際際滷

際際滷Share a Scribd company logo
2019 YEAR IN REVIEW: ITS A WRAP
OPENACC HIGHLIGHTS
2
WHAT IS OPENACC?
main()
{
<serial code>
#pragma acc kernels
{
<parallel code>
}
}
Add Simple Compiler Directive
POWERFUL & PORTABLE
Directives-based
programming model for
parallel
computing
Designed for
performance and
portability on
CPUs and GPUs
SIMPLE
Open Specification Developed by OpenACC.org Consortium
3
silica IFPEN, RMM-DIIS on P100
OPENACC GROWING MOMENTUM
Wide Adoption Across Key HPC Codes
ANSYS Fluent
Gaussian
VASP
LSDalton
MPAS
GAMERA
GTC
XGC
ACME
FLASH
COSMO
Numeca
200 APPS* USING OpenACC
Prof. Georg Kresse
Computational Materials Physics
University of Vienna
For VASP, OpenACC is the way forward for GPU
acceleration. Performance is similar to CUDA, and
OpenACC dramatically decreases GPU
development and maintenance efforts. Were
excited to collaborate with NVIDIA and PGI as an
early adopter of Unified Memory.
 
VASP
Top Quantum Chemistry and Material Science Code
* Applications in production and development
4
OPENACC: READYING HPC COMMUNITY
FOR EXASCALE
LEARN MORE
In 2019, OpenACC.org announced a series of efforts to increase
community engagement, specification development and user
adoption, including:
 Electing Jack Wells, Director of Science at Oak Ridge
Leadership Computing Facility (OLCF), as the new vice
president;
 Electing Thomas Schulthess, director of the Swiss National
Supercomputing Center (CSCS), to the OpenACC Board of
Directors; and
 Expanding the OpenACC organizations vision and mission.
With OpenACCs growing
adoption across key HPC
applications and by focusing on
user needs, I believe we can
make a positive difference in the
scientific community.
Jack Wells
OpenACC Vice President and
Deputy Director of Science,
OLCF
The expanded vision of the OpenACC
organization created an opportunity for
me to bring my experience to help
continue the focus on the needs of the
research community maintaining high
standards to push productive
accelerated computing and training at
all levels of the computational
scientific journey.
Thomas Schulthess
OpenACC Board Member and
Director, CSCS
5
READ BLOG
At SC19, we announced the latest update to the OpenACC
language specification, version 3.0. This includes a number of
minor updates in response to user requests, in addition to
some clarifications.
We decided several years ago to be calendar-driven, instead
of feature-driven. That means we try to have a new version of
the spec every year around the time of the SC
conference. Whatever features we finish working on by the
end of that year will be part of the next version. Other features
that are still under discussion, or that require more work, will be
deferred until the next version.
UPDATED SPECIFICATION:
OPENACC 3.0
6
READ BLOG
The GPU Hackathon program has grown exponentially, helping
to advance scientific discovery and engineering progress. 2019
saw 15 GPU Hackathon events across the United States,
Europe, Asia and India.
We launched www.gpuhackathons.org to support the growing
program and bring resources right to your fingertips.
2019 A RECORD-BREAKING YEAR: GLOBAL
HACKATHONS AND A NEW WEBSITE
The team made tremendous progress during the week, and that momentum has carried the
project forward ever since. In four days they managed to obtain a two-orders-of-magnitude
increase in performance, and since then they've made further optimizations that gained them
another order of magnitude. This speedup has opened up an entirely new class of problems
in quasistatic global geophysics.
Jeroen Tromp, Blair Professor of Geology and Professor of
Applied & Computational Mathematics at Princeton University
7
WATCH VIDEO
GPU Bootcamps are an exciting and unique way
for scientists and researchers to learn the skills
needed to start quickly accelerating scientific codes
on GPUs.
Recently launched, these 1- or 2-day events
introduce available GPU libraries, programming
models, and platforms where attendees learn the
basics of GPU programming through extensive
hands-on collaboration based on a real-life code
using the OpenACC programming model.
OpenACC held over 20 GPU Bootcamps worldwide
in 2019.
GPU BOOTCAMPS LAUNCHED IN 2019
DELIVERING HANDS-ON SKILLS BUILDING
8
OPENACC AT GTC 2019
MARCH 18-20 | SAN JOSE, CA USA
VIEW SESSIONS
From weather and climate to computational biology to
supercomputing and HPC, OpenACC sessions featured
at GTC19 helped to facilitate discussions, educate
attendees and encourage networking and collaboration.
Tutorials delivered hands-on skills building, a Birds of a
Feather panel provided an open forum to discuss the
specification, and the Meet the Experts sessions
enabled one-on-one deep dives into using OpenACC to
solve specific challenges.
Additionally, our user community enjoyed an evening of
comradery at the OpenACC User Group Reception.
9
A SNAPSHOT FROM PASC19
JUNE 12-14 | ZURICH, SWITZERLAND
SEE INTERVIEW
OpenACCs own Sunita Chandrasekaran, User
Adoption Chair, was Technical Papers co-chair at
PASC19, hosted by ETH Zurich.
An interdisciplinary conference, and this year saw 36
papers submitted spanning multiple scientific
domains.
Best paper was awarded to Sam Hatfield from the
University of Oxford for Accelerating High-Resolution
Weather Models with Deep-Learning Hardware.
10
VIEW TALKS
Hosted by RIKEN Center for Computational Science (RIKEN
R-CCS) in Kobe, Japan, the 2019 OpenACC Annual
Meeting brought together researchers and developers to
discuss how to improve the specification, help accelerate
scientific efforts using the OpenACC programming model,
and grow the OpenACC organization and community.
 100+ Attendees
 2 Keynote Presentations
 11 Curated Talks
 1 GPU Bootcamp at RIKEN R-CCS
OPENACC ANNUAL MEETING
SEPTEMBER 2-5 | KOBE, JAPAN
11
READ BLOG
Co-located with SC19, WACCPD has been one of the
major forums to bring together programming model
users, developers, and tools community to share
knowledge and experiences to tackle emerging
complex parallel computing systems.
The workshop highlights the state-of-art through
accepted papers, showcases all aspects of
heterogeneous systems, and discusses innovative
features, techniques and lessons learned.
SC19: SIXTH WORKSHOP ON ACCELERATOR
PROGRAMMING USING DIRECTIVES
12
TRAINING AND EDUCATION
WATCH RECORDING
ON-DEMAND WEBINAR:
OpenACC: Directives For The Next-generation
Workforce
Presented by Sunita Chandrasekaran, University of Delaware
Ever-evolving computer architectures are demanding that the
next generation workforce be knowledgeable and proficient at
employing the best ways to achieve parallelism and performance
portability.
Watch this webinar to learn more about OpenACC and discover
the latest tools and resources to help you get started
accelerating your code with OpenACC.
13
TRAINING AND EDUCATION
SEE TUTORIALS
VIDEO TUTORIAL:
OpenACC Selected Topics
Presented by Michael Wolfe, NVIDIA/PGI Compilers and Tools
These engaging video tutorials cover a wide range of topics
including:
 Pool allocator using OpenACC;
 Managed memory using OpenACC;
 Implicit deep copy;
 True deep copy;
 The future of data management;
 Multicore execution; and
 Pinned memory.
14
TRAINING AND EDUCATION
VIEW COURSE
ONLINE VIDEO COURSE:
OpenACC Programming
Presented by Appentra
Explore Appentras new online video course.
Presented as a series and broken into simple, easy-to-
follow steps, these videos succinctly explain how to
quickly get up and running on OpenACC code
parallelization.
Learn best practices for parallel programming using
OpenACC, how to decompose codes into parallel
patterns and a practical step-by-step process based on
patterns for parallelizing any code.
15
STAY IN THE KNOW:
JOIN THE OPENACC COMMUNITY
JOIN TODAY
The OpenACC specification is designed for, and
by, users meaning that the OpenACC organization
relies on our users active participation to shape
the specification and to educate the scientific
community on its use.
Take an active role in influencing the future of both
the OpenACC specification and the organization
itself by becoming a member of the community.
WWW.OPENACC.ORG
Learn more at

