| CPU2006 license: | 55 | Test date: | Sep-2007 |
|---|---|---|---|
| Test sponsor: | Dell Inc. | Hardware Availability: | Dec-2006 |
| Tested by: | Dell Inc. | Software Availability: | Oct-2007 |
| Hardware | |
|---|---|
| CPU Name: | AMD Opteron 8218 |
| CPU Characteristics: | |
| CPU MHz: | 2666 |
| FPU: | Integrated |
| CPU(s) enabled: | 8 cores, 4 chips, 2 cores/chip |
| CPU(s) orderable: | 2,4 chips |
| Primary Cache: | 64 KB I + 64 KB D on chip per core |
| Secondary Cache: | 1 MB I+D on chip per core |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 32 GB (16x2GB, DDR2-667 CL5 ECC Dual Rank) |
| Disk Subsystem: | 1 x 250 GB SATA 7200 RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | 64-Bit SUSE LINUX Enterprise Server 10 SP1 |
| Compiler: | The Portland Group (PGI) PGI pgcc 7.1-0 C Compiler PGI pgCC 7.1-0 C++ Compiler |
| Auto Parallel: | No |
| File System: | ReiserFS |
| System State: | Multi-user, run level 3 |
| Base Pointers: | 32/64-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | SmartHeap 8.0 32-bit Library for Linux |
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 | 885 | 88.3 | 885 | 88.3 | 885 | 88.3 | 8 | 739 | 106 | 726 | 108 | 729 | 107 |
| 401.bzip2 | 8 | 1106 | 69.8 | 1108 | 69.7 | 1111 | 69.5 | 8 | 1089 | 70.9 | 1090 | 70.8 | 1091 | 70.8 |
| 403.gcc | 8 | 1212 | 53.1 | 1211 | 53.2 | 1215 | 53.0 | 8 | 894 | 72.0 | 891 | 72.3 | 894 | 72.1 |
| 429.mcf | 8 | 1057 | 69.0 | 1057 | 69.0 | 1057 | 69.0 | 8 | 783 | 93.2 | 784 | 93.0 | 783 | 93.2 |
| 445.gobmk | 8 | 818 | 103 | 818 | 103 | 818 | 103 | 8 | 759 | 111 | 760 | 110 | 760 | 110 |
| 456.hmmer | 8 | 556 | 134 | 557 | 134 | 556 | 134 | 8 | 521 | 143 | 526 | 142 | 523 | 143 |
| 458.sjeng | 8 | 999 | 96.9 | 996 | 97.2 | 993 | 97.5 | 8 | 945 | 102 | 945 | 102 | 945 | 102 |
| 462.libquantum | 8 | 1118 | 148 | 1110 | 149 | 1106 | 150 | 8 | 1118 | 148 | 1110 | 149 | 1106 | 150 |
| 464.h264ref | 8 | 1235 | 143 | 1236 | 143 | 1243 | 142 | 8 | 1235 | 143 | 1236 | 143 | 1243 | 142 |
| 471.omnetpp | 8 | 818 | 61.1 | 820 | 61.0 | 823 | 60.8 | 8 | 818 | 61.1 | 820 | 61.0 | 823 | 60.8 |
| 473.astar | 8 | 828 | 67.8 | 822 | 68.4 | 821 | 68.4 | 8 | 828 | 67.8 | 822 | 68.4 | 821 | 68.4 |
| 483.xalancbmk | 8 | 583 | 94.7 | 582 | 94.8 | 584 | 94.6 | 8 | 581 | 95.1 | 580 | 95.2 | 581 | 95.1 |
General Notes
