Re: [Beowulf] What services do you run on your cluster nodes?

On Fri, 26 Sep 2008, Donald Becker wrote: But that rule doesn't continue when we move to higher core counts. We still want a little observability, but a number for each of a zillion cores is useless. Perhaps worse than useless, because each tool has to make its own decision about how to summar

Re: XML alternatives [was Re: [Beowulf] What services do you run on your cluster nodes?]

Ashley Pittman wrote: On Thu, 2008-09-25 at 19:11 -0400, Joe Landman wrote: YAML is nice until you accidentally change indentation. Then the game is over. Having bolluxed up many fortran codes in my (distant past), I have to say "just say no" to things *requiring* indentation. In YAML's cas

Re: XML alternatives [was Re: [Beowulf] What services do you run on your cluster nodes?]

Another bit of FORTRAN trivia: the first card readers could only read 36 columns at a time and they made two passes over the card; hence, 72 columns per card. ...we now return you to your regularly scheduled off-topic diversion. george wm turner high performance systems 812 855 5156 On Se

Re: [Beowulf] What services do you run on your cluster nodes?

On Thu, 25 Sep 2008, Robert G. Brown wrote: > Ugly vs pretty is irrelevant. The data is clearly and cleanly > organized, easy to extract and put into structures, capable of dealing > with the natural variability of hardware configuration on the sending > side -- multiple interfaces, CPUs all hand

Re: [Beowulf] mpich vs hp mpi performance

At 19:47 26.09.2008, Joshua Baker-LePain wrote: DYNA has 2 main solver types -- explicit (iteration based) and implicit (matrix inversion, essentially). If you're only using one system and the explicit solver, then there's no need for any MPI, as ls-dyna is multi-threaded. I ran both SMP v

Re: [Beowulf] mpich vs hp mpi performance

On Fri, 26 Sep 2008 at 6:28pm, Linus Harling wrote Thank you for the input! The system is a Dell T605 2x4core Opteron 2354 with 32GB RAM. With a second one like it being added (probably using ethernet) if need arises. Which means most/all of the communication will be by shared memory to begin wi

Re: [Beowulf] mpich vs hp mpi performance

Igor Kozin skrev: > Hi Linus, > check out IMB/PMB results in our benchmarking database > http://www.cse.scitech.ac.uk/disco/dbd/ Thanks! Great resource! /Linus > Best, > Igor > > 2008/9/25 Linus Harling <[EMAIL PROTECTED] > > > Hello List. > > I've been search

Re: [Beowulf] mpich vs hp mpi performance

Alan Louis Scheinine skrev: > It depends very much on hardware and on the specific program. > My experience with MPICH on Ethernet and MVAPICH, MVAPICH2, > OpenMPI on Infiniband is that different programs find one > or the other better, no clear winner. Also, small changes > in some options that o

[Beowulf] Benchmark/apps showing benefit from instruction set advances

Hi, I am searching an application or benchmark written in C, C++, or some ftn dialect. The benchmark must show that the more advanced instruction set enabled by the compiler, the increased performance is achieved whilst running on a modern machine. For example, using the Intel compiler an

Re: XML alternatives [was Re: [Beowulf] What services do you run on your cluster nodes?]

On 9/26/08 3:56 AM, "Simon Cross" <[EMAIL PROTECTED]> wrote: > > Fortran was bad because you had to line things up in particular > columns [1]. The "significant whitespace is bad" was really just a > slogan for "we don't like how early versions of Fortran did things". > Python (and Haskell) don'

Re: XML alternatives [was Re: [Beowulf] What services do you run on your cluster nodes?]

On Fri, Sep 26, 2008 at 1:11 AM, Joe Landman <[EMAIL PROTECTED]> wrote: > YAML is nice until you accidentally change indentation. Then the game is > over. Having bolluxed up many fortran codes in my (distant past), I have to > say "just say no" to things *requiring* indentation. In YAML's case,

Re: XML alternatives [was Re: [Beowulf] What services do you run on your cluster nodes?]

On Thu, 2008-09-25 at 19:11 -0400, Joe Landman wrote: > YAML is nice until you accidentally change indentation. Then the game > is over. Having bolluxed up many fortran codes in my (distant past), I > have to say "just say no" to things *requiring* indentation. In YAML's > case, it suffers fr