SPEC® MPIM2007 Result

SGI

SGI ICE X
(Intel Xeon E5-2690 v3, 2.6 GHz)

SPECmpiM_peak2007 = 115

SPECmpiM_base2007 = 112

MPI2007 license:	14	Test date:	Jul-2014
Test sponsor:	SGI	Hardware Availability:	Sep-2014
Tested by:	SGI	Software Availability:	Apr-2014

Results Table

Benchmark	Base							Peak
Benchmark	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	800	10.7	146	10.7	146	10.7	146	800	10.7	146	10.7	146	10.7	146
107.leslie3d	800	28.8	181	28.8	182	27.9	187	800	28.8	181	28.8	182	27.9	187
113.GemsFDTD	800	219	28.8	219	28.8	219	28.8	160	146	43.2	146	43.2	146	43.1
115.fds4	800	10.7	182	10.9	179	10.9	179	800	10.7	182	10.9	179	10.9	179
121.pop2	800	74.3	55.5	74.3	55.5	74.5	55.4	800	74.3	55.5	74.3	55.5	74.5	55.4
122.tachyon	800	22.8	123	22.8	123	22.8	123	800	22.8	123	22.8	123	22.8	123
126.lammps	800	99.1	29.4	101	28.8	99.0	29.4	800	99.1	29.4	101	28.8	99.0	29.4
127.wrf2	800	29.6	263	29.6	263	29.4	265	800	29.6	263	29.6	263	29.4	265
128.GAPgeofem	800	9.04	228	9.18	225	9.21	224	800	9.04	228	9.18	225	9.21	224
129.tera_tf	800	23.2	119	23.2	119	23.3	119	800	23.2	119	23.2	119	23.3	119
130.socorro	800	35.3	108	35.3	108	35.2	108	800	35.3	108	35.3	108	35.2	108
132.zeusmp2	800	24.2	128	24.1	129	24.2	128	800	24.2	128	24.1	129	24.2	128
137.lu	800	23.7	155	23.7	155	23.6	156	800	23.7	155	23.7	155	23.6	156

Hardware Summary
Type of System:	Homogeneous
Compute Node:	SGI ICE X IP-131 Compute Node
Interconnect:	InfiniBand (MPI and I/O)
File Server Node:	SGI Rackable C1103-TY12
Total Compute Nodes:	40
Total Chips:	80
Total Cores:	960
Total Threads:	960
Total Memory:	5 TB
Base Ranks Run:	800
Minimum Peak Ranks:	160
Maximum Peak Ranks:	800

Software Summary
C Compiler:	Intel C++ Composer XE 2013 for Linux, Version 14.0.3.174 Build 20140422
C++ Compiler:	Intel C++ Composer XE 2013 for Linux Version 14.0.3.174 Build 20140422
Fortran Compiler:	Intel Fortran Composer XE 2013 for Linux, Version 14.0.3.174 Build 20140422
Base Pointers:	64-bit
Peak Pointers:	Not Applicable
MPI Library:	SGI MPT 2.09 Patch 11049
Other MPI Info:	OFED 1.5.4
Pre-processors:	None
Other Software:	None

Node Description: SGI ICE X IP-131 Compute Node

Hardware
Number of nodes:	40
Uses of the node:	compute
Vendor:	SGI
Model:	SGI ICE X (Intel Xeon E6-2690 v3, 2.6 GHz)
CPU Name:	Intel Xeon E5-2690 v3
CPU(s) orderable:	1-2 chips
Chips enabled:	2
Cores enabled:	24
Cores per chip:	12
Threads per core:	1
CPU Characteristics:	12 Core, 2.60 GHz, 9.6 GT/s QPI Intel Turbo Boost Technology up to 3.50 GHz Hyper-Threading Technology disabled
CPU MHz:	2600
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:	128 GB (8 x 16 GB 2Rx4 PC4-17000R-15, ECC)
Disk Subsystem:	None
Other Hardware:	None
Adapter:	Mellanox MT27500 with ConnectX-3 ASIC (PCIe x8 Gen3 8 GT/s)
Number of Adapters:	2
Slot Type:	PCIe x8 Gen3
Data Rate:	InfiniBand 4x FDR
Ports Used:	2
Interconnect Type:	InfiniBand

