==================================================================================
Scaling results of SMDEP on two different architectures 
using different compilers. P corresponds to the ammount
of SMDEP that has been parallelized according to Amdahl's
law. 28 sept 2008.
==================================================================================

1000 particles HARPERTOWN: Intel(R) Xeon(R) CPU E5430  @ 2.66GHz icc 9.1 -fast

serial => 127 s
 1 cpu => 280 s 
 2 cpu => 146 s, P=0.95714
 3 cpu =>  99 s, P=0.96964
 4 cpu =>  76 s, P=0.97143
 5 cpu =>  62 s, P=0.97321
 6 cpu =>  53 s, P=0.97286
 7 cpu =>  46 s, P=0.97500
 8 cpu =>  42 s, P=0.97143

P=0.970102

--------------------------------------

1000 particles HARPERTOWN: Intel(R) Xeon(R) CPU E5430  @ 2.66GHz gcc 4.1.2 -O3 -funroll-loops -march=core2

serial => 599 s
 1 cpu => 676 s 
 2 cpu => 360 s, P=0.93491
 4 cpu => 198 s, P=0.94280
 8 cpu => 118 s, P=0.94336

P=0.94036

--------------------------------------

1000 particles HUYGENS: POWER6 (architected), altivec supported @ 4.704GHz gcc 4.2.1 -O3 -funroll-loops -mpowerpc64

serial => 1000 s
 1 cpu => 1026 s
 2 cpu =>  557 s, P=0.91423
 4 cpu =>  308 s, P=0.93307
 8 cpu =>  185 s, P=0.93679  
16 cpu =>  128 s, P=0.93359
32 cpu =>   97 s, P=0.93467
64 cpu =>   90 s, P=0.92676

P=0.92985

--------------------------------------

1000 particles HUYGENS: POWER6 (architected), altivec supported @ 4.704GHz xlc_r 9.0 -qarch=auto -qcache=auto -qtune=auto -O5 -qhot -qnostrict -qmaxmem=-1

serial => 546 s 
 1 cpu => 580 s 
 2 cpu => 295 s, P=0.98276
 3 cpu => 197 s, P=0.99052
 4 cpu => 151 s, P=0.98621
 5 cpu => 122 s, P=0.98707
 6 cpu => 103 s, P=0.98690
 7 cpu =>  89 s, P=0.96749
 8 cpu =>  80 s, P=0.98522
 9 cpu =>  84 s, P=0.96207
10 cpu =>  66 s, P=0.98467
11 cpu =>  61 s, P=0.98431
12 cpu =>  58 s, P=0.98182
13 cpu =>  62 s, P=0.96753
14 cpu =>  58 s, P=0.96923
15 cpu =>  48 s, P=0.98276
16 cpu =>  46 s, P=0.98207
17 cpu =>  43 s, P=0.98373
18 cpu =>  46 s, P=0.97485
19 cpu =>  45 s, P=0.97366
20 cpu =>  39 s, P=0.98185
21 cpu =>  38 s, P=0.98121
22 cpu =>  37 s, P=0.98079
23 cpu =>  37 s, P=0.97876
24 cpu =>  35 s, P=0.98051
28 cpu =>  33 s, P=0.97803
32 cpu =>  33 s, P=0.97353
64 cpu =>  35 s, P=0.95457

P=0.97854