[Wien] Wien2k benchmark.
Peter Blaha
pblaha at zeus.theochem.tuwien.ac.at
Fri Sep 26 09:04:31 CEST 2003
I have repeated this benchmark on our dual-chanel-memory P4s.
As expected they give the best performance. (The fastest Itanium processors
might be similar, but probably much more expensive)
> We have run one test on different platforms using various
> compilers and libraries. The result is in seconds and represents the CPU
> time taken by lapw1c to calculate one k-point for the test case.
>
> Compaq-Alpha 666 1092 sec Dec compiler
> G4 4877 sec /-O Absoft compiler
> G5(dual CPU64bits 2GHz) 690 sec Absoft/-O
> G5 350 sec /xlf compiler /G5 64bits
> libraries /-O5 -qhot -arch=g5
> Athlon XP3000+ (2.17 GHz) 1444 sec (PGI, PGI compiled LAPACK
> and BLAS)
> Athlon XP3000+ (2.17 GHz) 705 sec ifc, mkl
> Athlon XP3000+ (2.17 GHz) 541 sec ifc, ifc compiled LAPACK,
> Athlon ATLAS)
> Athlon XP3000+ (2.17 GHz) 515 sec (PGI, PGI compiled LAPACK,
> Athlon ATLAS)
P4, 2.5 GHz, Asus P4G8X Delux/GD 347 sec ifc, mkl
Board (FSB533/400, dual chanel
memory architecture (THIS IS
IMPORTANT!!)
P4, 3GHz (similar dual memory b.) 288 sec ifc, mkl
For more details I list part of the timings of the 2.5 GHz run.
Time for al,bl: 9.00000000000001D-002
nlo 304
Time for los: 4.38000000000000
Time for LOOP 260: 76.7500000000000
HNS: Loop 140: 0.0 Loop 90: 0.2 matrix update: 1.0loop 30: 0.0(wall time)
211 Eigenvalues computed
Seclr4(Cholesky complete (CPU)) : 17.390 3234.083 Mflops
Seclr4(Transform to eig.problem (CPU)) : 63.800 2644.548 Mflops
Seclr4(Compute eigenvalues (CPU)) : 135.000 1666.391 Mflops
Seclr4(Backtransform (CPU)) : 3.430 745.412 Mflops
TIME HAMILT (CPU) = 81.2, HNS = 43.4, DIAG = 219.9
K= 0.00000 0.00000 0.00000 1
MATRIX SIZE 3481 WEIGHT= 1.00 PGR:
EIGENVALUES ARE:
-0.7692397 -0.7059677 -0.6756879 -0.6738878 -0.6598848
===> TOTAL CPU TIME: 346.1 (INIT = 1.5 + K-POINTS = 344.6)
> SUM OF WALL CLOCK TIMES: 359.5 (INIT = 1.6 + K-POINTS = 357.9)
Maximum WALL clock time: 360.093922019005
Maximum CPU time: 346.340000000000
P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671 FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------
More information about the Wien
mailing list