SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

SGI

SGI Altix ICE 8400EX
(AMD Opteron 6180 SE, 2.5GHz)

MPI2007 license: 4 Test date: Jun-2011
Test sponsor: SGI Hardware Availability: Mar-2011
Tested by: SGI Software Availability: Aug-2011
Benchmark results graph

Results Table

Benchmark Base Peak
Ranks Seconds Ratio Seconds Ratio Seconds Ratio Ranks Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
104.milc 768 27.5 56.9  25.8 60.6  25.7 60.9  768 27.5 56.9 25.8 60.6 25.7 60.9
107.leslie3d 768 95.2 54.8  94.1 55.4  94.8 55.1  768 95.2 54.8 94.1 55.4 94.8 55.1
113.GemsFDTD 768 747   8.45 749   8.42 749   8.42 192 409   15.4 407   15.5 407   15.5
115.fds4 768 25.4 76.7  25.4 76.7  25.9 75.3  696 27.6 70.8 25.1 77.7 25.2 77.4
121.pop2 768 192   21.5  190   21.7  195   21.1  768 192   21.5 190   21.7 195   21.1
122.tachyon 768 45.1 62.0  45.1 62.1  44.9 62.2  768 45.1 62.0 45.1 62.1 44.9 62.2
126.lammps 768 256   11.4  274   10.7  273   10.7  192 214   13.6 214   13.6 214   13.6
127.wrf2 768 102   76.5  101   77.3  101   77.3  768 102   76.5 101   77.3 101   77.3
128.GAPgeofem 768 29.0 71.1  29.1 71.0  29.2 70.7  768 29.0 71.1 29.1 71.0 29.2 70.7
129.tera_tf 768 61.1 45.3  61.1 45.3  61.8 44.8  768 61.1 45.3 61.1 45.3 61.8 44.8
130.socorro 768 127   30.0  126   30.2  128   29.9  768 127   30.0 126   30.2 128   29.9
132.zeusmp2 768 58.5 53.0  58.5 53.0  58.7 52.9  528 57.0 54.4 56.9 54.5 56.8 54.6
137.lu 768 58.0 63.4  58.0 63.4  57.9 63.5  504 55.5 66.2 55.3 66.4 55.4 66.4
Hardware Summary
Type of System: Homogeneous
Compute Node: SGI Altix ICE 8400EX Compute Node
Interconnect: InfiniBand (MPI and I/O)
File Server Node: SGI InfiniteStorage 4000
Total Compute Nodes: 32
Total Chips: 64
Total Cores: 768
Total Threads: 768
Total Memory: 2 TB
Base Ranks Run: 768
Minimum Peak Ranks: 192
Maximum Peak Ranks: 768
Software Summary
C Compiler: Intel C Compiler for Linux
Version 11.1, Build 20100806
C++ Compiler: Intel C++ Compiler for Linux
Version 11.1, Build 20100806
Fortran Compiler: Intel Fortran Compiler for Linux
Version 11.1, Build 20100806
Base Pointers: 64-bit
Peak Pointers: 64-bit
MPI Library: SGI MPT 2.04 Patch 10789
Other MPI Info: OFED 1.4.2
Pre-processors: None
Other Software: None

Node Description: SGI Altix ICE 8400EX Compute Node

Hardware
Number of nodes: 32
Uses of the node: compute
Vendor: SGI
Model: SGI Altix ICE 8400EX (AMD Opteron 6180 SE, 2.5GHz)
CPU Name: AMD Opteron 6180 SE
CPU(s) orderable: 1-2 chips
Chips enabled: 2
Cores enabled: 24
Cores per chip: 12
Threads per core: 1
CPU Characteristics: 12 Cores/chip, 2.5 GHz
CPU MHz: 2500
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: None
Other Hardware: None
Adapter: Mellanox MT26428 ConnectX IB QDR
(PCIe x8 Gen2 5 GT/s)
Number of Adapters: 1
Slot Type: PCIe x8 Gen2
Data Rate: InfiniBand 4x QDR
Ports Used: 2
Interconnect Type: InfiniBand
Software
Adapter: Mellanox MT26428 ConnectX IB QDR
(PCIe x8 Gen2 5 GT/s)
Adapter Driver: OFED-1.4.2
Adapter Firmware: 2.7.0
Operating System: SUSE Linux Enterprise Server 11 SP1 (x86_64)
Kernel 2.6.32.27-0.2-default
Local File System: NFSv3
Shared File System: NFSv3 IPoIB
System State: Run Level 3 (Multi-User)
Other Software: SGI Performance Suite 1.0,
Build 702r19.sles11-1010072114
SGI Tempo Compute Node 2.2,
Build 702r19.sles11-1010072114