Software
Adapter:	Mellanox MT27500 with ConnectX-3 ASIC (PCIe x8 Gen3 8 GT/s)
Adapter Driver:	OFED-1.5.4
Adapter Firmware:	2.30.3000
Operating System:	SUSE Linux Enterprise Server 11 SP3 (x86_64), Kernel 3.0.93-0.8-default
Local File System:	NFSv3
Shared File System:	NFSv3 IPoIB
System State:	Multi-user, run level 3
Other Software:	SGI Tempo Service Node 2.8.1, Build 709rp49.sles11sp3-1402182002

Node Description: SGI Rackable C1103-TY12

Hardware
Number of nodes:	1
Uses of the node:	fileserver
Vendor:	SGI
Model:	SGI Rackable C1103-TY12 (Intel Xeon X5670, 2.93 GHz)
CPU Name:	Intel Xeon X5670
CPU(s) orderable:	1-2 chips
Chips enabled:	2
Cores enabled:	12
Cores per chip:	6
Threads per core:	2
CPU Characteristics:	Intel Turbo Boost Technology up to 3.33 GHz Hyper-Threading Technology enabled
CPU MHz:	2933
Primary Cache:	32 KB I + 32 KB D on chip per core
Secondary Cache:	256 KB I+D on chip per chip
L3 Cache:	12 MB I+D on chip per chip
Other Cache:	None
Memory:	96 GB (12 * 8 GB 2Rx4 PC3-10600R-9, ECC)
Disk Subsystem:	12 TB RAID 6 12 x 1 TB SATA (Seagate Constellation, 7200RPM)
Other Hardware:	None
Adapter:	Mellanox MT27500 with ConnectX-3 ASIC (PCIe x8 Gen3 8 GT/s)
Number of Adapters:	2
Slot Type:	PCIe x8 Gen3
Data Rate:	InfiniBand 4x FDR
Ports Used:	2
Interconnect Type:	InfiniBand

Software
Adapter:	Mellanox MT27500 with ConnectX-3 ASIC (PCIe x8 Gen3 8 GT/s)
Adapter Driver:	OFED-1.5.2
Adapter Firmware:	2.30.3000
Operating System:	SUSE Linux Enterprise Server 11 SP1 (x86_64), Kernel 2.6.32.46-0.3-default
Local File System:	xfs
Shared File System:	--
System State:	Multi-user, run level 3
Other Software:	SGI Foundation Software 2.5, Build 705r10.sles11-1110192111

Interconnect Description: InfiniBand (MPI and I/O)

Hardware
Vendor:	Mellanox Technologies and SGI
Model:	None
Switch Model:	SGI FDR Integrated IB Switch Blade 2SW9x27 with Mellanox SwitchX device 51000
Number of Switches:	10
Number of Ports:	36
Data Rate:	InfiniBand 4x FDR
Firmware:	09.02.3000
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_IB_RAILS=2
   export MPI_CONNECTIONS_THRESHOLD=0
   ulimit -s unlimited

 BIOS settings:
   AMI BIOS version DY2E6044
   Hyper-Threading Technology disabled
   Intel Turbo Boost Technology enabled (default)
   Intel Turbo Boost Technology activated with
     modprobe acpi_cpufreq
     cpupower frequency-set -u 2601MHz -d 2601MHz -g performance

 Job Placement:
   Ten ranks were assigned to each CPU chip, leaving 2
   cores per chip idle. There were 10 switches used
   with a topologically compact configuration.

 Additional notes regarding interconnect:
   The Infiniband network consists of two independent planes,
   with half the switches in the system allocated to each plane.
   I/O traffic is restricted to one plane, while MPI traffic can
   use both planes.

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