SPEC CINT2000 Summary IBM Corporation IBM eServer p5 595 (1900 MHz, 64 CPU) Mon Oct 11 21:50:32 2004 SPEC License #11 Test date: Oct-2004 Hardware availability: Nov-2004 Tester: IBM Software availability: Nov-2004 Base Base Base Peak Peak Peak Benchmarks # Copies Run Time Rate # Copies Run Time Rate ------------ -------- -------- -------- -------- -------- -------- 164.gzip 128 250 832* 128 249 833 164.gzip 128 253 822 128 246 846 164.gzip 128 249 834 128 248 839* 175.vpr 128 238 873* 128 235 885 175.vpr 128 235 884 128 232 896 175.vpr 128 239 869 128 234 890* 176.gcc 128 201 811 128 128 1281 176.gcc 128 173 946* 128 130 1252 176.gcc 128 163 1001 128 128 1279* 181.mcf 128 204 1311 128 187 1428 181.mcf 128 193 1384* 128 185 1449 181.mcf 128 193 1386 128 187 1431* 186.crafty 128 168 884* 128 124 1193 186.crafty 128 168 883 128 127 1173 186.crafty 128 163 910 128 125 1191* 197.parser 128 245 1090 128 239 1117 197.parser 128 245 1091* 128 242 1106 197.parser 128 243 1099 128 240 1111* 252.eon 128 150 1291 128 129 1495 252.eon 128 146 1325 128 131 1468* 252.eon 128 149 1296* 128 132 1457 253.perlbmk 128 321 832 128 269 992 253.perlbmk 128 328 816* 128 272 983 253.perlbmk 128 329 811 128 272 984* 254.gap 128 148 1106* 128 166 985 254.gap 128 155 1053 128 161 1017* 254.gap 128 147 1113 128 160 1021 255.vortex 128 181 1559 128 190 1482* 255.vortex 128 196 1440 128 191 1480 255.vortex 128 190 1483* 128 187 1509 256.bzip2 128 185 1204 128 180 1235 256.bzip2 128 186 1195* 128 184 1213* 256.bzip2 128 188 1183 128 190 1172 300.twolf 128 407 1093 128 407 1095 300.twolf 128 408 1092* 128 410 1088* 300.twolf 128 410 1087 128 416 1070 ======================================================================== 164.gzip 128 250 832* 128 248 839* 175.vpr 128 238 873* 128 234 890* 176.gcc 128 173 946* 128 128 1279* 181.mcf 128 193 1384* 128 187 1431* 186.crafty 128 168 884* 128 125 1191* 197.parser 128 245 1091* 128 240 1111* 252.eon 128 149 1296* 128 131 1468* 253.perlbmk 128 328 816* 128 272 984* 254.gap 128 148 1106* 128 161 1017* 255.vortex 128 190 1483* 128 190 1482* 256.bzip2 128 186 1195* 128 184 1213* 300.twolf 128 408 1092* 128 410 1088* SPECint_rate_base2000 1063 SPECint_rate2000 1147 HARDWARE -------- Hardware Vendor: IBM Corporation Model Name: IBM eServer p5 595 (1900 MHz, 64 CPU) CPU: POWER5 CPU MHz: 1900 FPU: Integrated CPU(s) enabled: 64 cores, 32 chips, 2 cores/chip (SMT on) CPU(s) orderable: 16,24,32,40,48,56,64 Parallel: No Primary Cache: 64KBI+32KBD (on chip)/core Secondary Cache: 1920KB unified (on chip)/chip L3 Cache: 36MB unified (off-chip)/chip, 4 chips/MCM, 8 MCMs/SUT Other Cache: None Memory: 256 GB DDR1 Disk Subsystem: 3x36GB SCSI, 15K RPM Other Hardware: None SOFTWARE -------- Operating System: AIX 5L V5.3 Compiler: XL C/C++ Enterprise Edition V7.0 for AIX File System: AIX/JFS2 System State: Multi-user NOTES ----- Tested by IBM Portability Flags: 176.gcc: EXTRA_CFLAGS=-ma -DHOST_WORDS_BIG_ENDIAN 186.crafty: EXTRA_CFLAGS=-DAIX 252.eon: EXTRA_LDFLAGS=-I. -DNDEBUG 253.perlbmk: EXTRA_CFLAGS=-DSPEC_CPU2000_AIX 254.gap: EXTRA_CFLAGS=-DSYS_IS_BSD -DSYS_STRING_H -DSYS_HAS_TIME_PROTO -DSYS_HAS_MALLOC_PROTO -DSYS_HAS_CALLOC_PROTO 300.twolf: EXTRA_CFLAGS=-DHAVE_SIGNED_CHAR Base Optimization Flags: C: -qpdf1/pdf2 -O5 -blpdata -D_ILS_MACROS C++: -qpdf1/pdf2 -O5 -lhmu -qalign=natural Alternate Sources for Base & Peak: Approved alternate-source file 252.eon.fmax_errno.src.alt.tar.gz was used with 252.eon for POSIX-compatibility. Peak Optimization Flags: 164.gzip: -qpdf1/pdf2 -O5 -blpdata -D_ILS_MACROS -qfdpr fdpr -R3 175.vpr: -qpdf1/pdf2 -O5 -blpdata -qalign=natural -qhot=arraypad -Q 176.gcc: -qpdf1/pdf2 -O5 181.mcf: -O5 -blpdata -qfdpr -D_ILS_MACROS fdpr -R3 186.crafty: -qpdf1/pdf2 -O4 -q64 -qfdpr -qarch=pwr3 -qtune=pwr3 -D_ILS_MACROS fdpr -R3 197.parser: -qpdf1/pdf2 -O5 -blpdata -D_ILS_MACROS -qfdpr fdpr -R3 252.eon: -qpdf1/pdf2 -O4 -qarch=pwr4 -qtune=pwr4 -qalign=natural -D_ILS_MACROS 253.perlbmk: -qpdf1/pdf2 -O5 -lhmu -qalign=natural 254.gap: -qpdf1/pdf2 -O5 -lhmu -qalign=natural -D_ILS_MACROS -blpdata 255.vortex: -qpdf1/pdf2 -O5 -lhmu -qalign=natural -D_ILS_MACROS -blpdata 256.bzip2: -qpdf1/pdf2 -O5 -blpdata -D_ILS_MACROS -qfdpr fdpr -R3 300.twolf: -qpdf1/pdf2 -O5 -blpdata -D_ILS_MACROS SMT: Acronym for "Simultaneous Multi-Threading". A processor technology that allows the simultaneous execution of multiple thread contexts within a single processor core. (Enabled by default) MCM: Acronym for "Multi-Chip Module" (four dual-core processor chips + four L3-cache chips) This system contains 8 MCMs. SUT: Acronym for "System Under Test" C: IBM XL C for AIX invoked as xlc C++: IBM XL C++ for AIX invoked as xlC APAR IY60349 was applied to AIX to enable new hardware support. ulimits set to unlimited. Large page mode and memory affinity were set as follows: vmo -r -o lgpg_regions=7000 -o lgpg_size=16777216 -o memory_affinity=1 chuser capabilities=CAP_BYPASS_RAC_VMM,CAP_PROPAGATE $USER shutdown -r export MEMORY_AFFINITY=MCM The following config-file entry was used to assign each benchmark process to a core: submit = schedule.128 \$SPECUSERNUM $command with the "schedule.128" function defined as follows: #!/bin/ksh index=$1 shift 1 # Strip off the residual arguments; the rest is the command. if [[ $index -ge 64 ]] then target=$((1+2*(index-64))) else target=$((2*index)) fi bindprocessor $$ $target # Schedule this job to the corresponding core. $* # Now run the command. The "bindprocessor" AIX command binds a process to a CPU core. ----------------------------------------------------------------------------- For questions about this result, please contact the tester. For other inquiries, please contact webmaster@spec.org. Copyright 1999-2004 Standard Performance Evaluation Corporation Generated on Tue Nov 2 17:03:07 2004 by SPEC CPU2000 ASCII formatter v2.1