SPEC® ACCEL™ ACC Result

IBM Corporation (Test Sponsor: NVIDIA Corporation)

Tesla P100

IBM Power Systems S822LC for High Performance
Computing (8335-GTB)

ACCEL license:	019	Test date:	Sep-2016
Test sponsor:	NVIDIA Corporation	Hardware Availability:	Sep-2016
Tested by:	IBM Corporation	Software Availability:	Sep-2016

Hardware
CPU Name:	POWER8 with NVLink
CPU Characteristics:
CPU MHz:	3259
CPU MHz Maximum:	3857
FPU:	Integrated
CPU(s) enabled:	16 cores, 2 chips, 8 cores/chip, 8 threads/core
CPU(s) orderable:	2 chips
Primary Cache:	32 KB I + 64 KB D on chip per core
Secondary Cache:	512 KB I+D on chip per core
L3 Cache:	8 MB I+D on chip per chip
Other Cache:	16 MB I+D off chip per 4 DIMMs
Memory:	512 GB (16 x 32 GB RDIMMs) DDR4 1600 MHz
Disk Subsystem:	2x 1TB SATA 6.0Gb/s 7200 RPM
Other Hardware:	No

Accelerator
Accel Model Name:	Tesla P100
Accel Vendor:	NVIDIA
Accel Name:	Tesla P100
Type of Accel:	GPU
Accel Connection:	NVLink
Does Accel Use ECC:	Yes
Accel Description:	See Notes
Accel Driver:	NVIDIA UNIX ppc64le Kernel Module 361.85

Software
Operating System:	Ubuntu 16.04.1 LTS 4.4.0-34-generic
Compiler:	PGI Accelerator Fortran/C/C++ Server, Release 16.9
File System:	ext4
System State:	Run level 5 (multi-user)
Other Software:	None

Results Table

Benchmark	Base						Peak
Benchmark	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	19.5	7.44	19.5	7.42	19.5	7.43
304.olbm	42.1	10.8	42.1	10.8	42.0	10.8
314.omriq	108	8.87	108	8.88	108	8.87
350.md	19.7	12.8	19.7	12.8	19.7	12.8
351.palm	177	2.09	172	2.15	176	2.10
352.ep	64.7	8.20	64.6	8.20	64.7	8.19
353.clvrleaf	65.2	6.83	64.9	6.85	64.9	6.85
354.cg	56.4	7.23	56.4	7.23	56.4	7.24
355.seismic	46.7	7.92	46.7	7.92	46.7	7.93
356.sp	80.0	3.45	80.3	3.44	80.5	3.43
357.csp	30.7	8.79	30.8	8.77	30.7	8.81
359.miniGhost	67.3	5.48	67.4	5.48	67.8	5.44
360.ilbdc	40.9	8.97	40.9	8.97	41.0	8.95
363.swim	56.4	4.08	55.7	4.13	56.0	4.10
370.bt	12.7	17.6	12.5	17.8	12.7	17.6

Submit Notes

The config file option 'submit' was used.

Platform Notes

 Sysinfo program /home/user/SPECACCEL/Docs/sysinfo
 $Rev: 6874 $ $Date:: 2013-11-20 #$ 0953404ef7e75a5f9bbb534c6de3f831
 running on gar1 Fri Sep  2 20:50:00 2016

 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
    clock : 2061.000000MHz
    clock : 2094.000000MHz
    clock : 2128.000000MHz
    clock : 2194.000000MHz
    clock : 2360.000000MHz
    clock : 2527.000000MHz
    clock : 4023.000000MHz
    machine : PowerNV 8335-GTB
    model : 8335-GTB
    platform : PowerNV
    revision : 1.0 (pvr 004c 0100)
    cpu : POWER8NVL (raw), altivec supported
 *
 * 0 "physical id" tags found.  Perhaps this is an older system,
 * or a virtualized system.  Not attempting to guess how to
 * count chips/cores for this system.
 *
       128 "processors"
    cores, siblings (Caution: counting these is hw and system dependent.  The
    following excerpts from /proc/cpuinfo might not be reliable.  Use with
    caution.)

 From /proc/meminfo
    MemTotal:       535690880 kB
    HugePages_Total:       0
    Hugepagesize:      16384 kB

 /usr/bin/lsb_release -d
    Ubuntu 16.04.1 LTS

 From /etc/*release* /etc/*version*
    debian_version: stretch/sid
    os-release:
       NAME="Ubuntu"
       VERSION="16.04.1 LTS (Xenial Xerus)"
       ID=ubuntu
       ID_LIKE=debian
       PRETTY_NAME="Ubuntu 16.04.1 LTS"
       VERSION_ID="16.04"
       HOME_URL="http://www.ubuntu.com/"
       SUPPORT_URL="http://help.ubuntu.com/"

 uname -a:
    Linux gar1 4.4.0-34-generic #53-Ubuntu SMP Wed Jul 27 16:04:07 UTC 2016
    ppc64le ppc64le ppc64le GNU/Linux

 run-level 5 Sep 2 16:36

 SPEC is set to: /home/user/SPECACCEL
    Filesystem                Type  Size  Used Avail Use% Mounted on
    /dev/mapper/g82L--vg-root ext4  788G  231G  517G  31% /

 (End of data from sysinfo program)
 Information from pgaccelinfo
 CUDA Driver Version:           8000
 NVRM version:                  NVIDIA UNIX ppc64le Kernel Module  361.85
 Device Number:                 0
 Device Name:                   Tesla P100-SXM2-16GB
 Device Revision Number:        6.0
 Global Memory Size:            17071669248
 Number of Multiprocessors:     56
 Concurrent Copy and Execution: Yes
 Total Constant Memory:         65536
 Total Shared Memory per Block: 49152
 Registers per Block:           65536
 Warp Size:                     32
 Maximum Threads per Block:     1024
 Maximum Block Dimensions:      1024, 1024, 64
 Maximum Grid Dimensions:       2147483647 x 65535 x 65535
 Maximum Memory Pitch:          2147483647B
 Texture Alignment:             512B
 Clock Rate:                    1480 MHz
 Execution Timeout:             No
 Integrated Device:             No
 Can Map Host Memory:           Yes
 Compute Mode:                  default
 Concurrent Kernels:            Yes
 ECC Enabled:                   Yes
 Memory Clock Rate:             715 MHz
 Memory Bus Width:              4096 bits
 L2 Cache Size:                 4194304 bytes
 Max Threads Per SMP:           2048
 Async Engines:                 3
 Unified Addressing:            Yes
 Managed Memory:                Yes
 PGI Compiler Option:           -ta=tesla:cc60