SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

Cray

Cray XC30 (Intel Xeon E5-2697 v2)

SPECmpiM_peak2007 = Not Run

MPI2007 license: 3440A Test date: Mar-2017
Test sponsor: Indiana University Hardware Availability: Apr-2013
Tested by: Indiana University Software Availability: Feb-2017
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 45.5 34.4 44.7 35.0 42.4 36.9
107.leslie3d 768 93.2 56.0 92.6 56.4 95.6 54.6
113.GemsFDTD 768 529   11.9 530   11.9 532   11.9
115.fds4 768 161   12.1 159   12.2 160   12.2
121.pop2 768 288   14.3 296   14.0 279   14.8
122.tachyon 768 35.8 78.1 35.7 78.3 35.7 78.4
126.lammps 768 200   14.6 200   14.6 200   14.6
127.wrf2 768 83.1 93.8 81.2 96.0 83.7 93.1
128.GAPgeofem 768 90.2 22.9 89.4 23.1 90.2 22.9
129.tera_tf 768 48.0 57.7 47.5 58.2 50.4 54.9
130.socorro 768 102   37.3 101   37.7 102   37.4
132.zeusmp2 768 67.3 46.1 68.9 45.0 68.8 45.1
137.lu 768 60.0 61.3 62.7 58.7 62.8 58.5
Hardware Summary
Type of System: Homogeneous
Compute Node: Big Red II Plus Node
Interconnects: Infiniband (QDR)
Cray Aries
File Server Node: Data Capacitor II
Total Compute Nodes: 16
Total Chips: 32
Total Cores: 384
Total Threads: 768
Total Memory: 1 TB
Base Ranks Run: 768
Minimum Peak Ranks: --
Maximum Peak Ranks: --
Software Summary
C Compiler: Intel C Composer XE 2017 for Linux,
Version 17.0.2.174 Build 20170213
C++ Compiler: Intel C++ Composer XE 2017 for Linux,
Version 17.0.2.174 Build 20170213
Fortran Compiler: Intel Fortran Composer XE 2017 for Linux,
Version 17.0.2.174 Build 20170213
Base Pointers: 64-bit
Peak Pointers: 64-bit
MPI Library: Cray MPI (MPT) 7.5.0
Other MPI Info: None
Pre-processors: No
Other Software: None

Node Description: Big Red II Plus Node

Hardware
Number of nodes: 16
Uses of the node: compute
Vendor: Cray
Model: XC30
CPU Name: Intel Xeon E5-2697 v2
CPU(s) orderable: 1-2 chips
Chips enabled: 2
Cores enabled: 24
Cores per chip: 12
Threads per core: 2
CPU Characteristics: Intel Turbo Boost Technology disabled,
Hyper-Threading enabled
CPU MHz: 2700
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 256 KB I+D on chip per core
L3 Cache: 30 MB I+D on chip per chip
Other Cache: None
Memory: 64 GB (8 x 8 GB 2Rx4 PC3-14900R-13, ECC)
Disk Subsystem: None
Other Hardware: None
Adapter: Mellanox Technologies MT27500 ConnectX-3
Number of Adapters: 1
Slot Type: PCIe x16 Gen 3
Data Rate: 40Gbps
Ports Used: 1
Interconnect Type: 40 Gigabit Infiniband (QDR)
Adapter: Cray Aries
Number of Adapters: 1
Slot Type: PCIe x16 Gen 3
Data Rate: 126 Gbps
Ports Used: 4
Interconnect Type: Aries
Software
Adapter: Mellanox Technologies MT27500 ConnectX-3
Adapter Driver: 1.0-ofed1.5.4.1
Adapter Firmware: 2.33.5100
Adapter: Cray Aries
Adapter Driver: Proprietary Cray_kgni
Adapter Firmware: v004.r091
Operating System: SUSE Linux Enterprise Server 11 SP3 (x86_64),
Cray Linux Environment 5.2
3.0.101-0.46.1_1.0502.8871-cray_ari_c
Local File System: None
Shared File System: Lustre
System State: Multi-User
Other Software: Slurm 15.08.12

