SPEC(R) ACCEL(TM) OMP Summary Intel Intel Xeon Phi 7210 Intel Server System LADMP00AP Family (Xeon Phi 7210, 1.3 GHz, 64 cores, 4 threads) Test Sponsor: Technische Universitaet Dresden Mon Jul 24 04:45:33 2017 ACCEL License: 37A Test date: Jul-2017 Test sponsor: Technische Universitaet Dresden Hardware availability: Jun-2016 Tested by: Technische Universitaet Dresden Software availability: Dec-2016 Base Base Base Peak Peak Peak Benchmarks Ref. Run Time Ratio Ref. Run Time Ratio -------------- ------ --------- --------- ------ --------- --------- 503.postencil 109 43.9 2.48 * 109 35.9 3.04 S 503.postencil 109 44.1 2.47 S 109 38.6 2.82 S 503.postencil 109 43.3 2.51 S 109 36.0 3.03 * 504.polbm 122 23.1 5.27 S 122 21.6 5.65 * 504.polbm 122 22.6 5.40 * 122 22.1 5.52 S 504.polbm 122 22.5 5.42 S 122 21.5 5.67 S 514.pomriq 621 294 2.11 S 621 233 2.67 S 514.pomriq 621 293 2.12 * 621 230 2.70 S 514.pomriq 621 291 2.13 S 621 230 2.70 * 550.pmd 241 77.9 3.09 S 241 55.0 4.38 * 550.pmd 241 77.5 3.11 * 241 55.1 4.38 S 550.pmd 241 77.4 3.12 S 241 54.9 4.39 S 551.ppalm 544 4660 0.117 S 544 251 2.17 * 551.ppalm 544 4654 0.117 * 544 251 2.17 S 551.ppalm 544 4645 0.117 S 544 252 2.16 S 552.pep 231 97.4 2.37 * 231 61.6 3.75 S 552.pep 231 97.4 2.37 S 231 61.4 3.76 S 552.pep 231 97.5 2.37 S 231 61.5 3.76 * 553.pclvrleaf 1145 103 11.1 * 1145 113 10.1 S 553.pclvrleaf 1145 102 11.2 S 1145 113 10.2 S 553.pclvrleaf 1145 103 11.1 S 1145 113 10.1 * 554.pcg 333 210 1.59 S 333 195 1.71 S 554.pcg 333 211 1.58 * 333 195 1.70 * 554.pcg 333 212 1.57 S 333 195 1.70 S 555.pseismic 282 50.6 5.57 S 282 52.2 5.41 S 555.pseismic 282 51.2 5.51 * 282 52.3 5.40 * 555.pseismic 282 51.3 5.50 S 282 52.3 5.39 S 556.psp 818 42.5 19.2 * 818 38.8 21.1 S 556.psp 818 42.5 19.3 S 818 38.4 21.3 S 556.psp 818 42.8 19.1 S 818 38.6 21.2 * 557.pcsp 859 56.1 15.3 S 859 46.3 18.5 S 557.pcsp 859 55.6 15.4 S 859 44.4 19.3 S 557.pcsp 859 55.9 15.4 * 859 45.3 19.0 * 559.pmniGhost 397 68.0 5.84 * 397 68.4 5.81 * 559.pmniGhost 397 67.9 5.85 S 397 68.3 5.81 S 559.pmniGhost 397 68.4 5.80 S 397 68.7 5.78 S 560.pilbdc 653 76.6 8.53 S 653 68.1 9.58 * 560.pilbdc 653 76.8 8.50 * 653 67.8 9.63 S 560.pilbdc 653 77.7 8.41 S 653 68.8 9.49 S 563.pswim 159 32.8 4.85 * 159 29.4 5.41 S 563.pswim 159 32.9 4.83 S 159 29.1 5.46 S 563.pswim 159 32.8 4.85 S 159 29.2 5.44 * 570.pbt 780 30.8 25.3 * 780 25.5 30.6 S 570.pbt 780 30.7 25.4 S 780 25.5 30.6 * 570.pbt 780 30.9 25.2 S 780 25.6 30.5 S ============================================================================== 503.postencil 109 43.9 2.48 * 109 36.0 3.03 * 504.polbm 122 22.6 5.40 * 122 21.6 5.65 * 514.pomriq 621 293 2.12 * 621 230 2.70 * 550.pmd 241 77.5 3.11 * 241 55.0 4.38 * 551.ppalm 544 4654 0.117 * 544 251 2.17 * 552.pep 231 97.4 2.37 * 231 61.5 3.76 * 553.pclvrleaf 1145 103 11.1 * 1145 113 10.1 * 554.pcg 333 211 1.58 * 333 195 1.70 * 555.pseismic 282 51.2 5.51 * 282 52.3 5.40 * 556.psp 818 42.5 19.2 * 818 38.6 21.2 * 557.pcsp 859 55.9 15.4 * 859 45.3 19.0 * 559.pmniGhost 397 68.0 5.84 * 397 68.4 5.81 * 560.pilbdc 653 76.8 8.50 * 653 68.1 9.58 * 563.pswim 159 32.8 4.85 * 159 29.2 5.44 * 570.pbt 780 30.8 25.3 * 780 25.5 30.6 * SPECaccel_omp_base 4.39 SPECaccel_omp_peak 6.08 HARDWARE -------- CPU Name: Intel Xeon Phi 7210 CPU Characteristics: Intel Turbo Boost 2 Technology up to 1.50 GHz CPU MHz: 1300 CPU MHz Maximum: 1500 FPU: None CPU(s) enabled: 64 cores, 1 chip, 64 cores/chip, 4 threads/core CPU(s) orderable: 1 chip Primary Cache: 32 KB I + 32 KB D on chip per core Secondary Cache: 1 MB I+D on chip per 2 cores L3 Cache: 16 GB I+D on chip per chip Other Cache: None Memory: 96 GB (6 x 16 GB 2Rx4 PC4-2400T-R, running at 1066 MHz) Disk Subsystem: 275 GB INTEL SSDSC2BB30 Other Hardware: -- ACCELERATOR ----------- Accel Model Name: Xeon Phi 