SPEC(R) ACCEL_OCL Summary SGI NVIDIA Tesla C2070 SGI C3108-TY11 Test Sponsor: Technische Universitaet Dresden Wed Feb 26 14:47:27 2014 ACCEL License: 37A Test date: Feb-2014 Test sponsor: Technische Universitaet Dresden Hardware availability: May-2011 Tested by: Technische Universitaet Dresden Software availability: Feb-2012 Base Base Base Base Base Base Base Peak Peak Peak Peak Peak Peak Peak Benchmarks Ref. RunTime Ratio Energy MaxPwr AvgPwr ERatio Ref. RunTime Ratio Energy MaxPwr AvgPwr ERatio -------------- ------ ------- ------ ------ ------ ------ ------ ------ ------- ------ ------ ------ ------ ------ 101.tpacf 107 108 0.993 32.8 318 304 0.995 S 101.tpacf 107 107 1.00 32.8 319 307 0.995 S 101.tpacf 107 107 0.997 33.0 319 308 0.988 * 103.stencil 125 131 0.958 47.2 371 362 0.959 S 103.stencil 125 130 0.963 47.1 371 363 0.961 * 103.stencil 125 129 0.967 47.1 371 364 0.962 S 104.lbm 112 117 0.955 40.8 355 348 0.944 S 104.lbm 112 118 0.952 40.9 356 348 0.941 * 104.lbm 112 119 0.945 41.2 356 347 0.936 S 110.fft 111 112 0.995 41.5 381 372 0.988 * 110.fft 111 111 0.998 41.4 382 372 0.991 S 110.fft 111 112 0.992 41.7 382 373 0.982 S 112.spmv 147 148 0.996 52.8 377 358 0.996 S 112.spmv 147 148 0.992 52.9 377 357 0.995 * 112.spmv 147 150 0.980 53.6 377 357 0.982 S 114.mriq 109 111 0.985 36.5 337 330 0.988 * 114.mriq 109 109 0.996 36.3 343 332 0.993 S 114.mriq 109 111 0.984 36.8 337 332 0.981 S 116.histo 114 112 1.01 31.0 286 276 1.00 S 116.histo 114 117 0.970 32.4 284 275 0.962 * 116.histo 114 118 0.963 32.6 284 275 0.956 S 117.bfs 117 118 0.988 38.6 344 326 0.986 S 117.bfs 117 118 0.990 38.6 345 327 0.986 * 117.bfs 117 118 0.991 38.5 344 326 0.987 S 118.cutcp 99 99.5 0.995 33.4 343 335 0.993 S 118.cutcp 99 101 0.983 33.7 343 334 0.984 S 118.cutcp 99 100 0.986 33.7 343 335 0.985 * 120.kmeans 100 106 0.945 28.8 280 272 0.938 * 120.kmeans 100 104 0.965 28.2 279 272 0.958 S 120.kmeans 100 107 0.938 29.0 279 272 0.931 S 121.lavamd 109 111 0.983 40.2 381 362 0.982 S 121.lavamd 109 109 0.996 39.7 381 363 0.993 S 121.lavamd 109 111 0.983 40.3 381 363 0.979 * 122.cfd 126 129 0.978 44.0 351 341 0.980 S 122.cfd 126 127 0.994 43.3 351 342 0.995 * 122.cfd 126 127 0.996 43.3 351 342 0.995 S 123.nw 115 114 1.01 36.0 321 315 1.00 S 123.nw 115 115 1.00 36.2 321 315 0.998 * 123.nw 115 116 0.995 36.3 322 314 0.996 S 124.hotspot 114 115 0.993 38.1 344 332 0.990 * 124.hotspot 114 114 0.997 38.2 344 334 0.989 S 124.hotspot 114 115 0.988 38.5 344 334 0.979 S 125.lud 119 121 0.987 44.6 382 370 0.983 S 125.lud 119 121 0.984 44.7 382 370 0.983 * 125.lud 119 122 0.974 45.0 381 368 0.976 S 126.ge 155 155 0.997 53.5 353 344 1.00 * 126.ge 155 155 0.997 53.7 353 345 0.999 S 126.ge 155 153 1.01 53.1 353 347 1.01 S 127.srad 114 114 0.997 40.2 362 352 0.991 S 127.srad 114 114 0.996 40.3 362 352 0.991 * 127.srad 114 116 0.980 40.7 361 350 0.980 S 128.heartwall 106 