| CPU2006 license: | 4 | Test date: | May-2009 |
|---|---|---|---|
| Test sponsor: | SGI | Hardware Availability: | Mar-2009 |
| Tested by: | SGI | Software Availability: | Feb-2009 |
| Hardware | |
|---|---|
| CPU Name: | Intel Xeon X5570 |
| CPU Characteristics: | Quad Core, 2.93 GHz Intel Turbo Boost Technology up to 3.33 GHz |
| CPU MHz: | 2933 |
| FPU: | Integrated |
| CPU(s) enabled: | 32 cores, 8 chips, 4 cores/chip, 2 threads/core |
| CPU(s) orderable: | 1,2 chips per blade, 2-16384 blades |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 256 KB I+D on chip per core |
| L3 Cache: | 8 MB I+D on chip per chip |
| Other Cache: | None |
| Memory: | 192 GB (4 x 12*4GB DDR3-1066 CL7 RDIMMs) |
| Disk Subsystem: | 13 TB Lustre Parallel Filesystem 1 Metadata Server and 6 Object Storage Servers 96 x 136 GB SAS (Seagate Cheetah 15000 rpm) |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | SUSE Linux Enterprise Server 10 (x86_64) SP2 with patch Linux kernel 20080917, Kernel 2.6.16.60-0.30-smp |
| Compiler: | Intel C++ Compiler 11.0 for Linux Build 20090131 Package ID: l_cproc_p_11.0.080 |
| Auto Parallel: | No |
| File System: | lustre v1.6.7 over DDR Infiniband |
| System State: | Multi-user, run level 3 |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | SGI ProPack 6 for Linux Service Pack 2 Microquill SmartHeap V8.1 Binutils 2.18.50.0.7.20080502 |
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 | 64 | 727 | 860 | 729 | 857 | 718 | 871 | 64 | 631 | 991 | 628 | 996 | 629 | 993 |
| 401.bzip2 | 64 | 993 | 622 | 991 | 623 | 991 | 623 | 64 | 945 | 653 | 945 | 654 | 944 | 654 |
| 403.gcc | 64 | 663 | 777 | 657 | 784 | 653 | 789 | 64 | 659 | 781 | 665 | 775 | 663 | 777 |
| 429.mcf | 64 | 540 | 1080 | 537 | 1090 | 540 | 1080 | 32 | 268 | 1090 | 268 | 1090 | 270 | 1080 |
| 445.gobmk | 64 | 700 | 959 | 700 | 960 | 701 | 957 | 64 | 638 | 1050 | 635 | 1060 | 635 | 1060 |
| 456.hmmer | 64 | 833 | 717 | 840 | 711 | 832 | 718 | 32 | 309 | 967 | 309 | 968 | 309 | 966 |
| 458.sjeng | 64 | 860 | 901 | 859 | 902 | 857 | 903 | 64 | 777 | 996 | 780 | 993 | 777 | 996 |
| 462.libquantum | 64 | 484 | 2740 | 485 | 2730 | 485 | 2740 | 64 | 484 | 2740 | 484 | 2740 | 484 | 2740 |
| 464.h264ref | 64 | 1133 | 1250 | 1154 | 1230 | 1153 | 1230 | 64 | 1070 | 1320 | 1080 | 1310 | 1074 | 1320 |
| 471.omnetpp | 64 | 578 | 692 | 577 | 693 | 577 | 693 | 64 | 578 | 692 | 577 | 693 | 577 | 693 |
| 473.astar | 64 | 735 | 611 | 734 | 612 | 735 | 611 | 64 | 672 | 668 | 672 | 668 | 673 | 667 |
| 483.xalancbmk | 64 | 421 | 1050 | 420 | 1050 | 419 | 1050 | 64 | 421 | 1050 | 420 | 1050 | 419 | 1050 |
Submit Notes
The config file option 'submit' was used. A submit.pl script was used to distribute benchmark copies across the 4 blades and to pin processes to cores using dplace. Each blade runs a separate instance of the operating system.
General Notes
Adjacent cache line prefetch enabled System has 4 blades with 2 chips/blade.
Base Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_IA32 |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Base Optimization Flags
C benchmarks:
| -xSSE4.2 -ipo -O3 -no-prec-div -static -inline-calloc -opt-malloc-options=3 -opt-prefetch |
C++ benchmarks:
| -xSSE4.2 -ipo -O3 -no-prec-div -opt-prefetch -Wl,-z,muldefs -L/spec/cpu2006.1.1/lib -lsmartheap |
Peak Compiler Invocation
C benchmarks (except as noted below):
| icc | |
| 401.bzip2: | /opt/intel/Compiler/11.0/080/bin/intel64/icc |
| 456.hmmer: | /opt/intel/Compiler/11.0/080/bin/intel64/icc |
| 458.sjeng: | /opt/intel/Compiler/11.0/080/bin/intel64/icc |
C++ benchmarks (except as noted below):
| icpc | |
| 473.astar: | /opt/intel/Compiler/11.0/080/bin/intel64/icpc |
Peak Portability Flags
| 400.perlbench: | -DSPEC_CPU_LINUX_IA32 |
| 401.bzip2: | -DSPEC_CPU_LP64 |
| 456.hmmer: | -DSPEC_CPU_LP64 |
| 458.sjeng: | -DSPEC_CPU_LP64 |
| 462.libquantum: | -DSPEC_CPU_LINUX |
| 473.astar: | -DSPEC_CPU_LP64 |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Peak Optimization Flags
C benchmarks:
C++ benchmarks:
| 471.omnetpp: | basepeak = yes |
| 473.astar: | -xSSE4.2(pass 2) -prof-gen(pass 1) -ipo(pass 2) -O3(pass 2) -no-prec-div(pass 2) -prof-use(pass 2) -ansi-alias -opt-ra-region-strategy=routine -auto-ilp32 -Wl,-z,muldefs -L/spec/cpu2006.1.1/lib -lsmartheap64 |
| 483.xalancbmk: | basepeak = yes |