PERFORMANCE METRICS for RS/6000 SYSTEMS

Section One covers SPEC and LINPACK results for most RS/6000 models and is relative to performance of a single processor for each system.

Notebooks

Workstations

Servers

SP Nodes

 

Section Two contains the Commercial performance metrics for most RS/6000 Server systems; in the case of Symmetric Multi-Processor (SMP) systems, performance metrics are shown for the various processor configurations.

Servers

SP Nodes

 

All performance measurements were made with systems running the AIX operating system. Systems are typically tested with the most recent version of AIX which was available at the time the test was performed. Practically all systems shown in these tables were tested with some release of AIX Version 4.

TPM estimates may be used for comparison and positioning with other systems. However care must be taken in how they are referenced in any documents, presentations or other materials. References and associations with the Transaction Processing Performance Council (TPC) are prohibited by TPC. The notation, tpm, must always be used with these estimates. An alternative comparison of performance in a commercial environment is now being used by IBM and is indicated as Relative OLTP (On-Line Transaction Processing) Performance. The performance is compared to that of a model 25S/250 performance for this workload.

Note that these performance measurements were all made with systems running the AIX operating system. Performance measurements are not yet available for systems running the Solaris or Windows NT operating systems.

More Performance Notes

 

Section 1 - SPEC and LINPACK PERFORMANCE

 

Notebooks (non-commercial) [top of page]

Model

Proc

MHz

L1

Cache

(KB)

L2

Cache

(MB)

SPECint95

SPECint_base95

SPECfp95

SPECfp_base95

LINPACK

DP

LINPACK

SP

LINPACK

TPP

820

603

100

16/16

0.25

--

--

--

--

--

--

--

850

603

100

16/16

0.25

--

2.53

--

2.07

--

--

--

860

603e

166

16/16

0.25

3.94

3.62

2.71

2.62

 

 

 

 

 

Workstations (non-commercial) [top of page]

Model

Proc

MHz

L1

Cache

(KB)

L2

Cache

(MB)

SPECint95

SPECint_base95

SPECfp95

SPECfp_base95

LINPACK

DP

LINPACK

SP

LINPACK

TPP

25T

601

66

32

0

1.82

1.69

2.32

2.23

12.7

19.5

33.9

25T

601

80

32

0

--

2.03

--

2.58

15.1

21.8

40.5

41T

601

80

32

0

--

2.03

--

2.58

15.1

21.8

40.5

41T

601

80

32

0.5

--

--

--

--

20.1

24.9

41.9

42W/T

604

120

16/16

0

3.21

2.93

2.74

2.57

--

26.9

42.3

42W/T

604

120

16/16

0.5

4.01

3.75

3.53

3.37

22.4

39.7

60.9

43P-100

604

100

16/16

0.25

3.59

3.36

3.20

3.04

--

27.9

56.4

43P-120

604

120

16/16

0.5

4.24

4.01

3.41

3.23

25.5

48.2

61.0

43P-132

604

133

16/16

0.5

4.72

4.55

3.76

3.59

27.8

51.7

67.5

43P-140

604e

166/

200

32/32

0.5/1

6.19/7.2

5.8/6.9

4.8/5.2

4.6/5.1

22/21.9

66/71

125/145

43P-140

upgrade

604e

233

32/32

1.0

8.66

7.77

5.63

5.30

22.6

77.1

156

43P-140new

604e

233

32/32

1.0

9.24

8.29

5.75

5.48

56.0

113.3

156

43P-140

604e

332

32/32

1.0

12.90

12.20

6.21

5.99

59.9

123.6

179

43P-150

604e

375

32/32

1.0

15.1

14.5

10.1

9.85

64.8

151.2

255.7

43P-240

604e/1

166

32/32

0.5

5.73

5.31

4.75

4.60

47.4

86.3

119

43P-240

604e/1

233

32/32

1.0

8.71

7.80

5.87

5.60

48.5

109.1

145

43P-260

Pow3

200

32/64

4.0

13.2

12.5

30.1

27.6

236.5

248.1

630

F40

604e/1

233

32/32

1.0

8.71

7.80

5.87

5.60

48.5

109.1

145

3AT

Pow2

59

32/64

0

2.84

2.71

7.20

6.81

49.7

64.7

166

3BT

Pow2

67

32/64

0

3.14

3.00

7.50

7.20

52.9

73.0

180

3BT

Pow2

67

32/64

1.0

3.25

3.12

7.71

7.35

55.1

73.0

182

3CT

Pow2

67

32/128

0

--

--

--

--

132.8

72.3

224

3CT

Pow2

67

32/128

2.0

3.31

3.20

9.35

8.75

133.6

72.3

231

397

P2SC

160

32/128

0

8.62

7.77

26.60

23.00

311.9

147.0

528

Servers (non-commercial performance) [top of page]

