Thumbnail-Übersicht
  nec_09mar2004...                                                                                                                                                                                                
 
Seite
 
Seite
 
Vectorization...
 
ex. 1: v = s +...
 
Seite
 
Chaining
 
ex. 1: v = s +...
 
ex. 1: v = s +...
 
ex. 1: v = s +...
 
ex. 2: v = v +...
 
Seite
 
ex. 3: v = v +...
 
Seite
 
ex. 3: v = v +...
 
Seite
 
ex. 4: s = s +...
 
ex. 4: s = s +...
 
ex. 5: matrix...
 
ex. s1: matrix...
 
Levels of Para...
 
Seite
 
Basic Rules to...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
Basic Rules fo...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
constant if
 
boundary if
 
ex. s1: matrix...
 
IO optimization
 
Basic Rules fo...
 
IO optimization
 
Basic Rules fo...
 
Seite
 
FORTRAN 90 !cd...
 
FORTRAN 90 !cd...
 
FORTRAN 90 !cd...
 
FORTRAN 90 !cd...
 
FORTRAN 90 !cd...
 
Stride for 1d-...
 
Bank busy time...
 
Stride for 1d-...
 
Bank busy time...
 
Stride for 1d-...
 
FORTRAN 90 !cd...
 
Stride for 1d-...
 
Bank busy time...
 
Seite
 
loop interchan...
 
loop interchan...
 
loop expansion
 
loop expansion
 
loop expansion
 
loop expansion
 
loop expansion
 
loop division
 
loop expansion
 
loop division
 
loop division
 
loop division
 
call to functi...
 
vectorization...
 
vectorization...
 
vectorization...
 
vectorization...
 
Seite
 
2D recursion
 
2D recursion...
 
Seite
 
Seite
 
Basic Rules fo...
 
Seite
 
Seite Seite Vectorization examples ex. 1: v = s + v (cont.) Seite Chaining ex. 1: v = s + v (cont.) ex. 1: v = s + v (cont.) ex. 1: v = s + v ex. 2: v = v + v Seite ex. 3: v = v + s * v Seite ex. 3: v = v + s * v Seite ex. 4: s = s + v * v ex. 4: s = s + v * v (cont.) ex. 5: matrix multiply ex. s1: matrix multiply Levels of Parallelism Seite Basic Rules to achieve Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance: Concurrency Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance Basic Rules for Performance: Pointers Basic Rules for Performance vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks vectorization of if-blocks constant if boundary if ex. s1: matrix multiply IO optimization Basic Rules for Performance IO optimization Basic Rules for Performance Seite FORTRAN 90 !cdir FORTRAN 90 !cdir FORTRAN 90 !cdir FORTRAN 90 !cdir FORTRAN 90 !cdir Stride for 1d-Arrays Bank busy time measured Stride for 1d-Arrays Bank busy time measured Stride for 1d-Arrays FORTRAN 90 !cdir Stride for 1d-Arrays Bank busy time measured Seite loop interchange loop interchange loop expansion loop expansion loop expansion loop expansion loop expansion loop division loop expansion loop division loop division loop division call to function vectorization example vectorization example (2) vectorization example (3) vectorization example (4) Seite 2D recursion 2D recursion (2) Seite Seite Basic Rules for Performance Seite
 
0:00
 
0:09
 
0:22
 
0:56
 
2:44
 
2:55
 
3:20
 
3:49
 
4:41
 
5:37
 
6:18
 
6:55
 
6:58
 
7:02
 
7:45
 
8:25
 
9:56
 
10:42
 
12:49
 
13:30
 
14:29
 
14:39
 
16:20
 
18:04
 
19:17
 
19:18
 
19:37
 
20:23
 
25:02
 
27:03
 
28:20
 
29:10
 
31:27
 
31:41
 
32:46
 
33:22
 
33:53
 
35:36
 
36:29
 
38:57
 
39:43
 
41:07
 
41:44
 
42:30
 
43:01
 
43:02
 
43:28
 
43:29
 
44:01
 
45:16
 
46:12
 
47:25
 
47:29
 
47:41
 
48:45
 
48:46
 
48:56
 
49:34
 
49:35
 
50:39
 
51:37
 
51:41
 
51:44
 
53:06
 
55:12
 
55:44
 
55:59
 
56:49
 
57:05
 
57:08
 
57:08
 
57:41
 
57:59
 
58:54
 
59:25
 
1:00:30
 
1:01:53
 
1:02:01
 
1:02:15
 
1:03:03
 
1:03:26
 
1:03:28
 
1:04:48
 
1:05:36
 
1:06:43
 
1:07:56
 
1:09:34
 
1:09:57
 
1:10:20
 
1:10:31
 
1:11:11
 
1:12:12
 
1:13:19
 
1:14:08
 
1:14:56
 
1:15:04