulf] Benchmark between Dell Poweredge 1950 And 1435 > Cc: beowulf@beowulf.org > > Renato S. Silva wrote: > >Hi > > > >What is the best compiler for AMD processors ? Bill is exactly correct... the answer depends on

[Beowulf] What is a "proper" machine count for a cluster

Hello all, A brand new noob entering this realm. I want to initiate a project involving a beowulf cluster. I have two engineers who are VERY interested. Anyway. I was wondering what would be a "decent" amount of equipment to start off and the preferred linux if any. I know the basic an

Re: [Beowulf] A start in Parallel Programming?

I have a few data points to add to the discussion: (1) FORTRAN is easy to learn. I think most of the college kids I work with could pick up 75% of the F77 language variant in an afternoon's reading (a good reference for learning the important parts of FORTRAN is "Classical Fortran" by Kupf

[Beowulf] Is Beowulf a standard?

Hi there, As as a newbie I have a question, "Is Beowulf a clustering standard?" if yes "What makes Beowulf a standard" As I read about Beowulf, it appeared to me as a method for starting Linux clustering, but some people call it a standard, I couldn't understand that what makes Beowulf a standard,

Re: [Beowulf] Re: A start in Parallel Programming?

On 3/14/07, Robert G. Brown <[EMAIL PROTECTED]> wrote: On Wed, 14 Mar 2007, Peter St. John wrote: > What General Physics I teaches about wiring would not be adequate to work as > an electrician in home construction, but it's adequate to do the bench > experiments that illustrate the concepts. T

Re: [Beowulf] Re: A start in Parallel Programming?

On Wed, 14 Mar 2007, Peter St. John wrote: What General Physics I teaches about wiring would not be adequate to work as an electrician in home construction, but it's adequate to do the bench experiments that illustrate the concepts. There are many purposes under the sun. I'd take issue with th

Re: [Beowulf] A start in Parallel Programming?

On Wed, Mar 14, 2007 at 12:20:12PM -0400, Robert G. Brown wrote: > Y'all are so cynical, really. Robert, When curmudgeonly addressing everyone, you want the super-plural, "all y'all". This message has been brought to you by the letter C and the number googol. -- g

Re: [Beowulf] Re: A start in Parallel Programming?

On Wed, 14 Mar 2007, David Mathog wrote: To follow the example of the automotive mechanic: Universities think they are producing Mechanical Engineers, not automobile mechanics. Train the Mechanical Engineering students entirely in theory, turn them loose at Ford, and unless somebody who real

Re: [Beowulf] Re: A start in Parallel Programming?

> "Robert G. Brown" <[EMAIL PROTECTED]> wrote > > >I'm kind of glad the folks who teach CS this way don't teach foreign >language too - they'd make the students learn a fair amount of Latin >before letting them enroll in a Spanish class! Sure Spanish is based >on Latin, but "Ubi latrina est?" i

Re: [Beowulf] Re: A start in Parallel Programming?

At 11:47 AM 3/14/2007, Peter St. John wrote: David, To follow the example of the automotive mechanic: Universities think they are producing Mechanical Engineers, not automobile mechanics. The somewhat condescending model I was exposed to was: The University of California produces researchers a

Re: [Beowulf] Re: A start in Parallel Programming?

At 11:29 AM 3/14/2007, David Mathog wrote: "Robert G. Brown" <[EMAIL PROTECTED]> wrote I'm kind of glad the folks who teach CS this way don't teach foreign language too - they'd make the students learn a fair amount of Latin before letting them enroll in a Spanish class! Sure Spanish is based

Re: [Beowulf] A start in Parallel Programming?

At 10:39 AM 3/14/2007, Joe Landman wrote: Robert G. Brown wrote: So I'm not holding my breath on ML running out this week or next. I'm more interested in speculating on when the next massive superML jump on TOP of ML will occur, when the next phase/paradigm shift is due that Ah Ok, my c

Re: [Beowulf] A start in Parallel Programming?

> How many Moore doublings until we reach it ... Robert ... ?? Seven. Exactly seven. I'm sure you're kidding, but I'm thinking: ballpark 10 32-bit GHz per kilogram, current commodity (surely not 100, considering power supply, rack mount, fan, etc). So ballpark 3 x 10^8 bits per second per gr

Re: [Beowulf] Weird problem with mpp-dyna

Joshua Baker-LePain wrote: On Wed, 14 Mar 2007 at 2:54pm, Joe Landman wrote Joshua Baker-LePain wrote: They also provide the matching pre-compiled LAM/MPI libraries on their site. For a fun little wrinkle, RHEL/CentOS ships LAM/MPI 7.0.6. However, the spec file in their RPM does *not* includ

Re: [Beowulf] Weird problem with mpp-dyna

On Wed, 14 Mar 2007 at 2:54pm, Joe Landman wrote Joshua Baker-LePain wrote: They also provide the matching pre-compiled LAM/MPI libraries on their site. For a fun little wrinkle, RHEL/CentOS ships LAM/MPI 7.0.6. However, the spec file in their RPM does *not* include the --enable-shared flag.

