benchmarking.

[IainS]

Thanks Dominque,
interesting, if somewhat depressing (most of my machines are PPC).

could I have a copy of  ".par" and pbharness.f90/pbvalid.f90 that  
you've used?

Iain

On 15 Jun 2008, at 14:57, Dominique Dhumieres wrote:

> On a G5 1.8Ghz I get:
>
> ====================================================================== 
> ==========
> Date & Time     : 19 May 2008 11:53:13
> Test Name       : pbharness
> Compile Command : gfc %n.f90 -m64 -O3 -ffast-math -funroll-loops - 
> ftree-loop-linear -fomit-frame-pointer -finline-limit=600 --param  
> min-vect-loop-bound=2 -o %n
> Benchmarks      : ac aermod air capacita channel doduc fatigue  
> gas_dyn induct linpk mdbx nf protein rnflow test_fpu tfft
> Maximum Times   :      300.0
> Target Error %  :      0.200
> Minimum Repeats :     2
> Maximum Repeats :     5
>
>    Benchmark   Compile  Executable   Ave Run  Number   Estim
>         Name    (secs)     (bytes)    (secs) Repeats   Err %
>    ---------   -------  ----------   ------- -------  ------
>           ac      9.91       34112    116.23       2  0.0030
>       aermod    265.08     1257488     91.16       4  0.2099
>          air     17.70       56576     41.27       5  0.8290
>     capacita      8.19       60200    185.99       2  0.2296
>      channel      5.89       30184     15.56       5  0.2732
>        doduc     38.49      134240     69.69       2  0.0689
>      fatigue     12.84       60040     21.46       3  0.1918
>      gas_dyn     17.03      679696     55.47       2  0.1695
>       induct     30.80      123592     66.08       2  0.0212
>        linpk      2.24       25832     34.05       4  0.1839
>         mdbx      8.94       56440     33.10       2  0.1541
>           nf     21.63       58728     56.96       2  0.1220
>      protein     24.75       85216     84.38       2  0.0646
>       rnflow     30.08      146512     84.22       4  0.1690
>     test_fpu     16.79       92464     35.70       2  0.1499
>         tfft      2.82       26144     20.81       3  0.1552
>
> Geometric Mean Execution Time =      51.37 seconds
>
> ====================================================================== 
> ==========
> Polyhedron Benchmark Validator
> Copyright (C) Polyhedron Software Ltd - 2004 - All rights reserved
>
> Dominique
>
> PS Note the very bad result for ac.f90, otherwise the timings are on
> average three times slower than on a Core2Duo 2.16Ghz