Node Description: SGI InfiniteStorage 4000

Hardware
Number of nodes: 1
Uses of the node: fileserver
Vendor: SGI
Model: SGI Altix 450 (Intel Itanium 2, 1.6GHz)
CPU Name: Intel Itanium 2 9030
CPU(s) orderable: 2-38 chips
Chips enabled: 2
Cores enabled: 4
Cores per chip: 2
Threads per core: 1
CPU Characteristics: 1.6GHz/8MB, 533MHz FSB
CPU MHz: 1600
Primary Cache: 16 KB I + 16 KB D on chip per core
Secondary Cache: 1 MB I + 256 KB D on chip per core
L3 Cache: 4 MB I+D on chip per core
Other Cache: None
Memory: 24 GB (12 x 2 GB, 2Rx4 PC2-3200-3, ECC)
Disk Subsystem: 16 TB RAID 5
32 x 500 GB SATA (Seagate Barracuda 7.2K)
Other Hardware: None
Adapter: Mellanox MT25208 InfiniHost III Ex
(PCIe x8 Gen1 2.5 GT/s)
Number of Adapters: 2
Slot Type: PCIe x8 Gen1
Data Rate: InfiniBand 4x DDR
Ports Used: 2
Interconnect Type: InfiniBand
Software
Adapter: Mellanox MT25208 InfiniHost III Ex
(PCIe x8 Gen1 2.5 GT/s)
Adapter Driver: OFED-1.4.2
Adapter Firmware: 5.3.0
Operating System: SUSE Linux Enterprise Server 11 SP1 (ia64)
Kernel 2.6.32.12-0.7-default
Local File System: xfs
Shared File System: --
System State: Run Level 3 (Multi-User)
Other Software: SGI ProPack 7SP1 for Linux,
Build 701r2.sles11-1005242307

Interconnect Description: InfiniBand (MPI and I/O)

Hardware
Vendor: Mellanox Technologies and SGI
Model: None
Switch Model: SGI QDR_1.5_HYPR_2454 with Mellanox Device 48438
(Infiniscale IV)
Number of Switches: 4
Number of Ports: 36
Data Rate: InfiniBand 4x QDR
Firmware: 5040005
Topology: Enhanced Hypercube
Primary Use: MPI and I/O traffic

Submit Notes

The config file option 'submit' was used.

General Notes

 Software environment:
   export MPI_REQUEST_MAX=65536
   export MPI_TYPE_MAX=32768
   export MPI_BUFS_THRESHOLD=1
   ulimit -s unlimited

 BIOS settings:
   AMI BIOS version 1.0a

 Job Placement:
   Each MPI job is assigned to a topologically compact
   set of nodes, i.e. the minimal needed number of switches was
   used for each job: 1 switch for up to 192 ranks, 2 switches
   for 384 ranks, 4 switches for 768 ranks, 8 switches for
   1536 ranks and 16 switches for 3072 ranks.

 Peak run:
   In the peak run, some benchmarks used different number of ranks
   from base. It is the only difference between base and peak.

Compiler Invocation

C benchmarks:

 icc 

C++ benchmarks:

126.lammps:  icpc 

Fortran benchmarks:

 ifort 

Benchmarks using both Fortran and C:

 icc   ifort 

Portability Flags

121.pop2:  -DSPEC_MPI_CASE_FLAG 
127.wrf2:  -DSPEC_MPI_CASE_FLAG   -DSPEC_MPI_LINUX 

Base Optimization Flags

C benchmarks:

 -O3   -xSSE2   -ipo   -no-prec-div 

C++ benchmarks:

126.lammps:  -O3   -xSSE2   -ipo   -no-prec-div   -ansi-alias 

Fortran benchmarks:

 -O3   -xSSE2   -ipo   -no-prec-div 

Benchmarks using both Fortran and C:

 -O3   -xSSE2   -ipo   -no-prec-div 

Peak Optimization Flags

C benchmarks:

104.milc:  basepeak = yes 
122.tachyon:  basepeak = yes 

C++ benchmarks:

126.lammps:  -O3   -xSSE2   -ipo   -no-prec-div   -ansi-alias 

Fortran benchmarks:

107.leslie3d:  basepeak = yes 
113.GemsFDTD:  -O3   -xSSE2   -ipo   -no-prec-div 
129.tera_tf:  basepeak = yes 
137.lu:  Same as 113.GemsFDTD 

Benchmarks using both Fortran and C:

115.fds4:  -O3   -xSSE2   -ipo   -no-prec-div 
121.pop2:  basepeak = yes 
127.wrf2:  basepeak = yes 
128.GAPgeofem:  basepeak = yes 
130.socorro:  basepeak = yes 
132.zeusmp2:  Same as 115.fds4 

Other Flags

C benchmarks:

 -lmpi 

C++ benchmarks:

126.lammps:  -lmpi 

Fortran benchmarks:

 -lmpi 

Benchmarks using both Fortran and C:

 -lmpi 

The flags file that was used to format this result can be browsed at
http://www.spec.org/mpi2007/results/flags/SGI_x86_64_Intel111_flags.20110715.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/mpi2007/results/flags/SGI_x86_64_Intel111_flags.20110715.xml.