Model

Proc

MHz

L1

Cache

(KB)

L2

Cache

(MB)

SPECint95

SPECint_base95

SPECfp95

SPECfp_

base95

LINPACK

DP

LINPACK

SP

LINPACK

TPP

25S

601

66

32

0

1.82

1.69

2.32

2.23

12.7

19.5

33.9

25S

601

80

32

0

--

2.03

--

2.58

15.1

21.8

40.5

C10

601

80

32

1

--

2.37

--

2.97

20.3

24.9

42.4

C20

604

120

16/16

1

--

3.85

--

3.50

22.7

40.7

62.7

43P

604

100

16/16

0.25

3.59

3.36

3.20

3.04

--

27.9

56.4

43P

604

120

16/16

0.50

4.24

4.01

3.41

3.23

25.5

48.2

61.0

43P

604

133

16/16

0.50

4.72

4.55

3.76

3.59

27.8

51.7

67.5

43P

604e

166

32/32

0.50

6.19

5.96

5.01

4.83

24.2

71.0

127.7

43P-140

604e

166

32/32

0.50

6.15

5.82

4.83

4.66

22.0

66.3

125.6

43P-140

604e

200

32/32

1.00

7.79

6.99

5.43

5.12

22.1

72.3

145.2

43P-140 upgrade

604e

233

32/32

1.00

8.66

7.77

5.63

5.30

22.6

77.1

156.2

43P-140 new

604e

233

32/32

1.00

9.24

8.29

5.75

5.48

56.0

113.3

156.1

43P-240

604e/1

166

32/32

0.50

5.73

5.31

4.75

4.6

47.4

86.3

119.1

43P-240

604e/1

233

32/32

1.00

8.71

7.80

5.87

5.6

48.5

109.1

145.6

43P-260

Power3/2

200

64/32

4.00

13.20

12.50

30.10

27.6

236.5

--

597.1

E20

604

100

16/16

0.50

3.67

3.43

3.13

3.06

23.5

39.5

62.3

E30

604

133

16/16

0.50

4.74

4.56

3.49

3.34

23.1

46.4

69.0

E30

604e

166

32/32

1.00

6..19

5.93

4.77

4.62

32.8

74.4

124.0

E30

604e

233

32/32

0.50

9.41

8.46

6.01

5.71

59.4

108.0

153.5

F30

604

133

16/16

0.50

4.74

4.56

3.49

3.34

23.1

46.4

69.0

F30

604e

166

32/32

1.00

6..19

5.93

4.77

4.62

32.8

74.4

124.0

F30/F3L

604e

233

32/32

0.50

9.41

8.46

6.01

5.71

59.4

108.0

153.5

F40

604e/1

166

32/32

0.50

5.73

5.31

4.75

4.6

47.4

86.3

119.1

F40

604e/1

233

32/32

1.00

8.71

7.80

5.87

5.6

48.5

109.1

145.6

H10

604e/1

166

32/32

0.50

5.73

5.31

4.75

4.6

47.4

86.3

119.1

H10

604e/1

233

32/32

1.00

8.71

7.80

5.87

5.6

48.5

109.1

145.6

F50

604e/1

166

32/32

0.25

7.52

6.79

8.52

8.11

70.2

98.9

166.4

F50/H50

6043/1

332

32/32

0.25

14.40

14.00

12.60

12.1

115.7

158.5

273.4

390

Pow2

67

32/64

0.00

3.14

3.00

7.50

7.20

52.9

73.0

180.6

390

Pow2

67

32/64

1.00

3.21

3.09

7.52

7.24

55.1

73.0

182.8

39H

Pow2

67

32/128

2.00

3.31

3.20

9.35

8.75

133.6

72.3

231.3

397

P2SC

160

32/128

0

8.62

7.77

26.60

23.0

311.9

147.0

528.0

58H

Pow2

55

32/256

0.00

--

--

--

