CFP2000 Result Copyright © 1999-2005 Standard Performance Evaluation Corporation |
Sun Microsystems Sun Fire X4200 |
|
SPEC license # 6 | Tested by: Sun Microsystems, Santa Clara | Test date: Mar-2006 | Hardware Avail: May-2006 | Software Avail: Mar-2006 |
Benchmark | Base Copies |
Base Runtime |
Base Ratio |
Copies | Runtime | Ratio | |
---|---|---|---|---|---|---|---|
168.wupwise | 2 | 51.0 | 72.7 | 2 | 48.8 | 76.1 | |
171.swim | 2 | 119 | 60.4 | 2 | 115 | 62.3 | |
172.mgrid | 2 | 91.3 | 45.7 | 2 | 87.8 | 47.6 | |
173.applu | 2 | 124 | 39.1 | 2 | 83.3 | 58.5 | |
177.mesa | 2 | 71.1 | 45.7 | 2 | 53.9 | 60.3 | |
178.galgel | 2 | 82.3 | 81.8 | 2 | 74.5 | 90.3 | |
179.art | 2 | 61.1 | 98.7 | 2 | 56.1 | 108 | |
183.equake | 2 | 61.9 | 48.7 | 2 | 62.2 | 48.5 | |
187.facerec | 2 | 67.8 | 65.0 | 2 | 67.8 | 65.0 | |
188.ammp | 2 | 133 | 38.5 | 2 | 122 | 41.9 | |
189.lucas | 2 | 106 | 43.6 | 2 | 83.7 | 55.4 | |
191.fma3d | 2 | 112 | 43.6 | 2 | 112 | 43.6 | |
200.sixtrack | 2 | 119 | 21.5 | 2 | 106 | 24.2 | |
301.apsi | 2 | 143 | 42.2 | 2 | 134 | 45.1 | |
SPECfp_rate_base2000 | 50.0 | ||||||
SPECfp_rate2000 | 55.6 |
|
|
Notes / Tuning Information |
---|
Portability flags: 178.galgel (pgf90) : -Mfixed Feedback Optimization +FDO: PGI : PASS1=-Mpfi PASS2=-Mpfo PathSale: PASS1=-fb_create fbdata PASS2=-fb_opt fbdata Baseline Optimization Flags: C programs : pgcc -fastsse -Mipa=fast,inline +FDO Fortran programs: pgf90 -fastsse -Mipa=fast,inline +FDO Peak Tuning Flags: 168.wupwise: pathf95 -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 -CG:local_fwd_sched=on -m3dnow 171.swim: pathf95 -Ofast -CG:local_fwd_sched=on -LNO:fusion=2 -m3dnow 172.mgrid: pathf95 -Ofast -CG:gcm=off -OPT:IEEE_a=3:unroll_size=200 -LNO:fusion=2:fission=1:blocking=off:prefetch_ahead=2 -WOPT:mem_opnds=on:aggstr=0 173.applu: pathf95 -Ofast -CG:local_fwd_sched=on -OPT:ro=3 -TENV:X=3 -LNO:fusion=2:fission=2:full_unroll_size=10000:prefetch=3 +FDO 177.mesa: pathf95 -O2 -ipa -OPT:Ofast -fno-math-errno -CG:local_fwd_sched=on -WOPT:mem_opnds=on +FDO 178.galgel: pgf90 -fastsse -O4 -Mipa=fast,inline ONESTEP=yes +FDO RM_SOURCES=lapak.f90 EXTRALIBS=-lacml 179.art: pgcc -fastsse -Munroll=n:9 -Mipa=fast,inline -tp k8-32 183.equake: pgcc -fastsse -Mflushz -Mnovect -Mipa=fast,inline ONESTEP=yes +FDO 187.facerec: pgf90 basepeak=1 188.ammp: pathcc -O3 -OPT:alias=disjoint:unroll_times_max=8:Ofast:ro=3 -fno-math-errno -TENV:X=4 +FDO 189.lucas: pathf95 -O3 -OPT:ro=3:fast_nint=off:unroll_size=256 -WOPT:mem_opnds=on +FDO 191.fma3d: pgf90 basepeak=1 200.sixtrack: pathf95 -O3 -OPT:Ofast:Olimit=6000:early_intrinsics=on -fno-math-errno -CG:load_exe=1 +FDO 301.apsi: pathf95 -Ofast -CG:load_exe=0 -LNO:prefetch=0:simd=2 taskset has been used to bind processes to CPUs Default BIOS settings was used. This result was measured on the Sun Fire X4200. In addition, Sun has submitted the same result for the Sun Fire X4100, which is electronically equivalent to the Sun Fire X4200. |