SPEC Seal of Reviewal OMPM2001 Result
Copyright © 1999-2002 Standard Performance Evaluation Corporation
SGI
SGI Altix 3000 (1500MHz, Itanium 2)
SPECompMpeak2001 = 35042      
SPECompMbase2001 = 31276      
SPEC license # HPG0014 Tested by: SGI Test site: SGI Test date: Aug-2003 Hardware Avail: Jun-2003 Software Avail: Jul-2003
Benchmark Reference
Time
Base
Runtime
Base
Ratio
Peak
Runtime
Peak
Ratio
Graph Scale
310.wupwise_m 6000 72.6  82666       72.6  82666       310.wupwise_m base result bar (82666)
310.wupwise_m peak result bar (82666)
312.swim_m 6000 53.5  112087        53.5  112087        312.swim_m base result bar (112087)
312.swim_m peak result bar (112087)
314.mgrid_m 7300 218    33462       218    33462       314.mgrid_m base result bar (33462)
314.mgrid_m peak result bar (33462)
316.applu_m 4000 420    9535      342    11701       316.applu_m base result bar (9535)
316.applu_m peak result bar (11701)
318.galgel_m 5100 604    8440      462    11035       318.galgel_m base result bar (8440)
318.galgel_m peak result bar (11035)
320.equake_m 2600 112    23270       79.9  32552       320.equake_m base result bar (23270)
320.equake_m peak result bar (32552)
324.apsi_m 3400 86.5  39284       77.7  43782       324.apsi_m base result bar (39284)
324.apsi_m peak result bar (43782)
326.gafort_m 8700 313    27831       265    32834       326.gafort_m base result bar (27831)
326.gafort_m peak result bar (32834)
328.fma3d_m 4600 184    25039       155    29631       328.fma3d_m base result bar (25039)
328.fma3d_m peak result bar (29631)
330.art_m 6400 51.9  123248        51.9  123248        330.art_m base result bar (123248)
330.art_m peak result bar (123248)
332.ammp_m 7000 490    14299       490    14299       332.ammp_m base result bar (14299)
332.ammp_m peak result bar (14299)
SPECompMbase2001 31276        
  SPECompMpeak2001 35042        

Hardware
Hardware Vendor: SGI
Model Name: SGI Altix 3000 (1500MHz, Itanium 2)
CPU: Intel Itanium 2
CPU MHz: 1500
FPU: Integrated
CPU(s) enabled: 64
CPU(s) orderable: 4-64
Primary Cache: 16KBI + 16KBD (on chip) per CPU
Secondary Cache: 256KB (on chip) per CPU
L3 Cache: 6.0MB (on chip) per CPU
Other Cache: N/A
Memory: 128 GB (16*512MB DIMMS per 4cpu module)
Disk Subsystem: 1 x 36 GB SCSI (Seagate Cheetah 15k rpm)
Other Hardware: None
Software
OpenMP Threads: 64
Parallel: OpenMP
Operating System: SGI ProPack(TM) v2.2
Compiler: Intel(R) Fortran Compiler for Linux 7.1 (Build 20030701)
Intel(R) C++ Compiler for Linux 7.1 (Build 20030701)
File System: xfs
System State: Single-user
Notes / Tuning Information
 Baseline optimization flags: 
   C programs:       -openmp -O3 -ipo -ansi -ansi_alias (ONESTEP)
   Fortran programs: -openmp -O3 -ipo -stack_temps (ONESTEP)
 
 Portability Flags:
   318.galgel_m: -FI -132
 
 Extra Flags:
   330.art_m: -DINTS_PER_CACHELINE=32 -DDBLS_PER_CACHELINE=16

 Baseline user environment:
   OMP_NUM_THREADS=64
   limit stacksize 64000
   KMP_STACKSIZE 31M
   KMP_LIBRARY TURNAROUND 

 Peak optimization flags:
    310.wupwise_m: basepeak=true
    312.swim_m: basepeak=true
    314.mgrid_m: basepeak=true
    316.applu_m: -openmp -O3 -ipo -stack_temps (ONESTEP)
                 OMP_NUM_THREADS=32                
    318.galgel_m: -openmp -O3 -ipo -stack_temps (ONESTEP)
                  OMP_NUM_THREADS=32
    320.equake_m: -openmp -O3 -ipo -ansi -ansi_alias (ONESTEP)
    324.apsi_m: -openmp -O3 -ipo -stack_temps (ONESTEP)
    326.gafort_m: -openmp -O3 -ipo -stack_temps (ONESTEP)
    328.fma3d_m: -openmp -O3 -ipo -stack_temps (ONESTEP)
    330.art_m: basepeak=true
    332.ammp_m: basepeak=true

 Alternate sources:

 ISO C and C++ compatible sources 
 Available as SPEC OMPM2001 source: ompm2001-isoc-20020619.tar.gz
 Used for base: 320.equake_m, 330.art_m, 332.ammp_m  
 and peak: 330.art_m, 332.ammp_m 

 Peak sources:
 
 SPEC OMPL2001 source for 64bit systems modified for SPEC OMPM2001.
 Available as SPEC OMPM2001 Source: ompm2001-srcl64bit-20020807.tar.gz.
 Used for 320.equake_m, 324.apsi_m, 326.gafort_m, and 328.fma3d_m.
 
For all benchmarks threads were bound to CPUs using the following submit command:
dplace -e -c0,x,x,1-NTM1 $command,
where NTM1 is the number of threads minus 1.
 



For questions about this result, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright © 1999-2002 Standard Performance Evaluation Corporation

First published at SPEC.org on 03-Sep-2003

Generated on Wed Sep 3 10:55:36 2003 by SPEC OMP2001 HTML formatter v1.01