CINT2000 Result Copyright © 1999-2005 Standard Performance Evaluation Corporation |
IBM Corporation IBM System p5 550 (2100 Mhz, 4 CPU, SLES) |
|
SPEC license # 11 | Tested by: IBM Austin | Test date: Oct-2006 | Hardware Avail: Aug-2006 | Software Avail: Dec-2006 |
Benchmark | Base Copies |
Base Runtime |
Base Ratio |
Copies | Runtime | Ratio | |
---|---|---|---|---|---|---|---|
164.gzip | 8 | 211 | 61.7 | 8 | 211 | 61.7 | |
175.vpr | 8 | 174 | 74.9 | 8 | 174 | 74.9 | |
176.gcc | 8 | 106 | 96.3 | 8 | 106 | 96.3 | |
181.mcf | 8 | 134 | 125 | 8 | 134 | 125 | |
186.crafty | 8 | 138 | 67.1 | 8 | 113 | 81.8 | |
197.parser | 8 | 242 | 69.1 | 8 | 225 | 74.2 | |
252.eon | 8 | 131 | 92.0 | 8 | 138 | 87.5 | |
253.perlbmk | 8 | 253 | 66.1 | 8 | 247 | 67.6 | |
254.gap | 8 | 122 | 83.9 | 8 | 122 | 83.9 | |
255.vortex | 8 | 130 | 136 | 8 | 130 | 136 | |
256.bzip2 | 8 | 153 | 90.7 | 8 | 153 | 90.7 | |
300.twolf | 8 | 317 | 87.9 | 8 | 317 | 87.9 | |
SPECint_rate_base2000 | 85.0 | ||||||
SPECint_rate2000 | 86.7 |
|
|
Notes / Tuning Information |
---|
+FDO Feedback directed optimization enabled by: PASS1=-qpdf1 PASS2=-qpdf2 Integer suite C: invoked as cc C++: invoked as xlC Integer Portability Flags: 176.gcc: -DHOST_WORDS_BIG_ENDIAN 186.crafty: -DLINUX_PPC32 252.eon: -DHAS_ERRLIST 253.perlbmk: -DSPEC_CPU2000_LINUX_PPC32 -DSPEC_CPU2000_NEED_BOOL 254.gap: -DSYS_IS_USG -DSYS_HAS_IOCTL_PROTO -DSYS_HAS_CALLOC_PROTO 300.twolf: -DHAVE_SIGNED_CHAR Additional Peak Portability Flags: 252.eon: -DSPEC_CPU2000_LP64 (for 64-bit compilation) 253.perlbmk: -DSPEC_CPU2000_LP64 (for 64-bit compilation) Integer Base Optimization Flags: C: +FDO -O5 C++: +FDO -O5 Integer Peak Optimization Flags 164.gzip basepeak=1 175.vpr basepeak=1 176.gcc basepeak=1 181.mcf basepeak=1 186.crafty +FDO -O4 -qarch=pwr4 -qtune=pwr4 -q64 197.parser +FDO -O5 -qstaticlink 252.eon +FDO -O5 -q64 253.perlbmk: +FDO -O5 -q64 254.gap basepeak=1 255.vortex basepeak=1 256.bzip2 basepeak=1 300.twolf basepeak=1 System Settings: -- ulimit stack size set to unlimited SMT: Acronym for 'Simultaneous Multi-Threading'. A processor technology that allows the simultaneous execution of multiple thread contexts within a single processor core. SMT is enabled by default. Large pages reserved as follows by root user: echo 240 > /proc/sys/vm/nr_hugepages System configured with libhugetlbfs library for application access to large pages Environment variables set as follows: export HUGETLB_MORECORE=yes export LD_PRELOAD=libhugetlbfs.so (export LD_PRELOAD=libhugetlbfs.so not used for --action build.) Each process was bound to a cpu using submit= with the taskset command submit = taskset -p -c \$SPECUSERNUM \$\$ >/dev/null ; $command |
First published at SPEC.org on 31-Oct-2006
Generated on Tue Oct 31 14:17:59 2006 by SPEC CPU2000 HTML formatter v1.01