'ulimit -s unlimited' was used to set environment stack size 'ulimit -l 2457600' was used to set environment locked pages in memory quantity 'numactl' was used to bind one copy per core, and memory to a local NUMA node Set vm/nr_hugepages=1200 in /etc/sysctl.conf mount -t hugetlbfs nodev /mnt/hugepages Environment variable PGI_HUGE_PAGES set to 150
Base Portability Flags
| 400.perlbench: | -DSPEC_CPU_LP64 -DSPEC_CPU_LINUX_X64 |
| 401.bzip2: | -DSPEC_CPU_LP64 |
| 403.gcc: | -DSPEC_CPU_LP64 |
| 429.mcf: | -DSPEC_CPU_LP64 |
| 445.gobmk: | -DSPEC_CPU_LP64 |
| 456.hmmer: | -DSPEC_CPU_LP64 |
| 458.sjeng: | -DSPEC_CPU_LP64 |
| 462.libquantum: | -DSPEC_CPU_LP64 -DSPEC_CPU_LINUX |
| 464.h264ref: | -DSPEC_CPU_LP64 |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Base Optimization Flags
C benchmarks:
| -fast -Mipa=fast -Mipa=inline -Mipa=noarg -Mfprelaxed -Msmartalloc=huge:64 -tp k8-64 -Bstatic_pgi |
C++ benchmarks:
| -fastsse -Mipa=fast -Mipa=inline -Mfprelaxed -Msmartalloc --zc_eh -tp k8-32 -Bstatic_pgi -lsmartheap |
Peak Portability Flags
| 400.perlbench: | -DSPEC_CPU_LP64 -DSPEC_CPU_LINUX_X64 |
| 401.bzip2: | -DSPEC_CPU_LP64 |
| 445.gobmk: | -DSPEC_CPU_LP64 |
| 456.hmmer: | -DSPEC_CPU_LP64 |
| 458.sjeng: | -DSPEC_CPU_LP64 |
| 462.libquantum: | -DSPEC_CPU_LP64 -DSPEC_CPU_LINUX |
| 464.h264ref: | -DSPEC_CPU_LP64 |
| 483.xalancbmk: | -DSPEC_CPU_LINUX |
Peak Optimization Flags
C benchmarks:
| 400.perlbench: | -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=inline(pass 2) -fast -O4 -Mfprelaxed -Msmartalloc=huge:8 -Mnounroll -tp k8-64 -Bstatic_pgi |
| 401.bzip2: | -Mpfi(pass 1) -Mpfo(pass 2) -fast -O4 -Msmartalloc=huge:64 -tp k8-64 -Bstatic_pgi |
| 403.gcc: | -fastsse -Mfprelaxed -Msmartalloc=huge:32 -Mipa=fast -Mipa=inline -tp k8-32 -Bstatic_pgi |
| 429.mcf: | -fastsse -Mipa=fast -Mipa=inline -Msmartalloc=huge:16 -tp k8-32 -Bstatic_pgi |
| 445.gobmk: | -Mpfi(pass 1) -Mpfo(pass 2) -Mipa=fast(pass 2) -fast -O4 -Msmartalloc=huge:32 -Mfprelaxed -Mnovect -tp k8-64 -Bstatic_pgi |
| 456.hmmer: | -fast -Msmartalloc=huge:32 -Mfprelaxed -Msafeptr -Mipa=const -Mipa=ptr -Mipa=arg -tp k8-64 -Bstatic_pgi |
| 458.sjeng: | -Mpfi(pass 1) -Mipa=fast(pass 2) -Mipa=inline(pass 2) -Mipa=noarg(pass 2) -Mpfo(pass 2) -fast -Msmartalloc=huge:32 -Mfprelaxed -tp k8-64 -Bstatic_pgi |
| 462.libquantum: | basepeak = yes |
| 464.h264ref: | basepeak = yes |
C++ benchmarks:
| 471.omnetpp: | basepeak = yes |
| 473.astar: | basepeak = yes |
| 483.xalancbmk: | -fastsse -O4 -Mipa=fast -Mipa=inline -Mfprelaxed -Msmartalloc --zc_eh -tp k8-32 -Bstatic_pgi -lsmartheap |