OMPM2001 Result: SGI SGI Rackable C1001-G5 (AMD Opteron 6174, 2.20 GHz)

Benchmark	Reference Time	Base Runtime	Base Ratio	Peak Runtime	Peak Ratio
310.wupwise_m	6000	70.8	84732	69.7	86068
312.swim_m	6000	134	44940	112	53601
314.mgrid_m	7300	128	56942	128	56942
316.applu_m	4000	109	36763	102	39107
318.galgel_m	5100	163	31271	163	31271
320.equake_m	2600	56.2	46282	47.5	54690
324.apsi_m	3400	56.5	60158	56.5	60158
326.gafort_m	8700	175	49671	175	49671
328.fma3d_m	4600	133	34594	133	34594
330.art_m	6400	41.0	156049	41.0	156049
332.ammp_m	7000	244	28701	244	28701
SPECompMbase2001	50410
	SPECompMpeak2001	52374

Benchmark

Reference
Time

Base
Runtime

Base
Ratio

Peak
Runtime

Peak
Ratio

310.wupwise_m

6000

70.8

84732

69.7

86068

312.swim_m

6000

134

44940

112

53601

314.mgrid_m

7300

128

56942

128

56942

316.applu_m

4000

109

36763

102

39107

318.galgel_m

5100

163

31271

163

31271

320.equake_m

2600

56.2

46282

47.5

54690

324.apsi_m

3400

56.5

60158

56.5

60158

326.gafort_m

8700

175

49671

175

49671

328.fma3d_m

4600

133

34594

133

34594

330.art_m

6400

41.0

156049

41.0

156049

332.ammp_m

7000

244

28701

244

28701

SPECompMbase2001

50410

SPECompMpeak2001

52374

Hardware

Hardware Vendor:

SGI

Model Name:

SGI Rackable C1001-G5 (AMD Opteron 6174, 2.20 GHz)

CPU:

AMD Opteron(tm) Processor 6174

CPU MHz:

2200

FPU:

Integrated

CPU(s) enabled:

24 cores, 2 chips, 12 cores/chip

CPU(s) orderable:

1-2 chips

Primary Cache:

64 KB I + 64 KB D on chip per core

Secondary Cache:

512 KB I+D on chip per core

L3 Cache:

12 MB I+D on chip per chip, 6 MB shared / 6 cores

Other Cache:

None

Memory:

64 GB (16 x 4 GB, 2Rx4 PC3-10600R-9, ECC)

Disk Subsystem:

5.8 TB RAID 5
20 x 146 GB + 40 x 73 GB FC (Seagate Cheetah 15K.4)

Other Hardware:

None

Software

OpenMP Threads:

Parallel:

OpenMP

Operating System:

SUSE Linux Enterprise Server 11 SP1 (x86_64),
Kernel 2.6.32.13-0.4-default

Compiler:

Open64 Compiler Suite: Version 4.2.4

File System:

NFSv3 IPoIB

System State:

Run Level 3 (Multi-User)

