WRF 2.5 Kilometer CONUS Benchmarks
This is November 2008 edition.
|
|
|
|
Normal scales |
Log scales |
Click above for larger images (data in thumbnails is not representative).
2.5 KM CONUS Benchmark
The domain is 2.5KM horizontal resolution on a 1500 by 1200 grid with 35 vertical levels, with a time step of 15 seconds. The benchmark is restarted on June 4, 2008 06Z and run for 3 hours and then the time per timestep is averaged over all but the first time step, which is discarded. The average time per time step is divided by the operation count for an average time step, 411.7 billion floating point operations[1], to determine floating point rate. Return to the WRF V3 Benchmarks Page for additional information.
System Descriptions
SGI ICE 8200 EX (NASA)
Results submitted by Gerardo Cisneros, SGI Corp., Summer 2008
1. Name of system (product name, hostname, institution): SGI ICE 8200 EX, pleiades, NASA Ames
2. Model version of WRF: 3.0
3. Operating system and version: SUSE Enterprise Linux 10.
4. Compiler and version: Intel Fortran and C/C++ Compilers, version 10.1.017.
5. Processor manufacturer, type, and speed: Xeon E5472 (Harpertown) 3GHz quad-cores with 6MB L2 per core pair, Intel 5400 series chipset (Seaburg)
6. Cores per socket and sockets per node: 4/2
7. Main memory per core: 2 GB
8. Interconnect type, product name, topology: Infiniband, ConnectX HCAs, hypercube
9. Other information:
Endeavor Cluster (Intel)
Results submitted by Roman S. Dubtsov, Intel Corp. November, 2008
1. Name of system (product name, hostname, institution): Endeavor, Intel
2. Model version of WRF: 3.0.1
3. Operating system and version: RHEL 4U4
4. Compiler and version: Intel Fortran Compiler 10.1.015
5. Processor manufacturer, type, and speed: Intel Xeon E5462, 2.8 GHz 12MB L2 cache
6. Cores per socket and sockets per node: 4/2
7. Main memory per core: 16 GB FB DIMM
8. Interconnect type, product name, topology: DDR Infiniband, fat tree
9. Other information: Intel Xeon 5400 chipset, 1600 MT/s FSB
Blue Gene/L (NY Blue)
Results submitted by Andrew Vogelmann, Brookhaven National Laboratory, July 2008
1. Name of system (product name, hostname, institution): Blue Gene/L, NY Blue, Brookhaven NL/SUNY Stonybrook
2. Model version of WRF: 3.0
3. Operating system and version: SUSE Linux Enterprise Server 9 (Front End), Compute Node Kernel (Compute Nodes)
4. Compiler and version: IBM Fortran, -O2 optimization
5. Processor manufacturer, type, and speed: PowerPC 440, 700 MHz
6. Cores per socket and sockets per node: 1/2
7. Main memory per core: 1 GB (in co-processor mode, using 1 core per node)
8. Interconnect type, product name, topology: Proprietary, 3D-Torus
9. Other information: run in co-processor mode so only 1 core used per node
Created Nov. 9, 2008. John Michalakes, NCAR. michalak@ucar.edu