| CPU2006 license: | 9006 | Test date: | Apr-2008 |
|---|---|---|---|
| Test sponsor: | NEC Corporation | Hardware Availability: | Oct-2007 |
| Tested by: | NEC Corporation | Software Availability: | Feb-2008 |
| Hardware | |
|---|---|
| CPU Name: | Intel Xeon X5355 |
| CPU Characteristics: | 2.66 GHz, 2x4 MB L2 shared, 1333 MHz bus |
| CPU MHz: | 2667 |
| FPU: | Integrated |
| CPU(s) enabled: | 8 cores, 2 chips, 4 cores/chip |
| CPU(s) orderable: | 1,2 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 (6x2 GB PC2-5300F, 2 rank, CL5-5-5, ECC) |
| Disk Subsystem: | 2x73.2 GB SAS, 15000RPM, Software RAID Level1 |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Red Hat Enterprise Linux AS release 4 (Nahant Update 5), Kernel 2.6.9-55.0.12.ELsmp on an X86_64 |
| Compiler: | Intel C++ and Fortran Compiler for Linux32 and Linux64 version 10.1 Build 20070913 Package ID: l_cc_p_10.1.008, l_fc_p_10.1.008 |
| Auto Parallel: | Yes |
| File System: | ext3 |
| System State: | Run level 3 (multi-user) |
| Base Pointers: | 64-bit |
| Peak Pointers: | 32/64-bit |
| Other Software: | binutils-2.17.tar.gz, Version 2.17 ft Server Control Software 5.0-0231 |
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. | ||||||||||||||
| 410.bwaves | 8 | 3494 | 31.1 | 3497 | 31.1 | 3512 | 31.0 | 4 | 1733 | 31.4 | 1730 | 31.4 | 1726 | 31.5 |
| 416.gamess | 8 | 1077 | 145 | 1077 | 145 | 1080 | 145 | 8 | 1067 | 147 | 1063 | 147 | 1063 | 147 |
| 433.milc | 8 | 3501 | 21.0 | 3521 | 20.9 | 3464 | 21.2 | 8 | 3403 | 21.6 | 3407 | 21.6 | 3409 | 21.5 |
| 434.zeusmp | 8 | 1263 | 57.7 | 1271 | 57.3 | 1254 | 58.0 | 8 | 1246 | 58.4 | 1235 | 59.0 | 1234 | 59.0 |
| 435.gromacs | 8 | 462 | 124 | 462 | 124 | 462 | 124 | 8 | 457 | 125 | 456 | 125 | 457 | 125 |
| 436.cactusADM | 8 | 1561 | 61.2 | 1572 | 60.8 | 1543 | 62.0 | 1 | 149 | 80.3 | 150 | 79.8 | 150 | 79.8 |
| 437.leslie3d | 8 | 3398 | 22.1 | 3426 | 22.0 | 3392 | 22.2 | 8 | 3316 | 22.7 | 3309 | 22.7 | 3310 | 22.7 |
| 444.namd | 8 | 590 | 109 | 589 | 109 | 591 | 109 | 8 | 586 | 109 | 586 | 109 | 588 | 109 |
| 447.dealII | 8 | 757 | 121 | 747 | 123 | 735 | 125 | 8 | 710 | 129 | 741 | 124 | 721 | 127 |
| 450.soplex | 8 | 2402 | 27.8 | 2421 | 27.6 | 2395 | 27.9 | 8 | 2199 | 30.3 | 2198 | 30.4 | 2197 | 30.4 |
| 453.povray | 8 | 260 | 164 | 259 | 164 | 258 | 165 | 8 | 224 | 190 | 224 | 190 | 223 | 191 |
| 454.calculix | 8 | 652 | 101 | 656 | 101 | 658 | 100 | 8 | 486 | 136 | 485 | 136 | 488 | 135 |
| 459.GemsFDTD | 8 | 4465 | 19.0 | 4327 | 19.6 | 4322 | 19.6 | 8 | 4206 | 20.2 | 4231 | 20.1 | 4241 | 20.0 |
| 465.tonto | 8 | 1099 | 71.6 | 1143 | 68.9 | 1128 | 69.8 | 8 | 1094 | 72.0 | 1072 | 73.4 | 1071 | 73.5 |
| 470.lbm | 8 | 7288 | 15.1 | 7167 | 15.3 | 7077 | 15.5 | 4 | 3142 | 17.5 | 3145 | 17.5 | 3140 | 17.5 |
| 481.wrf | 8 | 2149 | 41.6 | 2106 | 42.4 | 2093 | 42.7 | 8 | 2094 | 42.7 | 2094 | 42.7 | 2100 | 42.5 |
| 482.sphinx3 | 8 | 3774 | 41.3 | 3747 | 41.6 | 3741 | 41.7 | 4 | 1191 | 65.5 | 1190 | 65.5 | 1188 | 65.6 |
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 OMP_NUM_THREADS set to number of cores
Platform Notes
This Express5800/320Fc-MR is a fault-tolerant server. Two modules are installed in this server and each module has "2CPU chips,12GB memory", so total "4CPU chips,24GB memory" are on this server. With lockstep technology, these two modules communicate each other and handle the same instructions at the same time, then logically the "CPU,Memory" is recognized as "2CPU chips,12GB memory" by the OS.
