SPEC OMPL2001 Summary SGI SGI Altix 4700 Density System (1600MHz 24M L3, DC Itanium2 9050) Mon Jul 17 11:38:24 2006 SPEC License #HPG0014 Tester: SGI Test date: Jul-2006 Test Site: SGI Hardware availability: Jul-2006 Software availability: Jul-2006 Base Base Base Peak Peak Peak Benchmarks Ref Time Run Time Ratio Ref Time Run Time Ratio ------------- -------- -------- -------- -------- -------- -------- 311.wupwise_l 9200 85.4 1723020 9200 85.4 1723020 311.wupwise_l 9200 85.7 1717280* 9200 85.7 1717280* 313.swim_l 12500 124 1606568* 12500 124 1606568* 313.swim_l 12500 124 1618820 12500 124 1618820 315.mgrid_l 13500 283 764544* 13500 258 837444* 315.mgrid_l 13500 280 770234 13500 258 838018 317.applu_l 13500 297 727215* 13500 297 727215* 317.applu_l 13500 296 728836 13500 296 728836 321.equake_l 13000 472 440372* 13000 472 440372* 321.equake_l 13000 470 442341 13000 470 442341 325.apsi_l 10500 186 903939 10500 174 966762* 325.apsi_l 10500 187 900515* 10500 173 973141 327.gafort_l 11000 298 590369 11000 298 590369 327.gafort_l 11000 300 587544* 11000 300 587544* 329.fma3d_l 23500 496 758395* 23500 496 758395* 329.fma3d_l 23500 489 769171 23500 489 769171 331.art_l 25000 122 3283637* 25000 122 3283637* 331.art_l 25000 120 3331835 25000 120 3331835 ======================================================================== 311.wupwise_l 9200 85.7 1717280* 9200 85.7 1717280* 313.swim_l 12500 124 1606568* 12500 124 1606568* 315.mgrid_l 13500 283 764544* 13500 258 837444* 317.applu_l 13500 297 727215* 13500 297 727215* 321.equake_l 13000 472 440372* 13000 472 440372* 325.apsi_l 10500 187 900515* 10500 174 966762* 327.gafort_l 11000 300 587544* 11000 300 587544* 329.fma3d_l 23500 496 758395* 23500 496 758395* 331.art_l 25000 122 3283637* 25000 122 3283637* SPECompLbase2001 987139 SPECompLpeak2001 1005076 HARDWARE -------- Hardware Vendor: SGI Model Name: SGI Altix 4700 Density System (1600MHz 24M L3, DC Itanium2 9050) CPU: Intel DC Itanium2 Processor 9050 CPU MHz: 1600 FPU: Integrated CPU(s) enabled: 256 cores, 128 chips, 2 cores/chip (Hyper-Threading Technology disabled) CPU(s) orderable: 2-1024 cores Primary Cache: 16KBI + 16KBD (on chip) per core Secondary Cache: 1MBI + 256KBD (on chip) per core L3 Cache: 12.0MB (on chip) per core Other Cache: N/A Memory: 512 GB (8*1G PC3200 DIMMS per 4 core module) Disk Subsystem: 1 x 147 GB SCSI (Seagate Cheetah 10k rpm) Other Hardware: None SOFTWARE -------- OpenMP Threads: 256 Parallel: OpenMP Operating System: SUSE Linux Enterprise Server 10 + SGI ProPack(TM) 5 Compiler: Intel(R) Fortran Compiler for Linux 9.1 (Build 20060707) Intel(R) C++ Compiler for Linux 9.1 (Build 20060707) File System: xfs System State: Multi-user NOTES ----- Baseline optimization flags: C programs: -openmp -O3 -IPF_fp_relaxed -ipo -ansi_alias (ONESTEP) Fortran programs: -openmp -O3 -IPF_fp_relaxed -ipo (ONESTEP) OpenMP runtime library libguide.a statically linked Extra Flags: 331.art_l: -DINTS_PER_CACHELINE=32 -DDBLS_PER_CACHELINE=16 User environment: OMP_NUM_THREADS 256 limit stacksize 128000 KMP_STACKSIZE 124M KMP_LIBRARY TURNAROUND OMP_DYNAMIC FALSE KMP_SCHEDULE static,balanced Peak optimization flags 311.wupwise_l: basepeak=true 313.swim_l: basepeak=true 315.mgrid_l: -openmp -O3 -IPF_fp_relaxed -ipo OpenMP runtime library libguide.a statically linked OMP_NUM_THREADS = 128 submit = dplace -x2 -e -c255-0:2 $command 317.applu_l: basepeak=true 321.equake_l: basepeak=true 325.apsi_l: -openmp -O3 -IPF_fp_relaxed -ipo OpenMP runtime library libguide.a statically linked OMP_NUM_THREADS = 128 submit = dplace -x2 -e -c255-0:2 $command 327.gafort_l: basepeak=true 329.fma3d_l: basepeak=true 331.art_l: basepeak=true Required alternate sources: Add critical region around update of linked list in parallel loop. Approved src.alt available as ompl-purdue1-20040324.tar.gz Used for 331.art_l, base and peak. For all benchmarks threads were bound to cores using the following submit command: dplace -x2 -e -cNTM1,0 $command, where NTM1 is the number of threads minus 1. This binds threads in order of creation, beginning with the master thread on core NTM1, the first slave thread on core NTM1-1, and so on. The -x2 flag instructs dplace to skip placement of the lightweight OpenMP monitor thread, which is created prior to the slave threads. For a description of SGI's compiler flags, portability flags, and system parameters used to generate this result, please refer to the SGI-20060801-Linux-Intel9.1-IPF.txt file in the flags directory. ----------------------------------------------------------------------------- For questions about this result, please contact the tester. For other inquiries, please contact webmaster@spec.org. Copyright 1999-2002 Standard Performance Evaluation Corporation Generated on Wed Aug 2 17:21:47 2006 by SPEC OMP2001 ASCII formatter v2.1