--

101.1

51.3

197.2

590

Pow2

66

32/256

0.00

3.33

3.19

10.40

9.69

131.8

73.2

237.2

59H

Pow2

66

32/128

1.00

3.31

3.20

9.34

8.74

132.0

73.1

227.1

591

Pow2

77

32/256

0.00

3.84

3.67

12.40

11.2

156.0

92.9

274.4

595

P2SC

135

32/128

0.00

6.17

5.90

17.60

15.4

262.1

124.0

443.2

R20

Pow2

66

32/128

1.00

3.31

3.20

9.34

8.74

132.0

73.1

227.1

R21

Pow2

77

32/256

0.00

3.84

3.67

12.40

11.2

156.0

92.9

274.4

R24

Pow2

71

32/128

2.00

3.53

3.41

9.98

9.33

141.0

78.2

245.3

H70

RS64II

340

64/64

4.0

16.0

13.7

21.2

20.2

187.6

124.0

498.3

S70

RS64

125

 

 

 

 

 

 

 

 

 

 

 

Scalable POWERparallel Models (characteristics of individual nodes) [top of page]

Model

Proc

MHz

L1

Cache

(KB)

L2

Cache

(MB)

SPECint95

SPECint_

base95

SPECfp95

SPECfp_

base95

LINPACK

DP

LINPACK

SP

LINPACK

TPP

Thin

Pow2

67

32/64

1.0

3.31

3.20

9.35

8.75

55.10

73.00

182.80

Thin

Pow2

67

32/128

2.0

3.42

3.29

10.20

9.61

133.60

72.30

231.30

Wide

Pow2

67

32/256

0

3.33

3.19

10.40

9.69

131.80

73.20

237.20

Wide

Pow2

77

32/256

0

3.84

3.67

12.40

11.20

156.00

92.90

274.40

Thin

P2SC

120

32/128

0

5.61

5.36

16.60

14.60

234.90

110.50

405.70

Wide

P2SC

135

32/128

0

6.17

5.90

17.60

15.40

262.10

124.00

443.20

Thin

P2SC

160

32/128

0

8.61

7.06

25.80

22.40

311.90

147.00

528.00

T/W 332

604e

332

32/32

.256

14.4

14.00

12.60

12.10

115.70

158.50

273.40

T/W

Pow3

200

32/64

4

16.0

13.7

21.2

20.2

187.6

124.0

498.3

 

Section 2 - COMMERCIAL PERFORMANCE - Servers [top of page]

Model

Proc

MHz

L1

Cache

(KB)

L2

Cache

(MB)

Rel

OLTP

Perf

tpm

SPECint_rate95

SPECint_ base_rate95

SPECfp_rate95

SPECfp_

base_rate95

SPEC

web96

ops/sec

25S

601

66

32

0

1.0

300

--

--

--

--

--

C10

601

80

32

1

1.8

485*

--

--

--

--

--

C20

604

120

16/16

1

2.6

620

--

--

--

--

--

43P

604

100

16/16

0.25

1.5

450

--

--

--

--

--

43P

604

120

16/16

0.5

1.9

560

--

--

--

--

--

43P

604

133

16/16

0.5

2.1

620

--

--

--

--

--

43P

604e

166

32/32

0.5

2.6

--

--

--

--

--

--

43P-140

604e

166

32/32

0.5

2.9

--

--

--

--

--

--

43P-140

604e

200

32/32

1

3.6

--

--

--

--

--

459

43P-140

upgrade

604e

233

32/32

1

3.7

--

--

--

--

--

--

43P-140new

604e

233

32/32

1

3.9

--

--

--

--

--

--

43P-140

604e

332

32/32

1

5.3

 

 

 

 

 

 

