SPEC® ACCEL™ OCL Result

Copyright 2014-2015 Standard Performance Evaluation Corporation

Intel (Test Sponsor: Intel Corporation)

Intel Xeon E5-2697 v3

R2208WTTYC1

SPECaccel_ocl_base = 2.09

SPECaccel_ocl_peak = 2.26

ACCEL license: 13 Test date: Feb-2015
Test sponsor: Intel Corporation Hardware Availability: Sep-2014
Tested by: Pavel Shelepugin, Alexander Bobyr Software Availability: Feb-2015
Benchmark results graph
Hardware
CPU Name: Intel Xeon E5-2697 v3
CPU Characteristics: Intel Turbo Boost Technology up to 3.6 GHz,
9.6 GT/s QPI, Hyper-Threading enabled
CPU MHz: 2600
CPU MHz Maximum: 3600
FPU: Integrated
CPU(s) enabled: 28 cores, 2 chips, 14 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: 256 KB I+D on chip per core
L3 Cache: 35 MB I+D on chip per chip, 35 MB
shared / 14 cores
Other Cache: None
Memory: 64 GB (8 x 8 GB 2Rx4 PC4-17000R-15, ECC)
Disk Subsystem: Panasas ActiveStor 12, PanFS 6.0
Other Hardware: None
Accelerator
Accel Model Name: Intel Xeon E5-2697 v3
Accel Vendor: Intel
Accel Name: Intel Xeon E5-2697 v3
Type of Accel: CPU
Accel Connection: 9.6 GT/s QPI
Does Accel Use ECC: yes
Accel Description: 2x Intel Xeon E5-2697 v3 CPUs with Hyper-Threading
Accel Driver: None
Software
Operating System: Red Hat Enterprise Linux Server
release 6.5 (Santiago)
2.6.32-358.6.2.el6.x86_64.crt1
Compiler: Intel C++ Composer XE 2015 for Linux,
Version 15.0.1.133 Build 20141023
File System: panfs
System State: Run level 3 (multi-user)
Other Software: Intel OpenCL SDK 2015 Version 15.1 Build 5.0.0.57

Results Table

Benchmark Base Peak
Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
101.tpacf 105     1.02   111     0.961  95.2   1.12   105     1.02   111     0.961  95.2   1.12  
103.stencil 76.9   1.63   76.7   1.63   76.6   1.63   76.9   1.63   76.7   1.63   76.6   1.63  
104.lbm 25.0   4.48   25.0   4.48   24.9   4.49   25.0   4.48   25.0   4.48   24.9   4.49  
110.fft 89.8   1.24   90.0   1.23   90.1   1.23   89.8   1.24   90.0   1.23   90.1   1.23  
112.spmv 132     1.12   133     1.11   132     1.11   132     1.12   133     1.11   132     1.11  
114.mriq 28.5   3.82   28.4   3.83   28.4   3.83   28.5   3.82   28.4   3.83   28.4   3.83  
116.histo 59.5   1.92   57.6   1.98   59.6   1.91   54.0   2.11   54.0   2.11   53.6   2.13  
117.bfs 28.9   4.05   28.9   4.05   30.4   3.85   26.7   4.38   27.4   4.28   27.3   4.29  
118.cutcp 66.6   1.49   66.6   1.49   66.6   1.49   69.2   1.43   66.6   1.49   66.5   1.49  
120.kmeans 46.2   2.17   45.5   2.20   45.9   2.18   46.2   2.17   45.5   2.20   45.9   2.18  
121.lavamd 19.6   5.55   19.6   5.56   19.5   5.58   17.9   6.10   18.0   6.06   17.9   6.09  
122.cfd 90.1   1.40   91.3   1.38   89.9   1.40   90.1   1.40   91.3   1.38   89.9   1.40  
123.nw 83.5   1.38   83.3   1.38   83.3   1.38   58.3   1.97   58.6   1.96   58.4   1.97  
124.hotspot 95.7   1.19   96.0   1.19   95.6   1.19   72.0   1.58   71.8   1.59   71.8   1.59  
125.lud 102     1.17   102     1.17   101     1.17   55.4   2.15   55.6   2.14   55.7   2.13  
126.ge 4.79  32.3    4.80  32.3    4.77  32.5    4.79  32.3    4.80  32.3    4.77  32.5   
127.srad 119     0.956  119     0.957  119     0.955  119     0.956  119     0.957  119     0.955 
128.heartwall 148     0.716  147     0.721  148     0.718  148     0.716  147     0.721  148     0.718 
140.bplustree 29.3   3.69   29.4   3.67   29.8   3.62   29.3   3.69   29.4   3.67   29.8   3.62  

Peak Tuning Notes

 The following compiler environment variables were used for the Peak run:
 (the description can be found in the flags file)
  CL_CONFIG_CPU_VECTORIZER_MODE=8 for 116.histo
  CL_CONFIG_CPU_VECTORIZER_MODE=4 for 117.bfs
  CL_CONFIG_CPU_RT_LOOP_UNROLL=16 for 118.cutcp
  CL_CONFIG_USE_FAST_RELAXED_MATH=1 for 121.lavamd

Submit Notes

The config file option 'submit' was used.
'numactl' utility was used for 'submit' option.

