CFP2000 Result: AMD Tyan Thunder n4250QE (S4985), Opteron (TM) 2220 (tested by Advanced Micro Devices)

Benchmark	Base Copies	Base Runtime	Base Ratio	Copies	Runtime	Ratio
168.wupwise	4	57.6	129	4	53.8	138
171.swim	4	174	82.7	4	170	84.8
172.mgrid	4	116	72.1	4	111	75.2
173.applu	4	107	91.2	4	93.4	104
177.mesa	4	61.8	105	4	57.2	114
178.galgel	4	92.8	145	4	87.2	154
179.art	4	110	110	4	74.6	162
183.equake	4	90.6	66.6	4	81.8	73.8
187.facerec	4	74.7	118	4	74.4	118
188.ammp	4	136	74.9	4	128	79.6
189.lucas	4	116	80.1	4	110	84.4
191.fma3d	4	130	75.1	4	131	74.5
200.sixtrack	4	117	43.8	4	112	45.7
301.apsi	4	149	80.7	4	142	85.0
SPECfp_rate_base2000	87.2
SPECfp_rate2000	94.3

Benchmark

Base
Copies

Base
Runtime

Base
Ratio

Copies

Runtime

Ratio

168.wupwise

57.6

129

53.8

138

171.swim

174

82.7

170

84.8

172.mgrid

116

72.1

111

75.2

173.applu

107

91.2

93.4

104

177.mesa

61.8

105

57.2

114

178.galgel

92.8

145

87.2

154

179.art

110

74.6

162

183.equake

90.6

66.6

81.8

73.8

187.facerec

74.7

118

74.4

118

188.ammp

136

74.9

128

79.6

189.lucas

116

80.1

110

84.4

191.fma3d

130

75.1

131

74.5

200.sixtrack

117

43.8

112

45.7

301.apsi

149

80.7

142

85.0

SPECfp_rate_base2000

87.2

SPECfp_rate2000

94.3

Hardware

Hardware Vendor:

AMD

Model Name:

Tyan Thunder n4250QE (S4985), Opteron (TM) 2220

CPU:

AMD Opteron (TM) 2220

CPU MHz:

2800

FPU:

Integrated

CPU(s) enabled:

4 cores, 2 chips, 2 cores/chip

CPU(s) orderable:

1,2

Parallel:

Primary Cache:

64KBI + 64KBD on chip per core

Secondary Cache:

1024KB (I+D) on chip per core

L3 Cache:

N/A

Other Cache:

N/A

Memory:

8x1GB, DDR2-667 CL4 ECC Reg Dual rank

Disk Subsystem:

SATA, 74GB

Other Hardware:

None

Software

Operating System:

SuSE Linux Enterprise Server 9 SP3 (64-bit)

Compiler:

PathScale EKOPath(TM) Compiler
Suite, Release 2.3

File System:

Linux/ext3

System State:

Multi-user, run level 3

Notes / Tuning Information
Tested by Advanced Micro Devices +FDO: PASS1= -fb_create fbdata PASS2= -fb_opt fbdata +ACML means -L/pathscale64/lib -lacml, which causes linking with AMD Core Math Library V2.7.0 Baseline optimization C programs: -Ofast -WOPT:mem_opnds=on +FDO Fortran programs: -Ofast -LNO:fusion=2 -OPT:fast_complex=on +FDO Portability Flags: 178.galgel: -fixedform Peak Tuning: 168.wupwise: -Ofast -LNO:prefetch_ahead=5:prefetch=3 -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 -IPA:linear=on:plimit=50000:callee_limit=5000 -INLINE:aggressive=on 171.swim: -Ofast -CG:local_fwd_sched=on -LNO:fusion=2 -m3dnow 172.mgrid: -Ofast -CG:gcm=off -OPT:IEEE_arith=3:unroll_size=200 -LNO:fusion=2:fission=1:blocking=off:prefetch_ahead=2 -WOPT:mem_opnds=on:aggstr=0 173.applu: -Ofast -CG:local_fwd_sched=on -OPT:ro=3 -TENV:X=3 -LNO:fusion=2:fission=2:full_unroll_size=10000 +FDO 177.mesa: -O2 -ipa -OPT:Ofast -fno-math-errno -CG:local_fwd_sched=on -WOPT:mem_opnds=on +FDO 178.galgel: -Ofast -OPT:fast_complex=on +ACML +FDO RM_SOURCES=lapak.f90 179.art: -O3 -OPT:Ofast -fno-math-errno -mno-sse2 -m32 183.equake: -Ofast -CG:load_exe=2 -WOPT:mem_opnds=on -m32 +FDO 187.facerec: -Ofast -LNO:fusion=2 -OPT:fast_complex=on:IEEE_NaN_Inf=off:unroll_size=0 +FDO 188.ammp: -O3 -OPT:alias=disjoint:unroll_times_max=8:Ofast:ro=3 -fno-math-errno -TENV:X=4 +FDO 189.lucas: -Ofast -OPT:ro=3:fast_nint=off:unroll_size=256 -WOPT:mem_opnds=on +FDO 191.fma3d: -O2 -ipa -CG:load_exe=1 -OPT:Ofast:IEEE_arith=3:ro=3 -WOPT:mem_opnds=on:retype_expr=on -IPA:pu_reorder=1 +FDO 200.sixtrack: -O3 -OPT:Ofast:Olimit=6000:early_intrinsics=on -fno-math-errno -CG:load_exe=1 +FDO 301.apsi: -Ofast -CG:load_exe=0 -LNO:prefetch=0:simd=2 System can be built with a 700W Zippy Power Supply All memory slots populated on all CPU(s) taskset utility used to bind CPU(s) to processes.

