SPEC OMPM2001 Summary IBM Corporation IBM eServer pSeries 690 (1900 MHz, 32 CPU) Thu Feb 19 01:22:11 2004 SPEC License #HPG0005 Tester: IBM Corporation Test date: Feb-2004 Test Site: IBM, Austin TX Hardware availability: Mar-2004 Software availability: Oct-2003 Base Base Base Peak Peak Peak Benchmarks Ref Time Run Time Ratio Ref Time Run Time Ratio ------------- -------- -------- -------- -------- -------- -------- 310.wupwise_m 6000 102 58991 6000 102 58991 310.wupwise_m 6000 103 58381 6000 103 58381 310.wupwise_m 6000 102 58713* 6000 102 58713* 312.swim_m 6000 149 40186 6000 118 50957 312.swim_m 6000 151 39660* 6000 118 50959* 312.swim_m 6000 152 39561 6000 118 51021 314.mgrid_m 7300 197 37069 7300 197 37069 314.mgrid_m 7300 197 37090* 7300 197 37090* 314.mgrid_m 7300 197 37132 7300 197 37132 316.applu_m 4000 96.9 41262* 4000 77.9 51363* 316.applu_m 4000 96.4 41479 4000 78.3 51065 316.applu_m 4000 97.4 41059 4000 77.6 51531 318.galgel_m 5100 288 17690 5100 191 26737 318.galgel_m 5100 295 17285 5100 194 26311* 318.galgel_m 5100 288 17678* 5100 196 26020 320.equake_m 2600 87.6 29695 2600 77.9 33355 320.equake_m 2600 87.2 29831* 2600 77.7 33463 320.equake_m 2600 86.6 30029 2600 77.8 33426* 324.apsi_m 3400 61.5 55291 3400 61.5 55291 324.apsi_m 3400 61.4 55333* 3400 61.4 55333* 324.apsi_m 3400 61.4 55411 3400 61.4 55411 326.gafort_m 8700 156 55930 8700 153 56902* 326.gafort_m 8700 156 55852 8700 153 56978 326.gafort_m 8700 156 55882* 8700 155 56251 328.fma3d_m 4600 160 28707 4600 160 28707 328.fma3d_m 4600 161 28576 4600 161 28576 328.fma3d_m 4600 161 28660* 4600 161 28660* 330.art_m 6400 73.8 86678 6400 50.7 126253* 330.art_m 6400 74.5 85903* 6400 49.6 129056 330.art_m 6400 74.8 85563 6400 52.5 121939 332.ammp_m 7000 387 18067* 7000 360 19446 332.ammp_m 7000 388 18035 7000 357 19600 332.ammp_m 7000 387 18081 7000 359 19477* ======================================================================== 310.wupwise_m 6000 102 58713* 6000 102 58713* 312.swim_m 6000 151 39660* 6000 118 50959* 314.mgrid_m 7300 197 37090* 7300 197 37090* 316.applu_m 4000 96.9 41262* 4000 77.9 51363* 318.galgel_m 5100 288 17678* 5100 194 26311* 320.equake_m 2600 87.2 29831* 2600 77.8 33426* 324.apsi_m 3400 61.4 55333* 3400 61.4 55333* 326.gafort_m 8700 156 55882* 8700 153 56902* 328.fma3d_m 4600 161 28660* 4600 161 28660* 330.art_m 6400 74.5 85903* 6400 50.7 126253* 332.ammp_m 7000 387 18067* 7000 359 19477* SPECompMbase2001 38278 SPECompMpeak2001 43708 HARDWARE -------- Hardware Vendor: IBM Corporation Model Name: IBM eServer pSeries 690 (1900 MHz, 32 CPU) CPU: POWER4+ CPU MHz: 1900 FPU: Integrated CPU(s) enabled: 32 cores, 16 chips, 2 cores/chip, 4 chips/MCM CPU(s) orderable: 1,2,3,4 (order # of MCMs) Primary Cache: 64KBI+32KBD (on chip) per core Secondary Cache: 1536KB unified (on chip) per chip L3 Cache: 32MB unified (off-chip) per MCM, 4 MCMs in SUT Other Cache: None Memory: 128 GB Disk Subsystem: 1x36GB SCSI, 10K RPM Other Hardware: None SOFTWARE -------- OpenMP Threads: 32 Parallel: OpenMP Operating System: AIX 5L V5.2 Compiler: IBM C for AIX, Version 6.0 XL FORTRAN for AIX, Version 8.1.1.0 Other Software: ESSL 4.1.0.0, PESSL 2.3.0.2, MASS 3.2.1 File System: AIX/JFS2 System State: Multi-user NOTES ----- Tested by IBM Corporation Portability Flags & Environment Variables -bmaxdata:0x80000000 used in all cases except 330.art_m -qfixed used in: 310.wupwise_m, 312.swim_m, 314.mgrid_m, 316.applu_m, 324.apsi_m -qfixed=80 used in: 318.galgel_m -qsuffix=f=f90 used in: 318.galgel_m, 326.gafort_m, 328.fma3d_m -bmaxdata:0xD0000000 used in 330.art_m (for base and peak) Base Flags C: -q64 -O5 -qalign=natural -qipa=partition=large -qmaxmem=-1 -qsmp=omp -w FORTRAN:-O5 -qalign=natural -qipa=partition=large -qmaxmem=-1 -qsmp=omp -w Base & Peak User Environment: ENV_OMP_NUM_THREADS=32 ENV_OMP_DYNAMIC=FALSE ENV_XLSMPOPTS=SPINS=0:YIELDS=0:STACK=8000000:SCHEDULE=STATIC ENV_MALLOCMULTIHEAP=1 Peak Flags -qsmp=omp used in all cases 310.wupwise_m: basepeak=1 312.swim_m: -O3 -qalign=natural -qtune=pwr4 -qarch=pwr4 314.mgrid_m: basepeak=1 316.applu_m: -O3 -qalign=natural -qtune=pwr4 -qarch=pwr4 318.galgel_m: -O5 -qalign=natural -qhot=arraypad -qessl -qlibessl -lesslsmp -lpesslsmp -lblacssmp -lmass -qipa=partition=large -qmaxmem=-1 ENV_OMP_NUM_THREADS=16 320.equake_m: fdpr -v -R3 -O5 -qalign=natural -qhot=arraypad -Q -lmass 324.apsi_m: basepeak=1 326.gafort_m: -O5 -qalign=natural -qrsqrt -qhot=arraypad -qipa=partition=large -qmaxmem=-1 328.fma3d_m: basepeak=1 330.art_m: fdpr -v -R3 -q64 -O5 -qlargepage -blpdata -qhot=arraypad -Q -lmass EXTRA_CFLAGS= -DINTS_PER_CACHELINE=32 -DDBLS_PER_CACHELINE=16 EXTRA_LDFLAGS=-bmaxdata:0xD0000000 -blpdata 332.ammp_m: -O5 -qalign=natural -qhot=arraypad -Q -lmass Peak sources: SPEC OMPL2001 source for 32bit systems modified for SPEC OMPM2001 used with 312.swim_m, 316.applu_m, 320.equake_m, 326.gafort_m. MCM: Acronym for "Multi-Chip Module" SUT: Acronym for "System Under Test" L3 Cache: 32MB x 4 = 128MB, shared by all processors C: IBM VAC++ invoked as xlc_r Fortran 90: IBM XL Fortran for AIX invoked as xlf90_r Fixes APAR IY48265 were applied to AIX. Large page mode and memory affinity were set as follows: vmo -r -o lgpg_regions=4096 -o lgpg_size=16777216 -o memory_affinity=1 chuser capabilities=CAP_BYPASS_RAC_VMM,CAP_PROPAGATE $USER shutdown -r export MEMORY_AFFINITY=MCM ----------------------------------------------------------------------------- For questions about this result, please contact the tester. For other inquiries, please contact webmaster@spec.org. Copyright 1999-2002 Standard Performance Evaluation Corporation Generated on Wed Mar 17 13:22:58 2004 by SPEC OMP2001 ASCII formatter v2.1