Table 4. CPU Time Consumption (in Minutes) of Each Test Job by the AMD Opteron System and IBM P690.
|
System |
K8-32 a |
P690 |
|||||||||||
|
Compiler |
ifc 7.1 |
ifc 7.1 |
ifc 7.1 |
pgf77 5.0-1 |
pgf77 5.0-1 |
pgf90 5.0-1 |
pgf77 5.0-1 |
pgf90 5.0-1 |
pgf77 5.0-1 |
pgf90 5.0-1 |
xlf 8.0 |
||
|
Arch. opt.b |
(a) |
(a) |
(a) |
(b) |
(b) |
(b) |
(c) |
(c) |
(d) |
(d) |
g98 default |
||
|
Library |
ATLAS 3.4.7 |
MKL 6.0 |
ACML 1.0 |
BLAS |
ATLAS 3.4.7 |
ACML 1.0 |
ATLAS 3.4.7 |
ACML 1.0 |
ATLAS 3.4.7 |
ACML 1.0 |
ESSL |
||
|
322-1 |
2.10 |
1.92 |
2.45 |
2.66 |
2.72 |
3.44 |
2.36 |
3.1 |
2.36 |
3.1 |
2.08 |
||
|
322-2 |
49 |
49 |
49 |
72 |
72 |
70 |
62 |
60 |
62 |
60 |
38 |
||
|
322-3 |
76 |
74 |
74 |
124 |
125 |
123 |
94 |
93 |
94 |
93 |
90 |
||
|
338 |
12 |
12 |
12 |
20 |
19 |
19 |
15 |
15 |
15 |
15 |
12 |
||
|
339 |
14 |
14 |
14 |
23 |
22 |
23 |
17 |
18 |
17 |
18 |
15 |
||
|
364 |
17 |
17 |
17 |
24 |
24 |
25 |
18 |
19 |
18 d |
19 d |
15 |
||
|
397 |
196 |
196 |
197 |
297 |
297 |
296 |
232 |
231 |
232 |
231 |
246 |
||
|
415-1 |
9 |
10 |
10 |
14 |
13 |
14 |
11 |
12 |
11 |
12 |
9 |
||
|
415-2 |
20 |
21 |
22 |
30 |
26 |
29 |
22 |
25 |
23 |
25 |
17 |
||
|
420-1 |
43 |
43 |
43 |
64 |
62 |
61 |
49 |
49 |
49 |
49 |
47 |
||
|
420-2 |
43 |
43 |
43 |
64 |
62 |
61 |
49 |
49 |
49 |
49 |
56 |
||
|
424-1 |
27 |
27 |
29 |
52 |
33 |
39 |
30 |
36 |
30 |
35 |
25 |
||
|
424-2 |
26 |
27 |
29 |
52 |
33 |
39 |
30 |
36 |
30 |
35 |
25 |
||
|
438 |
18 |
18 |
22 |
36 |
22 |
38 |
19 |
37 |
19 |
36 |
18 |
||
|
439-1 c |
39(2) |
39(2) |
39(2) |
68(2) |
64(2) |
64(2) |
89(4) |
48(2) |
89(4) |
49(2) |
91(4) |
||
|
439-2 |
30 |
31 |
30 |
49 |
49 |
49 |
38 |
33 |
38 |
33 |
39 |
||
|
447-1 |
64 |
64 |
65 |
109 |
98 |
102 |
74 |
77 |
74 |
78 |
72 |
||
|
447-2 |
40 |
41 |
41 |
74 |
60 |
62 |
47 |
49 |
47 |
49 |
38 |
||
|
559-4 |
8 |
9 |
8 |
8 |
8 |
9 |
8 |
8 |
8 |
8 |
13 |
||
|
560-4 |
9 |
10 |
9 |
9 |
9 |
10 |
9 d |
9 d |
9 |
9 |
15 |
||
|
561-4 |
25 |
25 |
25 |
30 |
29 |
31 |
25 |
27 |
25 |
27 |
38 |
||
|
Sum |
728 |
733 |
741 |
1154 |
1066 |
1103 |
851 d |
886 d |
852 d |
884 d |
830 |
||
a The 32-bit compilation under 64-bit operating system. b The global options of ifc is set to "-unroll -O2", plus: (a) "-tpp7 -axW -ipo -ipo_obj". The global optimization to pgf77/pgf90 is set to "-Munroll -O2", plus: (b) "-tp k8-32 -Mvect=cachesize:1048576"; (c) "-tp k8-32 -Mvect=cachesize:1048576 -fastsse"; (d) "-fastsse -time -Mreentrant -Mrecursive -Mnosave -Minfo -Mneginfo -Mscalarsse -Mvect=assoc,recog,prefetch,sse,cachesize:1048576". c The numbers in the parentheses are the step numbers required in the geometrical optimization calculations; this CPU time is excluded from sum. d Estimated.