7210 Accel Vendor: Intel Accel Name: Intel Xeon Phi 7210 Type of Accel: CPU Accel Connection: N/A Does Accel Use ECC: yes Accel Description: Intel Xeon Phi 7210, SMT ON, Turbo ON Cluster Mode: Quadrant, Memory Mode: Cache Accel Driver: SOFTWARE -------- Operating System: CentOS Linux release 7.3 3.10.0-514.21.2.el7.x86_64 Compiler: Intel Compiler C/C++/Fortran Version 17.0.1 20161005 File System: ext4 System State: Run level 3 (user-level) Other Software: FFTW 3.3.6pl1 Submit Notes ------------ The config file option 'submit' was used. submit = numactl -p 0 $command Platform Notes -------------- Sysinfo program /tmp/spec-accel/1.2/Docs/sysinfo $Rev: 6965 $ $Date:: 2015-04-21 #$ c05a7f14b1b1765e3fe1df68447e8a35 running on taurusknl28.taurus.hrsk.tu-dresden.de Mon Jul 24 10:45:36 2017 This section contains SUT (System Under Test) info as seen by some common utilities. To remove or add to this section, see: http://www.spec.org/accel/Docs/config.html#sysinfo From /proc/cpuinfo model name : Intel(R) Xeon Phi(TM) CPU 7210 @ 1.30GHz 1 "physical id"s (chips) 256 "processors" cores, siblings (Caution: counting these is hw and system dependent. The following excerpts from /proc/cpuinfo might not be reliable. Use with caution.) cpu cores : 64 siblings : 256 physical 0: cores 0 1 2 3 6 7 10 11 12 13 14 15 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 56 57 58 59 60 61 62 63 64 65 68 69 70 71 72 73 cache size : 1024 KB From /proc/meminfo MemTotal: 98707216 kB HugePages_Total: 0 Hugepagesize: 2048 kB /usr/bin/lsb_release -d CentOS Linux release 7.3.1611 (Core) From /etc/*release* /etc/*version* centos-release: CentOS Linux release 7.3.1611 (Core) centos-release-upstream: Derived from Red Hat Enterprise Linux 7.3 (Source) os-release: NAME="CentOS Linux" VERSION="7 (Core)" ID="centos" ID_LIKE="rhel fedora" VERSION_ID="7" PRETTY_NAME="CentOS Linux 7 (Core)" ANSI_COLOR="0;31" CPE_NAME="cpe:/o:centos:centos:7" redhat-release: CentOS Linux release 7.3.1611 (Core) system-release: CentOS Linux release 7.3.1611 (Core) system-release-cpe: cpe:/o:centos:centos:7 uname -a: Linux taurusknl28.taurus.hrsk.tu-dresden.de 3.10.0-514.21.2.el7.x86_64 #1 SMP Tue Jun 20 12:24:47 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux run-level 3 Jun 30 13:46 SPEC is set to: /tmp/spec-accel/1.2 Filesystem Type Size Used Avail Use% Mounted on /dev/sda1 ext4 275G 6.5G 255G 3% / Additional information from dmidecode: Warning: Use caution when you interpret this section. The 'dmidecode' program reads system data which is "intended to allow hardware to be accurately determined", but the intent may not be met, as there are frequent changes to hardware, firmware, and the "DMTF SMBIOS" standard. (End of data from sysinfo program) General Notes ------------- Used Environment Variables: ENV_KMP_AFFINITY=compact,0 - assign OpenMP Threads continously ENV_OMP_NUM_THREADS=128 - limits number of Threads to be started to 128 ENV_KMP_HW_SUBSET=1S,64C,2T - control Thread distribution accross sockets, cores and hw threads ENV_FORT_BUFFERED=true - enables buffered I/O for Fortran ENV_OMP_DYNAMIC - Enable or disable the dynamic adjustment of the number of threads within a team. If undefined, dynamic adjustment is disabled by default. ENV_KMP_LIBRARY - Selects the OpenMP runtime library throughput. The options for the variable value are: serial, turnaround, or throughput indicating the execution mode. ENV_KMP_BLOCKTIME - Sets the time, in milliseconds, that a thread should wait, after completing the execution of a parallel region, before sleeping. Base Compiler Invocation ------------------------ C benchmarks: icc Fortran benchmarks: ifort Benchmarks using both Fortran and C: icc ifort Base Portability Flags ---------------------- 503.postencil: -DSPEC_USE_INNER_SIMD 