Node Description: Data Capacitor II

Hardware
Number of nodes: 2
Uses of the node: fileserver
Vendor: DDN
Model: DDN SFA12K
CPU Name: Intel Xeon CPU E5-2620
CPU(s) orderable: 1-2 chips
Chips enabled: 2
Cores enabled: 12
Cores per chip: 6
Threads per core: 1
CPU Characteristics: Intel Turbo Boost Technology up to 2.50 GHz
CPU MHz: 2000
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 256 KB I+D on chip per core
L3 Cache: 15 MB I+D on chip per chip
Other Cache: None
Memory: 96 GB
Disk Subsystem: 30 TB RAID 6, 10 (8 + 2) x 3 TB SAS
Hitachi HUS724030ALS640, 7200RPM, 6.0Gbps
Other Hardware: None
Adapter: Mellanox ConnectX MHQH29-XTC
Number of Adapters: 1
Slot Type: PCIe x8 Gen 2
Data Rate: 40Gbps
Ports Used: 1
Interconnect Type: 40 Gigabit Infiniband (QDR)
Software
Adapter: Mellanox ConnectX MHQH29-XTC
Adapter Driver: 1.0-ofed1.5.4.1
Adapter Firmware: 2.9.1000
Operating System: CentOS 6.2
Local File System: Linux/ext4
Shared File System: lustre
System State: Multi-User
Other Software: None

Interconnect Description: Infiniband (QDR)

Hardware
Vendor: DDN
Model: Mellanox SX6506
Switch Model: Mellanox SX6506
Number of Switches: 1
Number of Ports: 108
Data Rate: 56 Gbps
Firmware: mellanox SX6506
Topology: switched
Primary Use: Lustre fileserver

Interconnect Description: Cray Aries

Hardware
Vendor: Cray
Model: Cray Aries
Switch Model: Cray Aries
Number of Switches: 144
Number of Ports: 48
Data Rate: 126 Gb/s
Firmware: v004.r091
Topology: Dragonfly
Primary Use: MPI traffic

Submit Notes

The config file option 'submit' was used.
submit = srun -c 1 -n $ranks -q $command

General Notes

130.socorro (base): "nullify_ptrs" src.alt was used.

MPI startup command:
  srun command was used to start MPI jobs.

export MPICH_NO_BUFFER_ALIAS_CHECK=true
  If set, the buffer alias error check for collectives is
  disabled. The MPI standard does not allow aliasing of type
  OUT or INOUT parameters on the same collective function
  call. The default is false.

Job placement:
  Slurm is used for job placement.
  Compute nodes are selected by Slurm.
  No specific node selection is used.

Base Compiler Invocation

C benchmarks:

 cc 

C++ benchmarks:

126.lammps:  CC 

Fortran benchmarks:

 ftn 

Benchmarks using both Fortran and C:

 cc   ftn 

Base Portability Flags

121.pop2:  -DSPEC_MPI_CASE_FLAG 
126.lammps:  -DMPICH_IGNORE_CXX_SEEK 
127.wrf2:  -DSPEC_MPI_CASE_FLAG   -DSPEC_MPI_LINUX 
130.socorro:  -assume nostd_intent_in 

Base Optimization Flags

C benchmarks:

 -O3   -ansi-alias   -no-prec-div   -ipo   -xhost   -fp-model fast=2 

C++ benchmarks:

126.lammps:  -O3   -ansi-alias   -no-prec-div   -ipo   -xhost   -fp-model fast=2 

Fortran benchmarks:

 -O3   -ansi-alias   -no-prec-div   -ipo   -xhost   -fp-model fast=2 

Benchmarks using both Fortran and C:

 -O3   -ansi-alias   -no-prec-div   -ipo   -xhost   -fp-model fast=2 

The flags file that was used to format this result can be browsed at
http://www.spec.org/mpi2007/flags/EM64T_Intel170_flags.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/mpi2007/flags/EM64T_Intel170_flags.xml.