Platforms (hardware+compiler) used for the benchmarks

Back to main benchmark page

Machine Processor Parallelism Compiler
BGP-GCC Blue Gene / P 850 MHz PowerPC 450 quad-core Quad mode GCC 4.1.2 (-O3)
AMD-8356-GCC Multi-Proc AMD Opteron 8356 quad-core Shared memory / OpenMPI GCC 4.2.4 (-O3)
I-X5570-GCC Multi-Proc Intel Xeon X5570 quad-core Shared memory / Mpich-shmem GCC 4.3.2 (-O3)
I-X5550-GCC Cluster Intel Xeon X5550 quad-core Infinband / OpenMPI GCC 4.3.2 (-O3)
I-E5520-GCC Cluster / MP Intel Xeon E5520 quad-core Dual-Proc+Infiniband / OpenMPI / RR(*) GCC 4.3.2 (-O3)
I-E5520-ICC Cluster / MP Intel Xeon E5520 quad-core Dual-Proc+Infiniband / OpenMPI / RR(*) ICC 11.1 (-O3 -ip -xHOST)

(*) RR stands for “round-robin” mode: MPI uses one core on the first multi-processor, then a core on the next MP, and so on until all MPs are used, after which it starts exploiting the second core on each MP, and so on.

 
plb_wiki/benchmark/platforms.txt · Last modified: 2011/10/29 19:14 (external edit)
 
RSS - 2011 © FlowKit.com
Except where otherwise noted, text, pictures and animations on this site are licensed under a Creative Commons Attribution-Share Alike 3.0 License.