Intel Core i7-3820 testing with a ASRock X79 Extreme11 and 4 x NVIDIA GeForce GTX 570 HD on CentOS 6.3 (Final) via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1210317-MJEA-GTX570M06 gtx570-memory-bandwidth - Phoronix Test Suite gtx570-memory-bandwidth Intel Core i7-3820 testing with a ASRock X79 Extreme11 and 4 x NVIDIA GeForce GTX 570 HD on CentOS 6.3 (Final) via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1210317-MJEA-GTX570M06&grt&sor .
gtx570-memory-bandwidth Processor Motherboard Chipset Memory Disk Graphics Audio Network OS Kernel Compiler File-System 1 x GTX570 2 x GTX570 3 x GTX570 4 x GTX570 Intel Core i7-3820 @ 3.60GHz (4 Cores) ASRock X79 Extreme11 Intel Xeon E5/Core 8 x 4096 MB DDR3-1333MHz 128GB M4-CT128M4SSD2 NVIDIA GeForce GTX 570 HD Intel C600/X79 Broadcom NetLink BCM57781 Gigabit PCIe CentOS 6.3 (Final) 2.6.32-279.el6.x86_64 (x86_64) GCC 4.4.6 20120305 + CUDA 5.0 ext4 OpenBenchmarking.org Compiler Details - --build=x86_64-redhat-linux --disable-dssi --disable-libjava-multilib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-unique-object --enable-java-awt=gtk --enable-java-maintainer-mode --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-libgcj-multifile --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-cloog --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/jre --with-ppl --with-tune=generic Processor Details - Scaling Governor: ondemand
gtx570-memory-bandwidth cuda-bandwidthTest: Pageable memory - Device to host cuda-bandwidthTest: Pageable memory - Host to device cuda-bandwidthTest: Pageable memory - Device to device cuda-bandwidthTest: Non-pageable system memory - Device to host cuda-bandwidthTest: Non-pageable system memory - Host to device cuda-bandwidthTest: Non-pageable system memory - Device to device 1 x GTX570 2 x GTX570 3 x GTX570 4 x GTX570 5648.80 5940.80 131168.47 6317.27 6350.73 131157.00 11154.70 11886.77 262267.60 12631.97 12696.27 262320.30 16988.03 17831.50 393474.77 18952.60 19051.43 393469.87 22606.00 23749.57 524728.10 25267.30 25393.30 524609.13 OpenBenchmarking.org
CUDA bandwidthTest Memory type: Pageable memory - Memory transfer path: Device to host OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Pageable memory - Memory transfer path: Device to host 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 5K 10K 15K 20K 25K SE +/- 12.22, N = 3 SE +/- 14.77, N = 3 SE +/- 120.20, N = 3 SE +/- 8.71, N = 3 22606.00 16988.03 11154.70 5648.80 1. (CXX) g++ options: -m64 -lcudart
CUDA bandwidthTest Memory type: Pageable memory - Memory transfer path: Host to device OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Pageable memory - Memory transfer path: Host to device 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 5K 10K 15K 20K 25K SE +/- 7.34, N = 3 SE +/- 5.36, N = 3 SE +/- 4.88, N = 3 SE +/- 2.29, N = 3 23749.57 17831.50 11886.77 5940.80 1. (CXX) g++ options: -m64 -lcudart
CUDA bandwidthTest Memory type: Pageable memory - Memory transfer path: Device to device OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Pageable memory - Memory transfer path: Device to device 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 110K 220K 330K 440K 550K SE +/- 32.43, N = 3 SE +/- 23.87, N = 3 SE +/- 32.06, N = 3 SE +/- 10.66, N = 3 524728.10 393474.77 262267.60 131168.47 1. (CXX) g++ options: -m64 -lcudart
CUDA bandwidthTest Memory type: Non-pageable system memory - Memory transfer path: Device to host OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Non-pageable system memory - Memory transfer path: Device to host 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 5K 10K 15K 20K 25K SE +/- 1.57, N = 3 SE +/- 1.16, N = 3 SE +/- 0.64, N = 3 SE +/- 0.73, N = 3 25267.30 18952.60 12631.97 6317.27 1. (CXX) g++ options: -m64 -lcudart
CUDA bandwidthTest Memory type: Non-pageable system memory - Memory transfer path: Host to device OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Non-pageable system memory - Memory transfer path: Host to device 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 5K 10K 15K 20K 25K SE +/- 3.38, N = 3 SE +/- 1.47, N = 3 SE +/- 2.52, N = 3 SE +/- 0.67, N = 3 25393.30 19051.43 12696.27 6350.73 1. (CXX) g++ options: -m64 -lcudart
CUDA bandwidthTest Memory type: Non-pageable system memory - Memory transfer path: Device to device OpenBenchmarking.org MB/sec, More Is Better CUDA bandwidthTest 5.0 Memory type: Non-pageable system memory - Memory transfer path: Device to device 4 x GTX570 3 x GTX570 2 x GTX570 1 x GTX570 110K 220K 330K 440K 550K SE +/- 24.07, N = 3 SE +/- 20.10, N = 3 SE +/- 16.52, N = 3 SE +/- 3.75, N = 3 524609.13 393469.87 262320.30 131157.00 1. (CXX) g++ options: -m64 -lcudart
Phoronix Test Suite v10.8.4