SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

IBM

iDP (Intel Xeon L5420, 2.50 GHz)

SPECmpiM_peak2007 = Not Run

MPI2007 license: 3440 Test date: Apr-2009
Test sponsor: Indiana university Hardware Availability: Sep-2008
Tested by: Scott Teige Software Availability: Jan-2009
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 512 96.1 16.3 84.8 18.4 84.9 18.4
107.leslie3d 512 178   29.3 180   28.9 180   29.1
113.GemsFDTD 512 579   10.9 578   10.9 579   10.9
115.fds4 512 37.4 52.2 36.9 52.9 37.5 52.0
121.pop2 512 233   17.7 234   17.7 235   17.6
122.tachyon 512 64.4 43.4 65.4 42.8 65.3 42.8
126.lammps 512 248   11.7 248   11.7 251   11.6
127.wrf2 512 156   50.1 155   50.3 157   49.6
128.GAPgeofem 512 38.5 53.6 39.1 52.8 39.8 51.9
129.tera_tf 512 106   26.1 106   26.1 105   26.3
130.socorro 512 193   19.8 193   19.8 195   19.6
132.zeusmp2 512 78.4 39.6 79.8 38.9 79.2 39.2
137.lu 512 80.5 45.7 81.1 45.3 80.6 45.6
Hardware Summary
Type of System: Homogeneous
Compute Node: iDP node
Interconnects: Gigabit Ethernet
IB Switch
Total Compute Nodes: 64
Total Chips: 128
Total Cores: 512
Total Threads: 512
Total Memory: 2 TB
Base Ranks Run: 512
Minimum Peak Ranks: --
Maximum Peak Ranks: --
Software Summary
C Compiler: Intel C++ Compiler 10.1 for Linux (10.1.013)
C++ Compiler: Intel C++ Compiler 10.1 for Linux (10.1.013)
Fortran Compiler: Intel Fortran Compiler 10.1 for Linux (10.1.013)
Base Pointers: 64-bit
Peak Pointers: 64-bit
MPI Library: Intel MPI 3.1
Other MPI Info: None
Pre-processors: No
Other Software: OFED 1.4 compat-dapl-1.2.13
Intel MPI Library 3.1 for Linux Multi-Purpose
Daemon (MPD)

Node Description: iDP node

Hardware
Number of nodes: 64
Uses of the node: compute
Vendor: IBM
Model: System x iDataPlex dx340
CPU Name: Intel Xeon L5420
CPU(s) orderable: 1-2 chips
Chips enabled: 2
Cores enabled: 8
Cores per chip: 4
Threads per core: 1
CPU Characteristics: 1333 MHz FSB
CPU MHz: 2500
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 12 MB I+D on chip per chip, 6 MB shared / 2 cores
L3 Cache: None
Other Cache: None
Memory: 32 GB (FBDIMM 8x4-GB 667 MHz)
Disk Subsystem: Western Digital 160 GB SATA WD160YS-23SHBO
Other Hardware: None
Adapter: Intel Corporation 80003ES2LAN Gigabit Ethernet
Controller (Copper) (rev 01)
Number of Adapters: 2
Slot Type: --
Data Rate: Gigabit Ethernet
Ports Used: 1
Interconnect Type: Ethernet
Adapter: Mellanox Technologies MT26418 [ConnectX IB DDR,
PCIe 2.0 5GT/s] (rev a0)
Number of Adapters: 1
Slot Type: PCIe x8 Gen2
Data Rate: InfiniBand 4x DDR
Ports Used: 1
Interconnect Type: InfiniBand
Software
Adapter: Intel Corporation 80003ES2LAN Gigabit Ethernet
Controller (Copper) (rev 01)
Adapter Driver: OS default (e1000, v7.3.20-k2-NAPI)
Adapter Firmware: 2.4-0
Adapter: Mellanox Technologies MT26418 [ConnectX IB DDR,
PCIe 2.0 5GT/s] (rev a0)
Adapter Driver: OFED 1.3.1
Adapter Firmware: 2.5.0
Operating System: Red Hat EL v4.7
2.6.9-67.0.22.EL_lustre.1.6.7custom
Local File System: Linux/ext3
Shared File System: IBM N5500 NAS via NFSv3
System State: Multi-User
Other Software: lustre 1.6.7 kernel patches

Interconnect Description: Gigabit Ethernet

Hardware
Vendor: ProCurve Networking
Model: HP ProCurve Switch 5406zl Intelligent Edge J8697A
Switch Model: HP ProCurve Switch 5406zl Intelligent Edge J8697A
Number of Switches: 1
Number of Ports: 144
Data Rate: 1Gbps Ethernet
Firmware: --
Topology: Single switch
Primary Use: I/O traffic

Interconnect Description: IB Switch

Hardware
Vendor: Cisco
Model: Cisco SFS 7024D
Switch Model: Cisco SFS 7024D
Number of Switches: 1
Number of Ports: 288
Data Rate: InfiniBand 4x DDR
Firmware: 4.1.1.1.11
Topology: Single switch
Primary Use: MPI traffic

Submit Notes

The config file option 'submit' was used.

Base Compiler Invocation

C benchmarks:

 mpiicc 

C++ benchmarks:

126.lammps:  mpiicpc 

Fortran benchmarks:

 mpiifort 

Benchmarks using both Fortran and C:

 mpiicc   mpiifort 

Base Portability Flags

121.pop2:  -DSPEC_MPI_CASE_FLAG 
126.lammps:  -DMPICH_IGNORE_CXX_SEEK 
127.wrf2:  -DSPEC_MPI_LINUX   -DSPEC_MPI_CASE_FLAG 

Base Optimization Flags

C benchmarks:

 -O3   -xT   -ipo   -no-prec-div 

C++ benchmarks:

126.lammps:  -O3   -xT   -ipo   -no-prec-div 

Fortran benchmarks:

 -O3   -xT   -ipo   -no-prec-div 

Benchmarks using both Fortran and C:

 -O3   -xT   -ipo   -no-prec-div 

The flags file that was used to format this result can be browsed at
http://www.spec.org/mpi2007/flags/EM64T_Intel101_flags.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/mpi2007/flags/EM64T_Intel101_flags.xml.