105 1.01 35.9 348 341 1.00 * 128.heartwall 106 105 1.01 35.9 348 341 1.00 S 128.heartwall 106 108 0.983 36.5 349 338 0.988 S 140.bplustree 108 110 0.985 36.0 337 328 0.985 * 140.bplustree 108 110 0.978 36.2 337 328 0.977 S 140.bplustree 108 109 0.991 35.8 337 328 0.989 S ======================================================================================================================== 101.tpacf 107 107 0.997 33.0 319 308 0.988 * 103.stencil 125 130 0.963 47.1 371 363 0.961 * 104.lbm 112 118 0.952 40.9 356 348 0.941 * 110.fft 111 112 0.995 41.5 381 372 0.988 * 112.spmv 147 148 0.992 52.9 377 357 0.995 * 114.mriq 109 111 0.985 36.5 337 330 0.988 * 116.histo 114 117 0.970 32.4 284 275 0.962 * 117.bfs 117 118 0.990 38.6 345 327 0.986 * 118.cutcp 99 100 0.986 33.7 343 335 0.985 * 120.kmeans 100 106 0.945 28.8 280 272 0.938 * 121.lavamd 109 111 0.983 40.3 381 363 0.979 * 122.cfd 126 127 0.994 43.3 351 342 0.995 * 123.nw 115 115 1.00 36.2 321 315 0.998 * 124.hotspot 114 115 0.993 38.1 344 332 0.990 * 125.lud 119 121 0.984 44.7 382 370 0.983 * 126.ge 155 155 0.997 53.5 353 344 1.00 * 127.srad 114 114 0.996 40.3 362 352 0.991 * 128.heartwall 106 105 1.01 35.9 348 341 1.00 * 140.bplustree 108 110 0.985 36.0 337 328 0.985 * SPECaccel_ocl_energy_base 0.982 SPECaccel_ocl_base 0.985 SPECaccel_ocl_energy_peak -- SPECaccel_ocl_peak Not Run HARDWARE -------- CPU Name: Intel Xeon E5620 CPU Characteristics: CPU MHz: 2400 CPU MHz Maximum: 2400 FPU: Integrated CPU(s) enabled: 8 cores, 2 chips, 4 cores/chip, 2 threads/core CPU(s) orderable: 1,2 chips Primary Cache: 32 KB I + 32 KB D on chip per core Secondary Cache: 1 MB I+D on chip per core L3 Cache: 12 MB I+D on chip per chip Other Cache: None Memory: 24 GB (6 x 4 GB 2Rx4 PC3-10600R-9, ECC, running at 1066 MHz) Disk Subsystem: 500GB Seagate ST3500514NS SATA disk Other Hardware: None ACCELERATOR ----------- Accel Model Name: Tesla C2070 Accel Vendor: NVIDIA Accel Name: NVIDIA Tesla C2070 Type of Accel: GPU Accel Connection: PCIe 2.0 16x Does Accel Use ECC: yes Accel Description: NVIDIA Tesla C2070, 448 CUDA cores, 1150 MHz, 6 GB GDDR5 RAM (Fermi Generation) Accel Driver: NVIDIA UNIX x86_64 Kernel Module 304.54 SOFTWARE -------- Operating System: SUSE Linux Enterprise Server 11 (x86_64) SLES11 SP2 3.0.101-0.7.15-default Kernel 3.0.101-0.5-default Compiler: Gnu Compiler Collection C/C++ Version 4.3.4 File System: ext4 System State: Run level 5 (OpenCL drivers loaded, no X11-Desktop running) Other Software: NVIDIA CUDA 5.0.35, driver 304.54 POWER ----- Power Supply: 1200 W Power Supply Details: SuperMicro PWS-1K21P-1R Max. Power (W): 382.13 Idle Power (W): 185.87 Min. Temperature (C): 20.19 POWER ANALYZER -------------- Power Analyzer: Power Analyzer Hardware Vendor: ZES Zimmer Model: LMG450 Serial Number: 07510905 Input Connection: RS232 via USB-adapter Metrology Institute: PTB (Physikalisch Technische