43P-240

604e/1

166

32/32

0.5

2.8

--

51.5

47.7

42.7

41.3

--

43P-240

604e/2

166

32/32

0.5

4.2

--

97.4

90.4

76.5

73.5

--

43P-240

604e/1

233

32/32

1

3.7

--

78

69.9

52

50.1

--

43P-240

604e/2

233

64/64

2

5.2

--

151

132

95.2

89.5

--

43P-260

604e/1

200

64/32

4

10.5

--

116

111

266

243

--

43P-260

604e/2

200

128/64

8

21.0

--

232

222

509

468

2,654

E20

604.00

100

16/16

0.5

2.5

735*

--

--

--

--

--

E30

604.00

133

16/16

0.5

2.8

850

--

--

--

--

--

E30

604e

166

32/32

1

3.7

--

--

--

--

--

--

E30

604e

233

32/32

0.5

4.7

--

--

--

--

--

--

F30

604.00

133

16/16

0.5

2.8

850

--

--

--

--

--

F30

604e

166

32/32

1

3.7

--

--

--

--

--

--

F30

604e

233

32/32

0.5

4.7

--

--

--

--

--

--

F40

604e/1

166

32/32

0.5

2.8

--

51.5

47.7

42.7

41.3

--

F40

604e/2

166

64/64

1

4.2

--

97.4

90.4

76.5

73.5

--

F40

604e/1

233

32/32

1

3.7

--

--

--

--

--

--

F40

604e/2

233

64/64

2

5.2

--

--

--

--

--

--

H10

604e/1

166

32/32

0.5

2.8

--

51.5

47.7

42.7

41.3

--

H10

604e/2

166

64/64

1

4.2

--

97.4

90.4

76.5

73.5

--

F50

604e/1

166

32/32

0.26

7.6

--

67.5

61

76.5

72.8

--

F50

604e/2

166

64/64

0.51

13.7

--

135

121

149

143

--

F50

604e/3

166

96/96

0.77

19*

--

186*

181*

216*

205*

--

F50

604e/4

166

128/128

1

24.4

8142*

237

241

283

267

2,148

F50/H50

604e/1

332

32/32

.2

10.0

 

131

126

113

109

 

F50/H50

604e/2

332

64/64

.4

17.9

 

260

249

217

206

 

F50/H50

604e/3

332

96/96

.6

25.2

 

388

369

307

292

 

F50/H50

604e/4

332

128/128

.8

32.8

9,853*

510

490

387

366

2,755

390

Pow2

67

32/64

1

4.30

900

--

--

--

--

--

39H

Pow2

67

32/128

1

4.40

1000

--

--

--

--

--

397

P2SC

160

32/128

1

6.70

 

 

 

 

 

 

G30

601/2

75

32.00

0.5

4.20

880

40.40

36.90

44.10

43.20

--

G30

601/4

75

64.00

1

7.00

1450

73.00

77.10

81.50

78.80

--

G40

601/1

112

16/16

0.5

2.60

--

33.60

32.20

28.20

26.90

--

G40

604/2

112

32/32

1

4.80

--

66.50

60.60

53.30

50.70

--

G40

604/4

112

64/64

2

8.80

--

122.00

110.00

-

-

--

G40

604e/2

187

32/32

2

6.60

--

129.00

113.00

87.10

80.70

--

G40

604e/4

187

64/64

4

10.90

--

254.00

217.00

154.00

149.00

--

58H

Pow2

55

32/256

0

3.20

960

--

--

--

--

--

590

Pow2

66

32/256

0

3.9

1170

--

--

--

--

--

59H

Pow2

66

32/128

1

4.4

1330

--

--

--

--

--

591

Pow2

77

32/256

0

4.5

1350

--

--

--

--

--

595

P2SC

135

32/128

0

5.8

1750

--

--

--

--

--

J30

601/2

75

32

1

4.7

1400

42.3

39.7

47.4

46.4

--

J30

601/4

75

64

2

7.8

2350

83

78.2

91.3

89.3

--

J30

601/6

75

96

3

10.7

3200

125

117.0

134

133

--

J30

601/8

75

128

4

12.1

3630

162

153.0

172

165

--

J40

604/2

112

16/16

1

5.3

--

71

64.9

57.3

53.4

--

J40

604/4

112

32/32

2

9.8

--

138

129

107

102

--

J40

604/6

112

48/48

3

14.5

--

205

195

159

154

--

J40

604/8

112

64/64

4

19.2

5,774*

258

244

200

189

--

J50

604e/2

200

16/16

2

8.7

--

137

121

92.5

90.3

--

J50

604e/4

200

32/32

4

15.9

--

268

244

183

176

--

J50

604e/6

200

48/48

6

22.6

--

396

343

261

248

--

J50

604e/8

200

64/64

8

28.5

9165*

509

445

332

320

--

R20

Pow2

66

32/128

1

4.4

1330

--

--

--

--

--

R21

Pow2

77

32/256

0

4.5

1350

--

--

--

--

--

R24

Pow2

71

32/128

2

4.9

1,470*

--

--

--

--

--

R30

601/2

75.

