SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

Dell Inc.

PowerEdge M605,
Gigabit Ethernet,
Platform MPI 5.6.4,
Intel 10.1 compilers

SPECmpiM_peak2007 = Not Run

MPI2007 license: 021 Test date: Nov-2008
Test sponsor: Platform Computing Inc. Hardware Availability: Jan-2008
Tested by: Platform Computing Inc. Software Availability: Jul-2008
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 64 387 4.04 381 4.11 387 4.05
107.leslie3d 64 1182 4.42 1181 4.42 1182 4.42
113.GemsFDTD 64 1063 5.93 1065 5.92 1069 5.90
115.fds4 64 387 5.04 388 5.03 389 5.01
121.pop2 64 1585 2.60 1596 2.59 1600 2.58
122.tachyon 64 578 4.84 582 4.81 581 4.82
126.lammps 64 743 3.92 748 3.90 744 3.92
127.wrf2 64 1066 7.31 1061 7.35 1066 7.32
128.GAPgeofem 64 345 5.99 342 6.04 350 5.89
129.tera_tf 64 548 5.05 551 5.02 548 5.05
130.socorro 64 1007 3.79 1002 3.81 1004 3.80
132.zeusmp2 64 537 5.78 539 5.76 537 5.78
137.lu 64 836 4.40 840 4.38 836 4.40
Hardware Summary
Type of System: Homogenous
Compute Node: PowerEdge M605
Interconnect: Gigabit Ethernet
File Server Node: PowerEdge M605
Head Node: PowerEdge M605
Total Compute Nodes: 8
Total Chips: 16
Total Cores: 64
Total Threads: 64
Total Memory: 128 GB
Base Ranks Run: 64
Minimum Peak Ranks: --
Maximum Peak Ranks: --
Software Summary
C Compiler: Intel C Compiler 10.1 for Linux (10.1.018)
C++ Compiler: Intel C++ Compiler 10.1 for Linux (10.1.018)
Fortran Compiler: Intel Fortran Compiler 10.1 for Linux (10.1.018)
Base Pointers: 64-bit
Peak Pointers: Not Applicable
MPI Library: Scali MPI Connect 5.6.4-59151
Other MPI Info: Platform Computing Inc has acquired Scali MPI
Connect, hence Platform MPI and Scali MPI Connect
are used synonymously.
Pre-processors: None
Other Software: None

Node Description: PowerEdge M605

Hardware
Number of nodes: 8
Uses of the node: compute, head, fileserver
Vendor: Dell Inc.
Model: M605
CPU Name: AMD Opteron CPU 2350
CPU(s) orderable: 1-2 chips
Chips enabled: 2
Cores enabled: 8
Cores per chip: 4
Threads per core: 1
CPU Characteristics: Quad-Core AMD Opteron Processor 2350 (Barcelona)
CPU MHz: 2000
Primary Cache: 64 KB I + 64 KB D on chip per core
Secondary Cache: 512 KB I+D on chip per core
L3 Cache: 2 MB I+D on chip per chip
Other Cache: None
Memory: 16 GB (4 x 4 GB DIMMs, 667 MHz)
Disk Subsystem: 10K RPM Serial-Attach SCSI 3Gbps 2.5-in HotPlug
Hard Drive
Other Hardware: None
Adapter: Broadcom Corporation NetXtreme II BCM5708S
Gigabit Ethernet (rev 12)
Number of Adapters: 1
Slot Type: PCIe x8
Data Rate: 1 Gbps
Ports Used: 1
Interconnect Type: Gigabit Ethernet
Software
Adapter: Broadcom Corporation NetXtreme II BCM5708S
Gigabit Ethernet (rev 12)
Adapter Driver: OS default (bnx2, v1.5.11)
Adapter Firmware: 4.4.1
Operating System: RHEL 5.1 (x86_64)
Kernel 2.6.18-53.el5
Local File System: ext3
Shared File System: NFSv3
System State: multi-user
Other Software: None

Interconnect Description: Gigabit Ethernet

Hardware
Vendor: Dell Inc.
Model: 24 port Gigabit Ethernet switch
Switch Model: PowerConnect 5324
Number of Switches: 1
Number of Ports: 24
Data Rate: 1 Gbps Ethernet
Firmware: sw: v2.0.0.39, boot: v1.0.2.02, hw: 00.00.02
Topology: Single Switch
Primary Use: MPI and file system traffic

Submit Notes

The config file option 'submit' was used.

General Notes

Dell PowerConnect 5324 configuration details:

   Ports 9-16 were used and configured the following way:
                                                Flow Link          Back   Mdix
   Port     Type         Duplex  Speed Neg      ctrl State       Pressure Mode
   -------- ------------ ------  ----- -------- ---- ----------- -------- -------
   g9       1G-Copper    Full    1000  Enabled  On   Up          Disabled Off


Broadcom NetXtreme II BCM5708S Coalesce Settings:

   The following coalesce settings were applied on all nodes:
   ethtool -C eth0 rx-usecs 7 rx-frames 2 rx-usecs-irq 14 rx-frames-irq 4 tx-usecs 20 tx-frames 5 tx -usecs-irq 40 tx-frames-irq 10


NFS server usage:

   The first node was used as NFS server. The other 7 nodes mounted the working directory from the first node.


Base Compiler Invocation

C benchmarks:

 mpicc -ccl icc 

C++ benchmarks:

126.lammps:  mpicc -ccl icpc 

Fortran benchmarks:

 mpif77 -ccl ifort 

Benchmarks using both Fortran and C:

 mpicc -ccl icc   mpif77 -ccl ifort 

Base Portability Flags

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

Base Optimization Flags

C benchmarks:

 -O3   -ipo   -no-prec-div   -axS 

C++ benchmarks:

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

Fortran benchmarks:

 -O3   -ipo   -no-prec-div   -axS 

Benchmarks using both Fortran and C:

 -O3   -ipo   -no-prec-div   -axS 

The flags files that were used to format this result can be browsed at
http://www.spec.org/mpi2007/flags/MPI2007_flags.20081204.html,
http://www.spec.org/mpi2007/flags/EM64T_Intel101_flags.20081204.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/mpi2007/flags/MPI2007_flags.20081204.xml,
http://www.spec.org/mpi2007/flags/EM64T_Intel101_flags.20081204.xml.