| MPI2007 license: | 3440 | Test date: | Dec-2011 |
|---|---|---|---|
| Test sponsor: | Indiana University | Hardware Availability: | Jun-2010 |
| Tested by: | Huian Li | Software Availability: | Jan-2011 |
| 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 | 32 | 462 | 3.39 | 459 | 3.41 | 456 | 3.43 | |||||||
| 107.leslie3d | 32 | 1348 | 3.87 | 1442 | 3.62 | 1417 | 3.68 | |||||||
| 113.GemsFDTD | 32 | 915 | 6.90 | 1035 | 6.09 | 968 | 6.51 | |||||||
| 115.fds4 | 32 | 560 | 3.48 | 563 | 3.46 | 566 | 3.45 | |||||||
| 121.pop2 | 32 | 927 | 4.45 | 904 | 4.57 | 898 | 4.59 | |||||||
| 122.tachyon | 32 | 1002 | 2.79 | 1006 | 2.78 | 1005 | 2.78 | |||||||
| 126.lammps | 32 | 1093 | 2.67 | 1097 | 2.66 | 1105 | 2.64 | |||||||
| 127.wrf2 | 32 | 1074 | 7.26 | 1063 | 7.34 | 1044 | 7.46 | |||||||
| 128.GAPgeofem | 32 | 432 | 4.79 | 440 | 4.70 | 433 | 4.77 | |||||||
| 129.tera_tf | 32 | 829 | 3.34 | 830 | 3.34 | 832 | 3.33 | |||||||
| 130.socorro | 32 | 920 | 4.15 | 951 | 4.01 | 922 | 4.14 | |||||||
| 132.zeusmp2 | 32 | 735 | 4.22 | 743 | 4.18 | 829 | 3.74 | |||||||
| 137.lu | 32 | 993 | 3.70 | 977 | 3.76 | 998 | 3.69 | |||||||
| Hardware Summary | |
|---|---|
| Type of System: | Homogeneous |
| Compute Node: | Mason Node |
| Interconnects: | 10Gigabit Ethernet Gigabit Ethernet |
| File Server Node: | HOME |
| Total Compute Nodes: | 1 |
| Total Chips: | 4 |
| Total Cores: | 32 |
| Total Threads: | 32 |
| Total Memory: | 512 GB |
| Base Ranks Run: | 32 |
| Minimum Peak Ranks: | -- |
| Maximum Peak Ranks: | -- |
| Software Summary | |
|---|---|
| C Compiler: | Intel C Composer XE 2011 for Linux Version 12.0, Build 20110112 |
| C++ Compiler: | Intel C++ Composer XE 2011 for Linux Version 12.0, Build 20110112 |
| Fortran Compiler: | Intel Fortran Composer XE 2011 for Linux Version 12.0, Build 20110112 |
| Base Pointers: | 64-bit |
| Peak Pointers: | 64-bit |
| MPI Library: | OpenMPI-1.4.3 |
| Other MPI Info: | None |
| Pre-processors: | No |
| Other Software: | None |
| Hardware | |
|---|---|
| Number of nodes: | 1 |
| Uses of the node: | compute |
| Vendor: | HP |
| Model: | Proliant DL580 G7 Server Series |
| CPU Name: | Intel Xeon L7555 |
| CPU(s) orderable: | 1-4 chips |
| Chips enabled: | 4 |
| Cores enabled: | 32 |
| Cores per chip: | 8 |
| Threads per core: | 1 |
| CPU Characteristics: | Intel Turbo Boost Technology enabled, 5.86 GT/s QPI |
| CPU MHz: | 1866 |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 256 KB I+D on chip per core |
| L3 Cache: | 24 MB I+D on chip per chip, 24 MB shared / 8 cores |
| Other Cache: | None |
| Memory: | 512 GB (64 x 8 GB 2Rx4 PC3-10600R, ECC running at 1066 MHz and CL9) |
| Disk Subsystem: | Two 500 GB 7200 RPM 2.5" SAS hard drives,in RAID 1 mirror |
| Other Hardware: | None |
| Adapter: | HP NC375i 1G w/NC524SFP 10G Module |
| Number of Adapters: | 1 |
| Slot Type: | PCIe x8 Gen2 |
| Data Rate: | 10Gbps |
| Ports Used: | 1 |
| Interconnect Type: | 10 Gigabit Ethernet |
| Adapter: | HP NC375i 1G |
| Number of Adapters: | 1 |
| Slot Type: | PCIe x8 Gen2 |
| Data Rate: | 1Gbps |
| Ports Used: | 1 |
| Interconnect Type: | 1 Gigabit Ethernet |
| Software | |
|---|---|