32.00

1

4.7

1400

42.3

39.7

47.4

46.4

--

R30

601/4

75

64.00

2

7.8

2350

83

78.2

91.3

89.3

--

R30

601/6

75

96.00

3

10.7

3200

125

117

134

133

--

R30

601/8

75

128.00

4

12.1

3630

162

153

172

165

--

R40

604/2

112

16/16

1

5.3

--

71.9

64.9

57.3

53.4

--

R40

604/4

112

32/32

2

9.8

--

138

129

107

102

--

R40

604/6

112

48/48

3

14.5

--

205

195

159

154

--

R40

604/8

112

64/64

4

19.2

5,774*

258

244

200

189

--

R50

604e/2

200

16/16

2

8.7

--

137

121

92.5

90.3

--

R50

604e/4

200

32/32

4

15.9

--

268

244

183

176

--

R50

604e/6

200

48/48

6

22.6

--

396

343

261

248

--

R50

604e/8

200

64/64

8

28.5

9,165

509

445

332

320

--

H70

RS64II/1

340

64/64

4

16.6

--

144

124

191

182

--

H70

RS64II/2

340

64/64

8

31.7

--

287

247

370

354

6,958

H70

RS64II/3

340

64/64

12

44.2

--

430

370

534

512

--

H70

RS64II/4

340

64/64

16

56.7

--

573

492

674

645

11,774

S70

RS64/4

125

64/64

4

24.2

--

 

 

 

 

 

S70

RS64/8

125

128/128

8

44.0

--

 

 

 

 

 

S70

RS64/12

125

192/192

12

60.5

18,667

 

 

 

 

4,075

S7A

RS64II/4

262

64/64

8

46.0

--

 

 

 

 

 

S7A

RS64II/8

262

128/128

16

82.7

 

 

 

 

 

 

S7A

RS64II/12

262

192/192

24

113.8

34,139

 

 

 

 

20,200

 

Scalable POWERparallel Models (characteristics of individual nodes) [top of page]

Model

Proc

MHz

L1 Cache

(KB)

L2 Cache

(MB)

 

 

Rel OLTP Perf

 

 

tpm

SPEC int_rate95

SPECint_ base_rate95

SPECfp_rate95

SPECfp_ base_rate95

SPEC web96 ops/sec

Thin

Pow2

66

32/128

2.00

4.4

1000

--

--

--

--

 

Thin 120

P2SC

120

32/128

0

5.8

--

--

--

--

--

 

Wide

Pow2

66

32/256

0

3.9

1170

--

--

--

--

 

Wide

Pow2

77

32/256

0

4.5

1350

--

--

--

--

 

Wide 135

P2SC

135

32/128

0

5.8

--

--

--

--

--

 

Thin 160

P2SC

160

32/128

0

6.7

 

 

 

 

 

 

High

604/2

112

16/16

1.00

5.8

--

71.9

64.9

57.3

53.4

 

High

604/4

112

32/32

2

10.0

--

138

129

107

102

 

High

604/6

112

48/48

3

14.5

--

205

195

159

154

 

Thin/Wide 332

604e/2

332

64/64

.512

17.9

 

 

249

 

206

 

High

604/8

112

64/64

4

19.2

5,774

258

244

200

189

 

High

604e/2

200

32/32

2

9.3

 

137

121

92.5

90.3

 

High

604e/4

200

64/64

4

17.0

 

268

244

183

176

 

High

604e/6

200

96/96

6

23.8

 

396

343

261

248

 

Thin/Wide P3

Power3/1

200

64/32

4

10.5

 

 

111

 

243

 

Thin/Wide P3

Power3/2

200

128/64

8

21.0

 

 

222

 

468

 

High

604e/8

200

128/128

8

30.6

 

509

445

332

320

 

Thin/Wide 332

604e/4

332

128/128

1

32.8

 

 

490

 

366

 

 

 

NOTES ON PERFORMANCE VALUES [top of page]

The performance benchmarks and the values shown here were derived using particular, well configured, development-level computer systems. Unless otherwise indicated for a system, the values were derived using 32-bit applications and external cache if external cache is supported on the system. All performance benchmark values are provided "as is" and no warranties or guarantees are stated or implied by IBM. Actual system performance may vary and is dependent upon many factors including system hardware configuration and software design and configuration. Buyers should consult other sources of information to evaluate the performance of systems they are considering buying and should consider conducting application-oriented testing. For additional information about the performance benchmarks, values, and systems tested, please contact your local IBM Branch Office or IBM Authorized Reseller or access the following on the Web:

For more information you can check these websites:

SPEC http://www.specbench.org

TPC http://www.tpc.org

Linpack http://www.netlib.no/netlib/benchmark/performance.ps

Unless otherwise indicated for a system, the performance benchmarks were conducted using AIX V4.2 or V4.3. IBM C for AIX V3.6.4.0 and XL Fortran V6.1.0.0 were the compilers used in the benchmark tests. The preprocessors used in the benchmark tests include KAP 3.2 for Fortran and KAP/C 1.4.2 from Kuck & Associates and VAST-2 V4.01X8 from Pacific-Sierra Research. The preprocessors were purchased separately from these vendors.

