SPEC® ACCEL™ ACC Result

Copyright 2015-2018 Standard Performance Evaluation Corporation

Cray (Test Sponsor: Indiana University)

NVIDIA Tesla K20

Cray XK7

SPECaccel_acc_base = 1.71

SPECaccel_acc_peak = Not Run

ACCEL license: 3440A Test date: May-2018
Test sponsor: Indiana University Hardware Availability: Apr-2013
Tested by: Indiana University Software Availability: Feb-2018
Benchmark results graph
Hardware
CPU Name: AMD Opteron 6276
CPU Characteristics: AMD Turbo CORE Technology up to 3.2GHz, Turbo
CORE off
CPU MHz: 2300
CPU MHz Maximum: 3200
FPU: Integrated
CPU(s) enabled: 16 cores, 1 chip, 16 cores/chip
CPU(s) orderable: 1 chip
Primary Cache: 32 KB I + 16 KB D on chip per core
Secondary Cache: 16 MB I+D on chip per chip, 2 MB shared / 2 cores
L3 Cache: 16 MB I+D on chip per chip, 8 MB shared / 8 cores
Other Cache: None
Memory: 32 GB (4 x 8 GB 2Rx4 PC3L-12800R-11, ECC)
Disk Subsystem: None
Other Hardware: None
Accelerator
Accel Model Name: Tesla K20
Accel Vendor: NVIDIA
Accel Name: NVIDIA Tesla K20
Type of Accel: GPU
Accel Connection: PCIe 2.0 16x
Does Accel Use ECC: yes
Accel Description: NVIDIA Tesla K20m GPU, 2496 CUDA cores,
706 MHz, 5 GB GDDR5 RAM
Accel Driver: NVIDIA UNIX x86_64 Kernel Module 352.68
Software
Operating System: SUSE Linux Enterprise Server 11 (x86_64), Cray
3.0.101-0.46.1_1.0502.8871-cray_gem_c
Compiler: PGI Accelerator Fortran/C/C++ Server, Release 14.1
File System: lustre
NFSv3 (DDN SFA12KE) over 10GB Ethernet
System State: Run level 3 (multi-user)
Other Software: NVIDIA CUDA 5.5.51

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.
303.ostencil 143    1.01  143    1.01  143    1.01 
304.olbm 307    1.48  307    1.48  307    1.48 
314.omriq 493    1.94  493    1.94  493    1.94 
350.md 124    2.04  124    2.04  124    2.04 
351.palm 326    1.14  326    1.14  326    1.14 
352.ep 415    1.28  415    1.28  415    1.28 
353.clvrleaf 205    2.17  205    2.17  205    2.17 
354.cg 227    1.80  227    1.80  227    1.80 
355.seismic 187    1.98  187    1.98  187    1.98 
356.sp 167    1.66  167    1.66  167    1.66 
357.csp 206    1.31  206    1.31  206    1.31 
359.miniGhost 187    1.97  187    1.98  187    1.98 
360.ilbdc 145    2.52  145    2.53  145    2.53 
363.swim 122    1.89  123    1.87  120    1.91 
370.bt 98.7  2.26  98.6  2.26  98.7  2.26 

Platform Notes

 Sysinfo program
 /N/dc2/projects/hpc/lijunj/SPEC/accel-1.2-run/br2/cuda55/Docs/sysinfo
 $Rev: 6965 $ $Date:: 2015-04-21 #$ c05a7f14b1b1765e3fe1df68447e8a35
 running on nid00696 Thu May  3 08:43:33 2018

 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 : AMD Opteron(TM) Processor 6276
       1 "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 : 8
       siblings  : 16
       physical 0: cores 0 1 2 3 4 5 6 7
    cache size : 2048 KB

 From /proc/meminfo
    MemTotal:       33083668 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 = 3

 uname -a:
    Linux nid00696 3.0.101-0.46.1_1.0502.8871-cray_gem_c #1 SMP Mon Feb 12
    13:56:55 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux


 SPEC is set to: /N/dc2/projects/hpc/lijunj/SPEC/accel-1.2-run/br2/cuda55
    Filesystem                             Type    Size  Used Avail Use% Mounted
    on
    10.10.0.171@o2ib:10.10.0.172@o2ib:/dc2 lustre  5.3P  4.4P  798T  85% /N/dc2

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

 (End of data from sysinfo program)
 (End of data from sysinfo program)

General Notes



No: The test sponsor attests, as of date of publication, that CVE-2017-5754 (Meltdown)
is mitigated in the system as tested and documented.
No: The test sponsor attests, as of date of publication, that CVE-2017-5753 (Spectre variant 1)
is mitigated in the system as tested and documented.
No: The test sponsor attests, as of date of publication, that CVE-2017-5715 (Spectre variant 2)
is mitigated in the system as tested and documented.

This benchmark result is intended to provide perspective on
past performance using the historical hardware and/or
software described on this result page.

The system as described on this result page was formerly
generally available.  At the time of this publication, it may
not be shipping, and/or may not be supported, and/or may fail
to meet other tests of General Availability described in the
SPEC HPG Policy document, http://www.spec.org/hpg/policy.html

This measured result may not be representative of the result
that would be measured were this benchmark run with hardware
and software available as of the publication date.


Base Compiler Invocation

C benchmarks:

 pgcc 

Fortran benchmarks:

 pgfortran 

Benchmarks using both Fortran and C:

 pgcc   pgfortran 

Base Optimization Flags

C benchmarks:

 -fast   -Mfprelaxed   -acc   -ta=tesla:cc35   -ta=tesla:cuda5.5 

Fortran benchmarks:

 -fast   -Mfprelaxed   -acc   -ta=tesla:cc35   -ta=tesla:cuda5.5 

Benchmarks using both Fortran and C:

353.clvrleaf:  -fast   -Mfprelaxed   -acc   -ta=tesla:cc35   -ta=tesla:cuda5.5 
359.miniGhost:  -fast   -Mfprelaxed   -acc   -ta=tesla:cc35   -ta=tesla:cuda5.5   -Mnomain 

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

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