General Notes
All benchmarks compiled in 64-bit mode except 437.leslie3d, 450.soplex, 470.lbm and 482.sphinx3, for peak, are compiled in 32-bit mode The NEC Express5800/320Fc-MR(Intel Xeon X5355) and the Bull NovaScale R630 (Intel Xeon X5355,2.66GHz) models are electronically equivalent. The results have been measured on a NEC Express5800/320Fc-MR(Intel Xeon X5355) model.
Base Compiler Invocation
C benchmarks:
| icc |
C++ benchmarks:
| icpc |
Fortran benchmarks:
| ifort |
Benchmarks using both Fortran and C:
| icc ifort |
Base Portability Flags
| 410.bwaves: | -DSPEC_CPU_LP64 |
| 416.gamess: | -DSPEC_CPU_LP64 |
| 433.milc: | -DSPEC_CPU_LP64 |
| 434.zeusmp: | -DSPEC_CPU_LP64 |
| 435.gromacs: | -DSPEC_CPU_LP64 -nofor_main |
| 436.cactusADM: | -DSPEC_CPU_LP64 -nofor_main |
| 437.leslie3d: | -DSPEC_CPU_LP64 |
| 444.namd: | -DSPEC_CPU_LP64 |
| 447.dealII: | -DSPEC_CPU_LP64 |
| 450.soplex: | -DSPEC_CPU_LP64 |
| 453.povray: | -DSPEC_CPU_LP64 |
| 454.calculix: | -DSPEC_CPU_LP64 -nofor_main |
| 459.GemsFDTD: | -DSPEC_CPU_LP64 |
| 465.tonto: | -DSPEC_CPU_LP64 |
| 470.lbm: | -DSPEC_CPU_LP64 |
| 481.wrf: | -DSPEC_CPU_LP64 -DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LINUX |
| 482.sphinx3: | -DSPEC_CPU_LP64 |
Base Optimization Flags
C benchmarks:
| -fast |
C++ benchmarks:
| -fast |
Fortran benchmarks:
| -fast |
Benchmarks using both Fortran and C:
| -fast |
Peak Compiler Invocation
C benchmarks (except as noted below):
| /opt/intel/cc/10.1.008/bin/icc -L/opt/intel/cc/10.1.008/lib -I/opt/intel/cc/10.1.008/include | |
| 433.milc: | icc |
C++ benchmarks (except as noted below):
| icpc | |
| 450.soplex: | /opt/intel/cc/10.1.008/bin/icpc -L/opt/intel/cc/10.1.008/lib -I/opt/intel/cc/10.1.008/include |
Fortran benchmarks (except as noted below):
| ifort | |
| 437.leslie3d: | /opt/intel/fc/10.1.008/bin/ifort -L/opt/intel/fc/10.1.008/lib -I/opt/intel/fc/10.1.008/include |
Benchmarks using both Fortran and C:
| icc ifort |
Peak Portability Flags
| 410.bwaves: | -DSPEC_CPU_LP64 |
| 416.gamess: | -DSPEC_CPU_LP64 |
| 433.milc: | -DSPEC_CPU_LP64 |
| 434.zeusmp: | -DSPEC_CPU_LP64 |
| 435.gromacs: | -DSPEC_CPU_LP64 -nofor_main |
| 436.cactusADM: | -DSPEC_CPU_LP64 -nofor_main |
| 444.namd: | -DSPEC_CPU_LP64 |
| 447.dealII: | -DSPEC_CPU_LP64 |
| 453.povray: | -DSPEC_CPU_LP64 |
| 454.calculix: | -DSPEC_CPU_LP64 -nofor_main |
| 459.GemsFDTD: | -DSPEC_CPU_LP64 |
| 465.tonto: | -DSPEC_CPU_LP64 |
| 481.wrf: | -DSPEC_CPU_LP64 -DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LINUX |
Peak Optimization Flags
C benchmarks:
| 433.milc: | -prof-gen(pass 1) -prof-use(pass 2) -fast -fno-alias -auto-ilp32 |
| 470.lbm: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll2 -scalar-rep- -prefetch -opt-malloc-options=3 |
| 482.sphinx3: | -fast -unroll2 |
C++ benchmarks:
| 444.namd: | -prof-gen(pass 1) -prof-use(pass 2) -fast -fno-alias -auto-ilp32 |
| 447.dealII: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll2 -ansi-alias -scalar-rep- |
| 450.soplex: | -prof-gen(pass 1) -prof-use(pass 2) -fast -opt-malloc-options=3 |
| 453.povray: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll4 -ansi-alias |
Fortran benchmarks:
| 410.bwaves: | -fast -prefetch |
| 416.gamess: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll2 -Ob0 -ansi-alias -scalar-rep- |
| 434.zeusmp: | -prof-gen(pass 1) -prof-use(pass 2) -fast |
| 437.leslie3d: | -prof-gen(pass 1) -prof-use(pass 2) -fast -prefetch -opt-malloc-options=3 |
| 459.GemsFDTD: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll2 -Ob0 -prefetch |
| 465.tonto: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll4 -auto |
Benchmarks using both Fortran and C:
| 435.gromacs: | -prof-gen(pass 1) -prof-use(pass 2) -fast -prefetch -auto-ilp32 |
| 436.cactusADM: | -prof-gen(pass 1) -prof-use(pass 2) -fast -unroll2 -prefetch -parallel -auto-ilp32 |
| 454.calculix: | -fast -unroll-aggressive -auto-ilp32 |
| 481.wrf: | -fast -auto-ilp32 |