Table 7. The CPU time that each test job consumes on the Intel Pentium 4, Alpha and SGI systems. The unit of the CPU time is in minutes; the numbers in the parentheses are the step numbers required in the geometry optimization calculations. (a)-tpp7 -axW (b) -tpp7 -axW -ipo -ipo_obj

System

P41.8G

Alpha500

Alpha667

SGI

Compiler

pgf77

pgf77

pgf77

pgf77

ifc

ifc

ifc

ifc

ifc

ifc

ifc

ifc

f90

f90

f77

Arch. opt. a

-tp p6

-tp p6

-tp piv

-tp piv

-tpp6

-tpp6

-tpp7

-tpp7

(a)

(a)

(b)

(b)

-tune ev6

-tune ev6

-mips4 -r10000

Library

BLAS

ATLAS

BLAS

ATLAS

BLAS

ATLAS

BLAS

ATLAS

BLAS

ATLAS

BLAS

ATLAS

DXML

DXML

System blas.a

322-1

4.25

4.46

4.23

4.55

3.39

3.69

3.38

3.65

3.21

3.56

3.22

3.53

7.87

6.37

5.47

322-2

127

130

127

128

77

77

78

78

75

75

74

74

154

136

158

322-3

192

193

192

196

127

127

126

126

127

127

125

125

225

202

265

338

31

31

31

31

23

22

22

22

22

22

22

21

36

32

42

339

35

35

35

35

25

25

25

25

24

24

24

24

43

39

53

364 f

44

44

44

44

- b

- b

- b

- b

30

31

30

31

48

39

53

397

487

487

485

487

363

366

363

362

341

339

339

338

658

602

737

415-1

19

17

19

17

16

14

16

14

15

14

15

13

24

21

33

415-2 f

- c

- c

- c

- c

- c

- c

- c

- c

- c

- c

- c

- c

52

46

71

420-1

91

88

92

90

74

69

73

69

71

68

71

68

113

107

156

420-2

92

88

92

88

74

70

73

69

72

73

72

69

136

128

188

424-1

61

49

61

49

52

41

52

41

47

39

47

39

76

65

98

424-2

61

49

61

49

52

41

52

40

47

39

47

40

75

65

98

438

44

34

43

32

39

27

38

27

27

26

28

26

75

64

58

439-1 f

188(4)

109(2)

185(4)

109(2)

72(2)

71(2)

71(2)

71(2)

69(2)

- d

68(2)

- d

123(2)

111(2)

156(2)

439-2

79

79

79

80

55

55

54

55

47

51 e

47

51 e

90

81

121

447-1

171

161

171

161

129

120

127

118

122

116

122

114

214

191

215

447-2

94

80

93

79

77

63

77

63

65

59

65

58

121

104

153

559-4

16

16

16

16

11

12

11

11

11

10

11

10

35

31

155

560-4

19

19

19

19

13

13

13

13

12

12

13

12

40

36

152

561-4

54

54

54

54

38

38

38

38

35

35

35

35

90

95

201

Sum

1677

1614

1674

1616

1248

1184

1241

1175

1163

1133

1160

1121

2213

2005

2888

a The optimization to ifc other than the above architectural optimization is "-unroll -O2".

b Unknown address error reported by ifc.

c Error due to failure to write scratch files larger than 2 GB.  

d Geometrical optimization did not converge.

e Restarted by a converged structure. 

f  Excluded from sum.