Notes / Tuning Information

Tested by Advanced Micro Devices
 
 +FDO: PASS1= -fb_create fbdata  PASS2= -fb_opt fbdata
 +ACML  means -L/pathscale64/lib -lacml, 
        which causes linking with AMD Core Math Library V2.7.0
 
 Baseline optimization 
   C programs:       -Ofast -WOPT:mem_opnds=on +FDO 
   Fortran programs: -Ofast -LNO:fusion=2 -OPT:fast_complex=on +FDO
 Portability Flags:
   178.galgel: -fixedform
  
Peak Tuning:
 168.wupwise:  -Ofast -LNO:prefetch_ahead=5:prefetch=3
               -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3
               -IPA:linear=on:plimit=50000:callee_limit=5000 
               -INLINE:aggressive=on
 171.swim:     -Ofast -CG:local_fwd_sched=on -LNO:fusion=2 -m3dnow
 172.mgrid:    -Ofast -CG:gcm=off -OPT:IEEE_arith=3:unroll_size=200
               -LNO:fusion=2:fission=1:blocking=off:prefetch_ahead=2
               -WOPT:mem_opnds=on:aggstr=0
 173.applu:    -Ofast -CG:local_fwd_sched=on -OPT:ro=3 -TENV:X=3
               -LNO:fusion=2:fission=2:full_unroll_size=10000 +FDO
 177.mesa:     -O2 -ipa -OPT:Ofast -fno-math-errno -CG:local_fwd_sched=on -WOPT:mem_opnds=on +FDO
 178.galgel:   -Ofast -OPT:fast_complex=on +ACML +FDO
                RM_SOURCES=lapak.f90
 179.art:      -O3 -OPT:Ofast -fno-math-errno -mno-sse2 -m32
 183.equake:   -Ofast -CG:load_exe=2 -WOPT:mem_opnds=on -m32 +FDO
 187.facerec:  -Ofast -LNO:fusion=2
               -OPT:fast_complex=on:IEEE_NaN_Inf=off:unroll_size=0 +FDO
 188.ammp:     -O3 -OPT:alias=disjoint:unroll_times_max=8:Ofast:ro=3
               -fno-math-errno -TENV:X=4 +FDO
 189.lucas:    -Ofast -OPT:ro=3:fast_nint=off:unroll_size=256 -WOPT:mem_opnds=on +FDO
 191.fma3d:    -O2 -ipa -CG:load_exe=1 -OPT:Ofast:IEEE_arith=3:ro=3
               -WOPT:mem_opnds=on:retype_expr=on -IPA:pu_reorder=1 +FDO 
 200.sixtrack: -O3 -OPT:Ofast:Olimit=6000:early_intrinsics=on
               -fno-math-errno -CG:load_exe=1 +FDO
 301.apsi:     -Ofast -CG:load_exe=0 -LNO:prefetch=0:simd=2
   
 System can be built with a 700W Zippy Power Supply
 All memory slots populated on all CPU(s)
 taskset utility used to bind CPU(s) to processes.

First published at SPEC.org on 06-Feb-2007

Generated on Fri Feb 9 18:30:21 2007 by SPEC CPU2000 HTML formatter v1.01