| Adapter: | HP NC375i 1G w/NC524SFP 10G Module |
| Adapter Driver: | netxen_nic v 4.0.75 |
| Adapter Firmware: | 4.0.544 |
| Adapter: | HP NC375i 1G |
| Adapter Driver: | netxen_nic v 4.0.75 |
| Adapter Firmware: | 4.0.544 |
| Operating System: | RHEL6.0 (x86_64) 2.6.32-71.14.1.el6 Kernel 2.6.32-71.14.1.el6 |
| Local File System: | Linux/ext2 |
| Shared File System: | NFS |
| System State: | Multi-User |
| Other Software: | TORQUE-2.5.7 |
| Hardware | |
|---|---|
| Number of nodes: | 1 |
| Uses of the node: | fileserver |
| Vendor: | IBM |
| Model: | IBM N5500 NAS |
| CPU Name: | Intel Xeon CPU |
| CPU(s) orderable: | 1-4 chips |
| Chips enabled: | 4 |
| Cores enabled: | 32 |
| Cores per chip: | 8 |
| Threads per core: | 1 |
| CPU Characteristics: | -- |
| CPU MHz: | 1866 |
| Primary Cache: | 32 KB I + 32 KB D on chip per chip |
| Secondary Cache: | 256 KB I+D on chip per core |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 6 GB |
| Disk Subsystem: | 10 disks, 320GB/disk, 2.6TB total |
| Other Hardware: | None |
| Adapter: | Intel 82546GB Dual-Port Gigabit Ethernet Controller |
| Number of Adapters: | 1 |
| Slot Type: | PCI-Express x8 |
| Data Rate: | 1Gbps Ethernet |
| Ports Used: | 1 |
| Interconnect Type: | Ethernet |
| Software | |
|---|---|
| Adapter: | Intel 82546GB Dual-Port Gigabit Ethernet Controller |
| Adapter Driver: | e1000 |
| Adapter Firmware: | N/A |
| Operating System: | RedHat EL 4 Update 4 |
| Local File System: | None |
| Shared File System: | NFS |
| System State: | Multi-User |
| Other Software: | None |
| Hardware | |
|---|---|
| Vendor: | HP |
| Model: | HP NC375i 1G w/NC524SFP 10G Module |
| Switch Model: | Cisco 7018 (Line card module: N7K-M132XP-12) |
| Number of Switches: | 1 |
| Number of Ports: | 16 |
| Data Rate: | 10 Gbps Ethernet |
| Firmware: | EPLD 5.0.2 |
| Topology: | switched |
| Primary Use: | MPI traffic and NFS traffic |
| Hardware | |
|---|---|
| Vendor: | HP |
| Model: | Cisco SGE2010 |
| Switch Model: | Cisco SGE2010 |
| Number of Switches: | 1 |
| Number of Ports: | 48 |
| Data Rate: | 1 Gbps Ethernet |
| Firmware: | 3.0.0.18 |
| Topology: | switched |
| Primary Use: | Network management |
The config file option 'submit' was used.
MPI startup command: mpirun command was used to start MPI jobs. eth0 (10 GigE) was specified at the mpirun command line for MPI message passing eth3 (1 GigE) was specified for non-MPI communication. BIOS settings: Intel Turbo Boost Technology (Turbo) : Enabled (the default) RAM configuration: Each compute node has 64x8-GB RDIMMs. Network: Four compute nodes connect to one Cisco Nexus 7018 switch via 10 GigE port. Job placement: Each MPI job was assigned to a topologically compact set of nodes, i.e. the minimal needed number of compute nodes was used for each job: 1 compute node for 32 ranks, 2 for 64 ranks, 4 for 128 ranks, and 8 for 256 ranks PBS Pro was used for job submission. It has no impact on performance. Can be found at: http://www.altair.com
| mpicc |
| 126.lammps: | mpicxx |
| mpif90 |
| mpicc mpif90 |
| 121.pop2: | -DSPEC_MPI_CASE_FLAG |
| 126.lammps: | -DMPICH_IGNORE_CXX_SEEK |
| 127.wrf2: | -DSPEC_MPI_LINUX -DSPEC_MPI_CASE_FLAG |