504.polbm: -DSPEC_USE_INNER_SIMD 514.pomriq: -DSPEC_USE_INNER_SIMD 550.pmd: -DSPEC_USE_INNER_SIMD -80 551.ppalm: -DSPEC_USE_INNER_SIMD 552.pep: -DSPEC_USE_INNER_SIMD 553.pclvrleaf: -DSPEC_USE_INNER_SIMD 554.pcg: -DSPEC_USE_INNER_SIMD 555.pseismic: -DSPEC_USE_INNER_SIMD 556.psp: -DSPEC_USE_INNER_SIMD 557.pcsp: -DSPEC_USE_INNER_SIMD 559.pmniGhost: -DSPEC_USE_INNER_SIMD -nofor_main 560.pilbdc: -DSPEC_USE_INNER_SIMD 563.pswim: -DSPEC_USE_INNER_SIMD 570.pbt: -DSPEC_USE_INNER_SIMD Base Optimization Flags ----------------------- C benchmarks: -O3 -g -qopenmp -xMIC-AVX512 -qopenmp-offload=host Fortran benchmarks: -O3 -g -qopenmp -xMIC-AVX512 -qopenmp-offload=host Benchmarks using both Fortran and C: -O3 -g -qopenmp -xMIC-AVX512 -qopenmp-offload=host Peak Compiler Invocation ------------------------ C benchmarks: icc Fortran benchmarks: ifort Benchmarks using both Fortran and C: icc ifort Peak Portability Flags ---------------------- 503.postencil: -DSPEC_USE_INNER_SIMD 504.polbm: -DSPEC_USE_INNER_SIMD 514.pomriq: -DSPEC_USE_INNER_SIMD 550.pmd: -DSPEC_USE_INNER_SIMD -80 551.ppalm: -DSPEC_USE_INNER_SIMD -DSPEC_HOST_FFTW3 552.pep: -DSPEC_USE_INNER_SIMD 553.pclvrleaf: -DSPEC_USE_INNER_SIMD 554.pcg: -DSPEC_USE_INNER_SIMD 555.pseismic: -DSPEC_USE_INNER_SIMD 556.psp: -DSPEC_USE_INNER_SIMD 557.pcsp: -DSPEC_USE_INNER_SIMD 559.pmniGhost: -DSPEC_USE_INNER_SIMD -nofor_main 560.pilbdc: -DSPEC_USE_INNER_SIMD 563.pswim: -DSPEC_USE_INNER_SIMD 570.pbt: -DSPEC_USE_INNER_SIMD Peak Optimization Flags ----------------------- C benchmarks: 503.postencil: -O3 -xCORE-AVX2 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=3 504.polbm: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=5 514.pomriq: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=2 552.pep: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-streaming-stores always 554.pcg: -O3 -xCORE-AVX2 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=2 -qopt-streaming-stores always 557.pcsp: Same as 504.polbm 570.pbt: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host Fortran benchmarks: 550.pmd: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=3 -no-prec-div -fimf-precision=low 551.ppalm: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -no-prec-sqrt -I/sw/taurus/libraries/fftw/3.3.6pl1-gcc5.3-intelmpi5.1/include -L/sw/taurus/libraries/fftw/3.3.6pl1-gcc5.3-intelmpi5.1/lib 555.pseismic: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host 556.psp: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=2 560.pilbdc: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=3 563.pswim: Same as 555.pseismic Benchmarks using both Fortran and C: 553.pclvrleaf: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-streaming-stores always 559.pmniGhost: -O3 -xMIC-AVX512 -g -qopenmp -qopenmp-offload=host -qopt-prefetch=3 -qopt-streaming-stores always Peak Other Flags ---------------- Fortran benchmarks: 551.ppalm: -lfftw3 The flags file that was used to format this result can be browsed at https://www.spec.org/accel/flags/icc2015-openmp.html You can also download the XML flags source by saving the following link: https://www.spec.org/accel/flags/icc2015-openmp.xml SPEC and SPEC ACCEL are registered trademarks or trademarks of the Standard Performance Evaluation Corporation. All other brand and product names appearing in this result are trademarks or registered trademarks of their respective holders. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- For questions about this result, please contact the tester. For other inquiries, please contact webmaster@spec.org. Copyright 2015-2017 Standard Performance Evaluation Corporation Tested with SPEC ACCEL v1.2. Report generated on Wed Aug 16 15:42:44 2017 by ACCEL ASCII formatter v1290. Originally published on 16 August 2017.