[Beowulf] Re: A start in Parallel Programming?

> To follow the example of the automotive mechanic: Universities think they > are producing Mechanical Engineers, not automobile mechanics. Train the Mechanical Engineering students entirely in theory, turn them loose at Ford, and unless somebody who really does work on cars steps in to prevent

Re: [Beowulf] A start in Parallel Programming?

Peter St. John wrote: Ah, great, I wanted something like the reference to Bremermann, thanks. The Wiki item cites no references (!) but the discussion page does. The item http://www.gwu.edu/~umpleby/recent_papers/2004_physical_relationships_among_matter_energy_information_umpleby.htm

Re: [Beowulf] A start in Parallel Programming?

Peter St. John wrote: I think we're zooming to the future, but there is a wall in our future. ... Alrighty then ... -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics LLC, email: [EMAIL PROTECTED] web : http://www.scalableinformatics.com phone: +1 734 786 8423 fax : +1 734 78

Re: [Beowulf] Re: A start in Parallel Programming?

> Let's see, what language is CS is using here these days? It has > been a while since I looked: > > CS 1 (Introduction to Computation, first quarter) uses Scheme. > CS 2 (Introduction to Programming Methods, 2nd quarter) seems > to be mostly Java. > CS 3 (Introduction to Software Engineering

Re: [Beowulf] Weird problem with mpp-dyna

Joshua Baker-LePain wrote: Do you use a statically linked binary or did you relink it with your mpich? Agh. I forgot to mention this little wrinkle. LSTC software distribution is... interesting. Yup. Caused us lot of fun at some customer sites. For mpp-dyna, they ship dynamically link

Re: [Beowulf] Weird problem with mpp-dyna

Joshua Baker-LePain wrote: Running a simulation via 'mpirun -np 12' works just fine. Running the same sim (on the same virtual machine, even, i.e. in the same 'lamboot' session) with -np > 12 leads to the following output: [...] *** Error the number of solid elements 13929 defined on the

Re: [Beowulf] Re: A start in Parallel Programming?

David, To follow the example of the automotive mechanic: Universities think they are producing Mechanical Engineers, not automobile mechanics. Duke does not have any courses (that I knew of when I was there) good for college credit, that taught repairing real cars, and Duke is not where I'd send a

Re: [Beowulf] A start in Parallel Programming?

On Wed, 14 Mar 2007, Richard Walsh wrote: Joe Landman wrote: Peter St. John wrote: So maybe the ceiling will be (Deliverable FLOPS times Recoverable Bytes)/ml I rather like the Strossian MIPS per kilogram ... :) I am partial to the Bremermann limit: Hans Bremermann's conjecture that ?no d

[Beowulf] Re: A start in Parallel Programming?

"Robert G. Brown" <[EMAIL PROTECTED]> wrote > On Wed, 14 Mar 2007, Richard Walsh wrote: > > Time makes fools of us all, but especially CS departments ... > > Yeah, but as David pointed out, they actively try... I wouldn't go quite that far, rather I think the situation is more like this: Mech

Re: [Beowulf] A start in Parallel Programming?

Richard Walsh wrote: Joe Landman wrote: Peter St. John wrote: So maybe the ceiling will be (Deliverable FLOPS times Recoverable Bytes)/ml I rather like the Strossian MIPS per kilogram ... :) I am partial to the Bremermann limit: Hans Bremermann's conjecture that “no data processing system,

RE: [Beowulf] Weird problem with mpp-dyna

On Wed, 14 Mar 2007 at 10:05am, Michael Will wrote You mentioned your own code does not exhibit the issue but mpp-dyna does. Yep. What does the support team from the software vendor think the problem could be? They say that we have an academic license which does not entitle us to any supp

Re: [Beowulf] A start in Parallel Programming?

Ah, great, I wanted something like the reference to Bremermann, thanks. The Wiki item cites no references (!) but the discussion page does. The item http://www.gwu.edu/~umpleby/recent_papers/2004_physical_relationships_among_matter_energy_information_umpleby.htm looks just like what I want, but su

Re: [Beowulf] A start in Parallel Programming?

Joe Landman wrote: (3 or 4, I forget) Zeroth: You must play the game. First: You can't win. Second: You can't break even. Third: You can't quit the game. Gahh attribution is the Wikipedia entry for thermodynamics. Heard this joke 20+ years ago ... -- Joseph Landman, Ph.D Founder an

Re: [Beowulf] A start in Parallel Programming?

