| CPU2006 license: | 20 | Test date: | Jul-2007 |
|---|---|---|---|
| Test sponsor: | Bull SAS | Hardware Availability: | Mar-2007 |
| Tested by: | Bull SAS | Software Availability: | May-2007 |
| Hardware | |
|---|---|
| CPU Name: | Intel Xeon E5320 |
| CPU Characteristics: | 1.86 GHz, 8 MB L2, 1066 MHz system bus |
| CPU MHz: | 1866 |
| FPU: | Integrated |
| CPU(s) enabled: | 8 cores, 2 chips, 4 cores/chip |
| CPU(s) orderable: | 1 to 2 chips |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 8 MB I+D on chip per chip, 4 MB shared / 2 cores |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 12 GB (12x1 GB) FB-DIMM PC2-4200F ECC CL4 |
| Disk Subsystem: | 1x73 GB SAS, 15000 RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | SUSE LINUX Enterprise Server 10 Kernel 2.6.16.21-0.8-smp for x86_64 |
| Compiler: | Intel C++ Compiler for Linux32 and Linux64 version 10.0 Build 20070426 Package ID: l_cc_p_10.0.023 |
| Auto Parallel: | No |
| File System: | ext2 |
| System State: | Multi-user run level 3 |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | SmartHeap library V8.1 Binutils 2.17.50.0.15 |
Results Table
| Benchmark | Base | Peak | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | |
| Results appear in the order in which they were run. Bold underlined text indicates a median measurement. | ||||||||||||||
| 400.perlbench | 8 | 806 | 97.0 | 803 | 97.4 | 817 | 95.7 | 8 | 687 | 114 | 684 | 114 | 685 | 114 |
| 401.bzip2 | 8 | 1384 | 55.8 | 1388 | 55.6 | 1377 | 56.1 | 8 | 1335 | 57.8 | 1331 | 58.0 | 1322 | 58.4 |
| 403.gcc | 8 | 1172 | 54.9 | 1161 | 55.5 | 1167 | 55.2 | 8 | 1172 | 54.9 | 1161 | 55.5 | 1167 | 55.2 |
| 429.mcf | 8 | 1711 | 42.6 | 1711 | 42.6 | 1709 | 42.7 | 8 | 1707 | 42.7 | 1704 | 42.8 | 1706 | 42.8 |
| 445.gobmk | 8 | 889 | 94.4 | 887 | 94.6 | 890 | 94.3 | 8 | 824 | 102 | 818 | 103 | 818 | 103 |
| 456.hmmer | 8 | 965 | 77.4 | 965 | 77.4 | 964 | 77.4 | 8 | 772 | 96.6 | 771 | 96.8 | 771 | 96.8 |
| 458.sjeng | 8 | 1126 | 86.0 | 1125 | 86.0 | 1118 | 86.6 | 8 | 1005 | 96.4 | 1002 | 96.6 | 1004 | 96.4 |
| 462.libquantum | 8 | 4387 | 37.8 | 4386 | 37.8 | 4386 | 37.8 | 8 | 4204 | 39.4 | 4196 | 39.5 | 4189 | 39.6 |
| 464.h264ref | 8 | 1182 | 150 | 1187 | 149 | 1183 | 150 | 8 | 1123 | 158 | 1122 | 158 | 1122 | 158 |
| 471.omnetpp | 8 | 1542 | 32.4 | 1546 | 32.3 | 1545 | 32.4 | 8 | 1540 | 32.5 | 1540 | 32.5 | 1541 | 32.4 |
| 473.astar | 8 | 1163 | 48.3 | 1154 | 48.7 | 1166 | 48.2 | 8 | 1155 | 48.6 | 1150 | 48.8 | 1152 | 48.7 |
| 483.xalancbmk | 8 | 714 | 77.3 | 714 | 77.3 | 716 | 77.1 | 8 | 714 | 77.3 | 714 | 77.3 | 716 | 77.1 |
Operating System Notes
'ulimit -s unlimited' was used to set the stacksize to unlimited prior to run '/usr/bin/taskset' used to bind processes to CPUs All benchmarks compiled in 32-bit mode except 401.bzip2 and 456.hmmer, for peak, are compiled in 64-bit mode
General Notes
The NovaScale R440 and the NovaScale R460 models are electronically equivalent. The results have been measured on a NovaScale R460 model.
Base Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_IA32 |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Base Optimization Flags
C benchmarks:
| -fast -Wl,-z,muldefs |
C++ benchmarks:
| -xT -ipo -O3 -no-prec-div -ansi-alias -Wl,-z,muldefs -L/spec/cpu2006/lib -lsmartheap |
Peak Compiler Invocation
C benchmarks (except as noted below):
C++ benchmarks:
| icpc |
Peak Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_IA32 |
| 401.bzip2: | -DSPEC_CPU_LP64 |
| 456.hmmer: | -DSPEC_CPU_LP64 |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Peak Optimization Flags
C benchmarks:
| 400.perlbench: | -prof_gen(pass 1) -prof_use(pass 2) -fast -ansi-alias -prefetch -Wl,-z,muldefs |
| 401.bzip2: | -prof_gen(pass 1) -prof_use(pass 2) -fast -Wl,-z,muldefs |
| 403.gcc: | basepeak = yes |
| 429.mcf: | -fast -prefetch -Wl,-z,muldefs |
| 445.gobmk: | -prof_gen(pass 1) -prof_use(pass 2) -xT -O2 -ipo -no-prec_div -ansi-alias -Wl,-z,muldefs |
| 456.hmmer: | -prof_gen(pass 1) -prof_use(pass 2) -fast -unroll2 -ansi-alias -Wl,-z,muldefs |
| 458.sjeng: | -prof_gen(pass 1) -prof_use(pass 2) -fast -unroll4 -Wl,-z,muldefs |
| 462.libquantum: | -prof_gen(pass 1) -prof_use(pass 2) -fast -unroll4 -Ob0 -prefetch -opt-streaming-stores always -Wl,-z,muldefs |
| 464.h264ref: | Same as 456.hmmer |
C++ benchmarks:
| 471.omnetpp: | -prof_gen(pass 1) -prof_use(pass 2) -xT -O3 -ipo -no-prec_div -ansi-alias -Wl,-z,muldefs -L/spec/cpu2006/lib -lsmartheap |
| 473.astar: | Same as 471.omnetpp |
| 483.xalancbmk: | basepeak = yes |