Notes / Tuning Information
Base and Peak User Environment (bash): export OMP_NUM_THREADS=24 export OMP_DYNAMIC=FALSE export O64_OMP_SPIN_COUNT=1000000000 export O64_OMP_SPIN_USER_LOCK=TRUE export O64_OMP_SET_AFFINITY=TRUE export O64_OMP_AFFINITY_MAP=" 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23" export OMP_SLAVE_STACK_SIZE=22M ulimit -s unlimited For Compiler/OpenMP flags description please refer: SGI-x86-open64-424-flags.html Base tuning: C : opencc -mp -Ofast -march=barcelona -mcmodel=medium Fortran : openf90 -mp -Ofast -march=barcelona -OPT:early_mp=on -mcmodel=medium ONESTEP=yes Extra flags 330.art_m: -DINTS_PER_CACHELINE=16 -DDBLS_PER_CACHELINE=8 Portability flags: 318.galgel_m: -fixedform -col80 Peak tuning: ONESTEP=yes for all benchmarks 310.wupwise_m: FOPTIMIZE= -mp -Ofast -msse3 -OPT:early_mp=on -LNO:prefetch_ahead=5:prefetch=3 -march=barcelona -mcmodel=medium -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 -IPA:linear=on:plimit=50000:callee_limit=5000 -INLINE:aggressive=on -CG:load_exe=0 312.swim_m: FOPTIMIZE= -mp -O3 -OPT:Ofast -m3dnow -mcmodel=medium -LNO:simd=0 314.mgrid_m: basepeak=yes 316.applu_m: FOPTIMIZE= -mp -Ofast -march=barcelona -OPT:early_mp=on -mcmodel=medium 318.galgel_m: basepeak=yes 320.equake_m: COPTIMIZE= -Ofast -mp -OPT:alias=disjoint -march=barcelona -mcmodel=medium 324.apsi_m: basepeak=yes 326.gafort_m: basepeak=yes 328.fma3d_m: basepeak=yes 330.art_m: basepeak=yes 332.ammp_m: basepeak=yes Alternate Sources : 314.mgrid_m (base,peak): intel 312.swim_m (peak): ompl.32 316.applu_m (peak): ompl 320.equake_m (peak): ompl Additional notes for disk subsystem: The disk subsystem is mounted through NFS over IPoIB via 4x DDR InfiniBand network. The local disk on the system is only for the OS.

Notes / Tuning Information

 Base and Peak User Environment (bash):
   export OMP_NUM_THREADS=24
   export OMP_DYNAMIC=FALSE
   export O64_OMP_SPIN_COUNT=1000000000
   export O64_OMP_SPIN_USER_LOCK=TRUE
   export O64_OMP_SET_AFFINITY=TRUE
   export O64_OMP_AFFINITY_MAP=" 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23" 
   export OMP_SLAVE_STACK_SIZE=22M
   ulimit -s unlimited
   
 For Compiler/OpenMP flags description please refer:   
 SGI-x86-open64-424-flags.html 
 
 Base tuning:
   C       : opencc -mp -Ofast -march=barcelona -mcmodel=medium
   Fortran : openf90 -mp -Ofast -march=barcelona -OPT:early_mp=on -mcmodel=medium
   ONESTEP=yes 
  
 Extra flags
   330.art_m: -DINTS_PER_CACHELINE=16 -DDBLS_PER_CACHELINE=8
  
 Portability flags: 
   318.galgel_m: -fixedform -col80   

 Peak tuning:
   ONESTEP=yes for all benchmarks
   
   310.wupwise_m: FOPTIMIZE= -mp -Ofast -msse3 -OPT:early_mp=on 
    -LNO:prefetch_ahead=5:prefetch=3  -march=barcelona -mcmodel=medium
    -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 
    -IPA:linear=on:plimit=50000:callee_limit=5000 -INLINE:aggressive=on -CG:load_exe=0
   312.swim_m: FOPTIMIZE= -mp -O3 -OPT:Ofast -m3dnow -mcmodel=medium -LNO:simd=0
   314.mgrid_m: basepeak=yes    
   316.applu_m: FOPTIMIZE= -mp -Ofast -march=barcelona -OPT:early_mp=on -mcmodel=medium
   318.galgel_m: basepeak=yes
   320.equake_m: COPTIMIZE= -Ofast -mp -OPT:alias=disjoint -march=barcelona -mcmodel=medium
   324.apsi_m: basepeak=yes 
   326.gafort_m: basepeak=yes 
   328.fma3d_m: basepeak=yes   
   330.art_m: basepeak=yes   
   332.ammp_m: basepeak=yes

 Alternate Sources :
   314.mgrid_m (base,peak): intel
   312.swim_m (peak): ompl.32
   316.applu_m (peak): ompl
   320.equake_m (peak): ompl

 Additional notes for disk subsystem:
    The disk subsystem is mounted through NFS over IPoIB via 4x DDR 
    InfiniBand network. The local disk on the system is only for the OS.

First published at SPEC.org on 26-Jan-2011

Generated on Wed Jan 26 15:20:22 2011 by SPEC OMP2001 HTML formatter v1.01