Well I like the vapor-deposition diamond thing. You'd slice the diamond into thin wafers and treat it just like silicon, except it conducts heat way better. I think it's already cheap enough to trivialize the cost of the thickness you'd want for a chip, but they are working on making usefully wide

Re: [Beowulf] A start in Parallel Programming?

Joe Landman wrote: Peter St. John wrote: So maybe the ceiling will be (Deliverable FLOPS times Recoverable Bytes)/ml I rather like the Strossian MIPS per kilogram ... :) I am partial to the Bremermann limit: Hans Bremermann's conjecture that “no data processing system, whether artificial o

Re: [Beowulf] A start in Parallel Programming?

Richard Walsh wrote: As far as what you CS department is teaching (from what you described and from the point of view of modern high-performance scientific computing), I would be careful not to fall in love with CS departmental fads ... cross check all local CS-temporo-sectarianisms here on this

Re: [Beowulf] A start in Parallel Programming?

Peter St. John wrote: Joe Landman wrote: "...if you buy into the notion that there may in fact be some fundamental physical limits on Moore's law due to the nasty combination of thermodynamics (stability of small structures with respect to temperature, and resistance of same to defect format

Re: [Beowulf] A start in Parallel Programming?

Robert G. Brown wrote: So I'm not holding my breath on ML running out this week or next. I'm more interested in speculating on when the next massive superML jump on TOP of ML will occur, when the next phase/paradigm shift is due that Ah Ok, my concern is that silicon may be running out o

Re: [Beowulf] Weird problem with mpp-dyna

I think I"d just grep for 13 in all the scripts and confiuration files. Of course with my luck it would be 013 :-) Peter On 3/14/07, Michael Will <[EMAIL PROTECTED]> wrote: You mentioned your own code does not exhibit the issue but mpp-dyna does. What does the support team from the software v

RE: [Beowulf] Weird problem with mpp-dyna

You mentioned your own code does not exhibit the issue but mpp-dyna does. What does the support team from the software vendor think the problem could be? Do you use a statically linked binary or did you relink it with your mpich? We have ran lstc ls-dyna mpp970 and mpp971 across more than 16 nod

Re: [Beowulf] Weird problem with mpp-dyna

On Wed, 14 Mar 2007, Peter St. John wrote: I just want to mention (not being a sysadmin professionally, at all) that you could get exactly this result if something were assigning IP addresses sequentially, e.g. node1 = foo.bar.1 node2 = foo.bar.2 ... and something else had already assigned 13 to

Re: [Beowulf] A start in Parallel Programming?

On Wed, 14 Mar 2007, Peter St. John wrote: Joe Landman wrote: "...if you buy into the notion that there may in fact be some fundamental physical limits on Moore's law due to the nasty combination of thermodynamics (stability of small structures with respect to temperature, and resistance of sam

Re: [Beowulf] A start in Parallel Programming?

On Wed, 14 Mar 2007, Richard Walsh wrote: As far as what you CS department is teaching (from what you described and from the point of view of modern high-performance scientific computing), I would be careful not to fall in love with CS departmental fads ... cross check all local CS-temporo-secta

Re: [Beowulf] Weird problem with mpp-dyna

I just want to mention (not being a sysadmin professionally, at all) that you could get exactly this result if something were assigning IP addresses sequentially, e.g. node1 = foo.bar.1 node2 = foo.bar.2 ... and something else had already assigned 13 to a public thing, say, a webserver that is not

Re: [Beowulf] A start in Parallel Programming?

Kyle Spaans wrote: I just read this article, , and it has rekindled my desire to get started in parallel and multi-threaded programming. By now, I'm plenty comfortable with Linux Kyle, One could argue that parallel programming parad

Re: [Beowulf] A start in Parallel Programming?

Joe Landman wrote: "...if you buy into the notion that there may in fact be some fundamental physical limits on Moore's law due to the nasty combination of thermodynamics (stability of small structures with respect to temperature, and resistance of same to defect formation), and quantum mechanic

Re: [Beowulf] Re: A start in Parallel Programming?

On 3/13/07, David Mathog <[EMAIL PROTECTED]> wrote: ... if(i==1){} else if(i==2){} etc. Hopefully things have improved since then. It is possible to force a sort of computed goto in C (transfer control staying within a function) but it involves pre-storing addresses of labels in an array and th

[Beowulf] Weird problem with mpp-dyna

I have a user trying to run a coupled structural thermal analsis using mpp-dyna (mpp971_d_7600.2.398). The underlying OS is centos-4 on x86_64 hardware. We use our cluster largely as a COW, so all the cluster nodes have both public and private network interfaces. All MPI traffic is passed on