| MPI2007 license: | 4 | Test date: | Apr-2010 |
|---|---|---|---|
| Test sponsor: | SGI | Hardware Availability: | May-2010 |
| Tested by: | SGI | Software Availability: | Feb-2010 |
| 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. | ||||||||||||||
| 121.pop2 | 3072 | 98.1 | 39.7 | 87.0 | 44.7 | 87.4 | 44.5 | 2048 | 63.1 | 61.7 | 63.5 | 61.2 | 63.4 | 61.3 |
| 122.tachyon | 3072 | 192 | 10.1 | 124 | 15.7 | 122 | 16.0 | 3072 | 193 | 10.1 | 124 | 15.6 | 125 | 15.5 |
| 125.RAxML | 3072 | 125 | 23.4 | 124 | 23.5 | 124 | 23.5 | 3072 | 125 | 23.4 | 125 | 23.4 | 126 | 23.2 |
| 126.lammps | 3072 | 46.3 | 53.1 | 42.3 | 58.2 | 42.4 | 58.0 | 3072 | 42.2 | 58.2 | 42.3 | 58.1 | 42.4 | 57.9 |
| 128.GAPgeofem | 3072 | 124 | 47.7 | 125 | 47.3 | 124 | 47.9 | 2048 | 112 | 53.1 | 111 | 53.6 | 111 | 53.7 |
| 129.tera_tf | 3072 | 51.2 | 21.5 | 51.0 | 21.5 | 51.3 | 21.4 | 3072 | 51.1 | 21.5 | 50.5 | 21.8 | 50.8 | 21.6 |
| 132.zeusmp2 | 3072 | 46.0 | 46.1 | 45.9 | 46.2 | 45.8 | 46.3 | 2048 | 34.7 | 61.1 | 34.6 | 61.2 | 34.8 | 61.0 |
| 137.lu | 3072 | 47.0 | 89.3 | 47.1 | 89.2 | 47.2 | 89.1 | 2048 | 34.1 | 123 | 34.1 | 123 | 34.3 | 122 |
| 142.dmilc | 3072 | 32.0 | 115 | 31.8 | 116 | 31.7 | 116 | 3072 | 31.8 | 116 | 31.9 | 116 | 32.1 | 115 |
| 143.dleslie | 3072 | 916 | 3.38 | 917 | 3.38 | 917 | 3.38 | 2048 | 37.6 | 82.4 | 36.6 | 84.7 | 38.6 | 80.2 |
| 145.lGemsFDTD | 3072 | 99.5 | 44.3 | 99.4 | 44.4 | 99.5 | 44.3 | 2048 | 84.0 | 52.5 | 83.2 | 53.0 | 83.7 | 52.7 |
| 147.l2wrf2 | 3072 | 180 | 45.7 | 162 | 50.6 | 162 | 50.6 | 2048 | 143 | 57.2 | 138 | 59.3 | 136 | 60.4 |
| Hardware Summary | |
|---|---|
| Type of System: | Homogeneous |
| Compute Node: | SGI Altix ICE 8400EX Compute Node |
| Interconnects: | InfiniBand (MPI) InfiniBand (I/O) |
| File Server Node: | SGI InfiniteStorage Nexis 2000 NAS |
| Total Compute Nodes: | 256 |
| Total Chips: | 512 |
| Total Cores: | 3072 |
| Total Threads: | 6144 |
| Total Memory: | 6 TB |
| Base Ranks Run: | 3072 |
| Minimum Peak Ranks: | 2048 |
| Maximum Peak Ranks: | 3072 |
| Software Summary | |
|---|---|
| C Compiler: | Intel C Compiler for Linux Version 11.1, Build 20100203 |
| C++ Compiler: | Intel C++ Compiler for Linux Version 11.1, Build 20100203 |
| Fortran Compiler: | Intel Fortran Compiler for Linux Version 11.1, Build 20100203 |
| Base Pointers: | 64-bit |
| Peak Pointers: | 64-bit |
| MPI Library: | SGI MPT 1.26 |
| Other MPI Info: | OFED 1.4.1 |
| Pre-processors: | None |
| Other Software: | None |
| Hardware | |
|---|---|
| Number of nodes: | 256 |
| Uses of the node: | compute |
| Vendor: | SGI |
| Model: | SGI Altix ICE 8400EX (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 core |
| L3 Cache: | 12 MB I+D on chip per chip |
| Other Cache: | None |
| Memory: | 24 GB (6*4GB DDR3-1333 CL9 RDIMMs) |
| 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-1.4.1 |
| Adapter Firmware: | 2.7.0 |