SPECint95: SPEC component-level benchmark that measures integer performance. Result is the geometric mean of eight tests that comprise the CINT95 benchmark suite. All of these are written in C language. SPECint_base95 is the result of the same tests in CINT95 with a maximum of four compiler flags that must be used in all eight tests. (Units are relative to SPARCstation 10, which has value of 1.0.)

SPECint_rate95: Geometric average of the eight SPEC rates from the SPEC integer tests (CINT95). SPECint_base_rate95 is result of the same tests as CINT95 with a maximum of four compiler flags that must be used in all eight tests.

SPECfp95: New SPEC component-level benchmark that measures floating point performance. Result is the geometric mean of ten tests, all written in FORTRAN, that are included in the CFP95 benchmark suite. SPECfp_base95 is the result of the same test in CFP95 with a maximum of four compiler flags that must be used in all ten tests.(Units are relative to SPARCstation 10, which has value of 1.0.)

SPECfp_rate95: Geometric average of the ten SPEC rates from SPEC floating-point tests (CFP95). SPECfp_base_rate95 is the result of the same tests as CFP95 with a maximum of four compiler flags that must be used in all ten tests.

LINPACK DP: Double precision, n=100 results with AIX XL FORTRAN compiler, with optimization. Units are megaflops (MFLOPS).

LINPACK SP: Single Precision, n=100 results with AIX XL FORTRAN compiler, with optimization. Units are megaflops (MFLOPS).

LINPACK TPP: Toward Peak Performance (TPP), n=1,000 results with AIX XL FORTRAN compiler, with optimization. Units are megaflops (MFLOPS).

SPECweb96 is the maximum number of HTTP operations per second achieved on the SPECweb96 benchmark without significant degradation of response time. The web server software is IBM HTTP Server 1.3.4 or Zeus V1.1 from Zeus Technology Ltd.

Rel. OLTP Perf - Relative OLTP performance is an estimate of commercial throughput using an IBM analytical model. This model simulates some of the system's operations of the CPU, caches, and memory in an OLTP environment but does not simulate the disk or network I/O operations. Although general database and operating systems parameters are used, the model does not represent specific databases or AIX versions. With these limitations, ROP may be used to compare RS/6000 performance. The model 250 is the reference system and has a value of 1.0.

The following Transaction Processing Council (TPC) benchmarks reflect the performance of the microprocessor, memory subsystem, disk subsystem, and some portions of the network:

"tpmC" - TPC Benchmark C throughput measured as the average number of transactions processed per minute during a valid TPC-C configuration run of at least twenty minutes.

$/tpmC is TPC Benchmark C price-performance ratio where the estimated 5-year total cost of ownership for system hardware, software, and maintenance is divided by the throughput, tpmC. For complete disclosure information please see TPC Web page at http://www.tpc.org.

SPECint95 and SPECfp95 are trademarks of the Standard Performance Evaluation Corporation. KAP is a trademark of Kuck Associates. VAST-2 is a registered trademark of Pacific Sierra Research Corporation. TPC-C is a registered trademark of the Transaction Processing Performance Council.

Notes on Performance Estimates:

Relative OLTP (ROLTP) is an estimate of commercial performance derived from an IBM analytical model. The model simulates some of the system's operations such as CPU, cache, and memory. However, the model does not simulate disk or network I/O operations. Although the model uses general database and operating system parameters, the model does not reflect specific databases or AIX versions or releases. Unless otherwise indicated for a system, the model assumes the use of 32-bit applications. Unless otherwise indicated for a system, ROLTP is estimated only at the time the system is introduced. An IBM RS/6000 Model 250 is the baseline reference system and has a value of 1.0. Although ROLTP may be used to compare estimated RS/6000 commercial processing performance, actual system performance may vary and is dependent upon many factors including system hardware configuration and software design and configuration. All performance benchmark values are provided "AS IS" and no warranties or guarantees are stated or implied by IBM. Buyers should consult other sources of information to evaluate the performance of systems they are considering buying.

[top of page]