[Wien] Benchmark on Xeon 5140, SMP 4 cores: 2chips, 2cores/chips

[Florent Boucher]

Laurence,
I did what you suggest.
FOPT =  -FR -mp1 -w -prec_div -pc80 -pad -ip -axT

I got the following for the complex version: (x lapw1 -c -p, matrix size 
3481)
1 k-point  : 132.4 s
2 k-points : 138.4 s (efficiency 95.6%) (137.7, 139.0)
3 k-points : 158.1 s (efficiency 83.7%) (144.9,163.4,166.0)
4 k-points : 181.4 s (efficiency 73.0%) (179.7,180.8,182.4,182.6)

On another test for the real version (x lapw1 -p, matrix size 3620)
1 k-point  : 70.6 s
2 k-points : 75.5 s (efficiency 93.5%) (75.2, 75.7)
3 k-points : 88.4 s (efficiency 79.9%) (77.5, 93.3, 94.2)
4 k-points :108.2 s (efficiency 65.2%) (108.0,108.1,108.3,108.5)


Regards
Florent

-- 
 -------------------------------------------------------------------------
| Florent BOUCHER                    |                                    |
| Institut des Matériaux Jean Rouxel | Mailto:Florent.Boucher at cnrs-imn.fr |
| 2, rue de la Houssinière           | Phone: (33) 2 40 37 39 24          |
| BP 32229                           | Fax:   (33) 2 40 37 39 95          |
| 44322 NANTES CEDEX 3 (FRANCE)      | http://www.cnrs-imn.fr             |
 -------------------------------------------------------------------------