| Operating System: | SUSE Linux Enterprise Server 10 (x86_64) SP3 Kernel 2.6.16.60-0.54.5-smp |
| Local File System: | NFSv3 |
| Shared File System: | NFSv3 IPoIB |
| System State: | Multi-user, run level 3 |
| Other Software: | SGI ProPack 6 for Linux Service Pack 6, SGI Tempo V 1.10 |
| Hardware | |
|---|---|
| Number of nodes: | 1 |
| Uses of the node: | fileserver |
| Vendor: | SGI |
| Model: | SGI Altix XE 240 (Intel Xeon 5140, 2.33 GHz) |
| CPU Name: | Intel Xeon 5140 |
| CPU(s) orderable: | 1-2 chips |
| Chips enabled: | 2 |
| Cores enabled: | 4 |
| Cores per chip: | 2 |
| Threads per core: | 1 |
| CPU Characteristics: | 1333 MHz FSB |
| CPU MHz: | 2333 |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 4 MB I+D on chip per chip |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 16 GB (8*2GB DDR2-667MHz DIMMS) |
| Disk Subsystem: | 4.3 TB RAID 5 48 x 146 GB SAS (Seagate Cheetah 15K.5) |
| 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-1.3 |
| Adapter Firmware: | 5.3.0 |
| Operating System: | SUSE Linux Enterprise Server 10 (x86_64) SP1 Kernel 2.6.16.54-0.2.5-smp |
| Local File System: | xfs |
| Shared File System: | -- |
| System State: | Multi-user, run level 5 |
| Other Software: | SGI ProPack 5 for Linux Service Pack 5 |
| Hardware | |
|---|---|
| Vendor: | Mellanox Technologies |
| Model: | MT26428 ConnectX |
| Switch Model: | Mellanox MT48436 InfiniScale-IV |
| Number of Switches: | 128 |
| Number of Ports: | 36 |
| Data Rate: | InfiniBand 4x QDR |
| Firmware: | 5030004 |
| Topology: | Bristle hypercube |
| Primary Use: | MPI traffic |
| Hardware | |
|---|---|
| Vendor: | Mellanox Technologies |
| Model: | MT26428 ConnectX |
| Switch Model: | Mellanox MT48436 InfiniScale-IV |
| Number of Switches: | 64 |
| Number of Ports: | 36 |
| Data Rate: | InfiniBand 4x QDR |
| Firmware: | 5030004 |
| Topology: | Bristle hypercube |
| Primary Use: | I/O traffic |
The config file option 'submit' was used. For peak benchmarks that used 2048 MPI ranks, four ranks were assigned to each CPU chip, leaving 2 cores per chip idle.
Software environment:
export MPI_REQUEST_MAX=65536
export MPI_TYPE_MAX=32768
export MPI_BUFS_THRESHOLD=1
export MPI_DSM_DISTRIBUTE=yes
export MPI_IB_RAILS=2
ulimit -s unlimited
BIOS settings:
AMI BIOS version 8.16
Hyper-Threading Technology enabled (default)
Intel Turbo Boost Technology enabled (default)
Intel Turbo Boost Technology activated in the OS via
/etc/init.d/acpid start
/etc/init.d/powersaved start
powersave -f
Job Placement:
Each MPI job was assigned to a topologically compact set
of nodes, i.e. the minimal needed number of switches was
used for each job: 2 switches for 96 ranks,
4 switches for 192 ranks, 8 switches for 384 ranks,
16 switches for 768 ranks, 32 switches for 1536 ranks,
64 switches for 3072 ranks.
| icc |
| 126.lammps: | icpc |
| ifort |
| icc ifort |
| 121.pop2: | -DSPEC_MPI_CASE_FLAG |