2. OpenACC is a directives-
based programming
approach to parallel
computing designed for
performance and portability
on CPUs and GPUs.
3. TOP HPC APPS ADOPTING OPENACC
Gaussian 16
2.25X speedup
Hardware: HPE server with dual Intel Xeon E5-2698 v3 CPUs (2.30GHz ; 16 cores/chip),
256GB memory and 4 Tesla K80 dual GPU boards (boost clocks: MEM 2505 and SM 875).
Gaussian source code compiled with PGI Accelerator Compilers (16.5) with OpenACC (2.5
standard).
ANSYS Fluent
30000
22500
15000
7500
0
T4 T8 T14 T28
Time
(S)
CPU
(cores)
CPU: (Haswell EP) Intel(R) Xeon(R) CPU E5-2695 v3 @2.30GHz, 2 sockets, 28 cores
GPU: Tesla K80 12+12 GB, Driver 346.46
Fluent Native Solver
Fluent HTC Solver K80 GPU
ANSYS Fluent ● Gaussian ● VASP ● GTC ● XGC ● ACME ● FLASH ● LSDalton ● COSMO ● ELEPHANT ● RAMSES ● ICON ● ORB5
5.15X speedup
ANSYS Fluent R18.0 Radiation Solver Valinomycin wB97xD/6-311+(2d,p) Freq
4. Join OpenACC at #SC17 for a user group
meeting, talks, workshops, BoF and labs
5. JOIN OPENACC USERS AT SC17 REGISTER NOW
Forth OpenACC User Group Meeting, Nov 14, Tag Restaurant
Have a Great Time!
Raghu Raj
Prasanna Kumar
National Center for
Atmospheric
Research (NCAR)
Lin Gan
Wuxi Supercomputing
Center
Jack Wells
Oak Ridge National
Laboratory
Invited SpeakersNetwork and Discuss
OpenACC-related research
Feedback on the Specification
OpenACC Trainings
Experiences and Best Practices
6. OPENACC TALKS AT SC 2017 COMPLETE SCHEDULE
Activity Date Location
BoF: OpenACC API User Experience, Vendor
Reaction, Relevance, and Roadmap
Nov 14, 2017
5:15 - 6:45PM
Room
210 - 212
Talk: Application Readiness Projects for the Summit
Architecture
Nov 14, 2017
2:00 - 2:20PM
NVIDIA Booth
Talk: Unstructured-Grid CFD Algorithms on the NVIDIA
Pascal and Volta Architectures
Nov 14, 2017
12:30 - 12:50PM
NVIDIA Booth
Talk: Accelerating HPC Programmer Productivity with
OpenACC and CUDA Unified Memory
Nov 15, 2017
10:30 - 10:50AM
NVIDIA Booth
Talk: An Approach to Developing MPAS on GPUs Nov 15, 2017
3:00 - 3:20PM
NVIDIA Booth
7. WORKSHOPS AND TUTORIALS AT SC 2017
COMPLETE SCHEDULE
Activity Date Location
Tutorial: Scalable Parallel Programming Using OpenACC for
Multicore, GPUs, and Manycore
Nov 13, 2017
8:30AM - 5:00PM
Room 302
Tutorial: Application Porting and Optimization on GPU-
Accelerated POWER Architectures
Nov 13, 2017
8:30AM - 5:00PM
Room 405
Workshop: Fourth Workshop on Accelerator Programming
Using Directives (WACCPD)
Nov 13, 2017
9:00AM - 5:30PM
Room
710 - 712
Workshop: EduHPC-17: Workshop on Education for High-
Performance Computing
Nov 13, 2017
9:00AM - 5:30PM
Room 505
Hands-on Labs: OpenACC labs at NVIDIA Booth at SC17 Nov 13 - 16, 2017
Show hours
NVIDIA Booth
8. RESOURCES
Paper: Accelerating lattice QCD simulations with 2 flavours of staggered
fermions on multiple GPUs using OpenACC - a first attempt
“We present the results of an effort to accelerate a Rational Hybrid Monte Carlo (RHMC) program
for lattice quantum chromodynamics (QCD) simulation for 2 flavours of staggered fermions on
multiple Kepler K20X GPUs distributed on different nodes of a Cray XC30. We do not use CUDA
but adopt a higher level directive based programming approach using the OpenACC platform. The
lattice QCD algorithm is known to be bandwidth bound; our timing results illustrate this clearly, and
we discuss how this limits the parallelization gains. We achieve more than a factor three speed-up
compared to the CPU only MPI program.”
READ NOW