Bundesanstalt) Calibration By: ZES Zimmer Calibration Label: 04300751001 Calibration Date: 11.05.2013 PTDaemon Version: 1.6.2 (372e138a; 2013-12-04) Setup Description: connected to the single power supply that powers the system Current Ranges Used: 1.2A, 2.5A Voltage Range Used: 250V TEMPERATURE METER ----------------- Temperature Meter: Temperature Meter Hardware Vendor: Digi Model: Watchport/H Serial Number: W00513713 Input Connection: USB PTDaemon Version: 1.6.2 (372e138a; 2013-12-04) Setup Description: positioned in front of intake fan Operating System Notes ---------------------- continuously upgraded via SLES update manager Platform Notes -------------- Sysinfo program /home/juckel/kit39/Docs/sysinfo $Rev: 6874 $ $Date:: 2013-11-20 #$ 0953404ef7e75a5f9bbb534c6de3f831 running on spec-accel Wed Feb 26 20:47:28 2014 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(R) CPU E5620 @ 2.40GHz 2 "physical id"s (chips) 16 "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 : 4 siblings : 8 physical 0: cores 0 1 9 10 physical 1: cores 0 1 9 10 cache size : 12288 KB From /proc/meminfo MemTotal: 24599776 kB HugePages_Total: 0 Hugepagesize: 2048 kB /usr/bin/lsb_release -d SUSE Linux Enterprise Server 11 (x86_64) From /etc/*release* /etc/*version* SuSE-release: SUSE Linux Enterprise Server 11 (x86_64) VERSION = 11 PATCHLEVEL = 2 uname -a: Linux spec-accel 3.0.101-0.7.15-default #1 SMP Wed Jan 15 22:13:35 UTC 2014 (0e621ee) x86_64 x86_64 x86_64 GNU/Linux run-level 5 Feb 10 10:30 last=S SPEC is set to: /home/juckel/kit39 Filesystem Type Size Used Avail Use% Mounted on /dev/sda2 ext3 457G 199G 258G 44% / Cannot run dmidecode; consider saying 'chmod +s /usr/sbin/dmidecode' (End of data from sysinfo program) General Notes ------------- Pinned to socket 1 using numactl to avoid QPI transfers (numactl --physcpubind=1 runspec ...) Base Runtime Environment ------------------------ C benchmarks: OpenCL Platform: NVIDIA CUDA, OpenCL 1.1 CUDA 4.2.1 OpenCL Device #0: Tesla C2070, v 304.54 C++ benchmarks: OpenCL Platform: NVIDIA CUDA, OpenCL 1.1 CUDA 4.2.1 OpenCL Device #0: Tesla C2070, v 304.54 Base Compiler Invocation ------------------------ C benchmarks: gcc C++ benchmarks: g++ Base Optimization Flags ----------------------- C benchmarks: -O2 -I/opt/cuda/5.0.35/include -lOpenCL C++ benchmarks: -O2 -I/opt/cuda/5.0.35/include -lOpenCL The flags file that was used to format this result can be browsed at http://www.spec.org/accel/flags/gcc_flags.html You can also download the XML flags source by saving the following link: http://www.spec.org/accel/flags/gcc_flags.xml SPEC is a registered trademark 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 2014-2015 Standard Performance Evaluation Corporation Tested with SPEC ACCEL v39. Report generated on Tue Mar 3 14:20:46 2015 by ACCEL ASCII formatter v1212. Originally published on 17 March 2014.