Platform Notes

 Sysinfo program /panfs/panfs3/users2/pshelepu/kit47/Docs/sysinfo
 $Rev: 6874 $ $Date:: 2013-11-20 #$ 0953404ef7e75a5f9bbb534c6de3f831
 running on ehk288 Tue Feb 24 02:18:44 2015

 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 E5-2697 v3 @ 2.60GHz
       2 "physical id"s (chips)
       56 "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 : 14
       siblings  : 28
       physical 0: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14
       physical 1: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14
    cache size : 35840 KB

 From /proc/meminfo
    MemTotal:       65860360 kB
    HugePages_Total:       0
    Hugepagesize:       2048 kB

 /usr/bin/lsb_release -d
    Red Hat Enterprise Linux Server release 6.5 (Santiago)

 From /etc/*release* /etc/*version*
    redhat-release: Red Hat Enterprise Linux Server release 6.5 (Santiago)
    system-release: Red Hat Enterprise Linux Server release 6.5 (Santiago)
    system-release-cpe: cpe:/o:redhat:enterprise_linux:6server:ga:server

 uname -a:
    Linux ehk288 2.6.32-358.6.2.el6.x86_64.crt1 #4 SMP Fri May 17 15:33:33 MDT
    2013 x86_64 x86_64 x86_64 GNU/Linux

 run-level 3 Feb 19 10:17

 SPEC is set to: /panfs/panfs3/users2/pshelepu/kit47
    Filesystem           Type   Size  Used Avail Use% Mounted on
    panfs://36.101.211.31/users2
                         panfs   76T   49T   28T  65% /panfs/panfs3/users2

 Cannot run dmidecode; consider saying 'chmod +s /usr/sbin/dmidecode'

 (End of data from sysinfo program)

Base Runtime Environment

C benchmarks:

 OpenCL Platform: Intel(R) OpenCL, OpenCL 1.2 LINUX   OpenCL Device #0: Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz, v 1.2.0.57 

C++ benchmarks:

 OpenCL Platform: Intel(R) OpenCL, OpenCL 1.2 LINUX   OpenCL Device #0: Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz, v 1.2.0.57 

Base Compiler Invocation

C benchmarks:

 icc 

C++ benchmarks:

 icpc 

Base Optimization Flags

C benchmarks:

 -O3   -xCORE-AVX2   -lintelocl 

C++ benchmarks:

 -O3   -xCORE-AVX2   -lintelocl 

Peak Runtime Environment

C benchmarks:

 OpenCL Platform: Intel(R) OpenCL, OpenCL 1.2 LINUX   OpenCL Device #0: Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz, v 1.2.0.57 

C++ benchmarks:

 OpenCL Platform: Intel(R) OpenCL, OpenCL 1.2 LINUX   OpenCL Device #0: Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz, v 1.2.0.57 

Peak Compiler Invocation

C benchmarks:

 icc 

C++ benchmarks:

 icpc 

Peak Optimization Flags

C benchmarks:

110.fft:  basepeak = yes 
114.mriq:  basepeak = yes 
116.histo:  -O3   -xCORE-AVX2   -lintelocl 
117.bfs:  Same as 116.histo 
118.cutcp:  Same as 116.histo 
121.lavamd:  Same as 116.histo 
124.hotspot:  -O3   -xCORE-AVX2   -DSPEC_ACCEL_WG_SIZE_0_0=40   -lintelocl 
127.srad:  basepeak = yes 
128.heartwall:  basepeak = yes 
140.bplustree:  basepeak = yes 

C++ benchmarks:

101.tpacf:  basepeak = yes 
103.stencil:  basepeak = yes 
104.lbm:  basepeak = yes 
112.spmv:  basepeak = yes 
120.kmeans:  basepeak = yes 
122.cfd:  basepeak = yes 
123.nw:  -O3   -xCORE-AVX2   -DSPEC_ACCEL_WG_SIZE_0_0=32   -lintelocl 
125.lud:  Same as 123.nw 
126.ge:  basepeak = yes 

The flags file that was used to format this result can be browsed at
http://www.spec.org/accel/flags/EM64T_Intel150_flags.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/accel/flags/EM64T_Intel150_flags.xml.