More Related Content

OpenACC Highlights: 2019 Year in Review

  • 1. 2019 YEAR IN REVIEW: ITS A WRAP OPENACC HIGHLIGHTS
  • 2. 2 WHAT IS OPENACC? main() { <serial code> #pragma acc kernels { <parallel code> } } Add Simple Compiler Directive POWERFUL & PORTABLE Directives-based programming model for parallel computing Designed for performance and portability on CPUs and GPUs SIMPLE Open Specification Developed by OpenACC.org Consortium
  • 3. 3 silica IFPEN, RMM-DIIS on P100 OPENACC GROWING MOMENTUM Wide Adoption Across Key HPC Codes ANSYS Fluent Gaussian VASP LSDalton MPAS GAMERA GTC XGC ACME FLASH COSMO Numeca 200 APPS* USING OpenACC Prof. Georg Kresse Computational Materials Physics University of Vienna For VASP, OpenACC is the way forward for GPU acceleration. Performance is similar to CUDA, and OpenACC dramatically decreases GPU development and maintenance efforts. Were excited to collaborate with NVIDIA and PGI as an early adopter of Unified Memory. VASP Top Quantum Chemistry and Material Science Code * Applications in production and development
  • 4. 4 OPENACC: READYING HPC COMMUNITY FOR EXASCALE LEARN MORE In 2019, OpenACC.org announced a series of efforts to increase community engagement, specification development and user adoption, including: Electing Jack Wells, Director of Science at Oak Ridge Leadership Computing Facility (OLCF), as the new vice president; Electing Thomas Schulthess, director of the Swiss National Supercomputing Center (CSCS), to the OpenACC Board of Directors; and Expanding the OpenACC organizations vision and mission. With OpenACCs growing adoption across key HPC applications and by focusing on user needs, I believe we can make a positive difference in the scientific community. Jack Wells OpenACC Vice President and Deputy Director of Science, OLCF The expanded vision of the OpenACC organization created an opportunity for me to bring my experience to help continue the focus on the needs of the research community maintaining high standards to push productive accelerated computing and training at all levels of the computational scientific journey. Thomas Schulthess OpenACC Board Member and Director, CSCS
  • 5. 5 READ BLOG At SC19, we announced the latest update to the OpenACC language specification, version 3.0. This includes a number of minor updates in response to user requests, in addition to some clarifications. We decided several years ago to be calendar-driven, instead of feature-driven. That means we try to have a new version of the spec every year around the time of the SC conference. Whatever features we finish working on by the end of that year will be part of the next version. Other features that are still under discussion, or that require more work, will be deferred until the next version. UPDATED SPECIFICATION: OPENACC 3.0
  • 6. 6 READ BLOG The GPU Hackathon program has grown exponentially, helping to advance scientific discovery and engineering progress. 2019 saw 15 GPU Hackathon events across the United States, Europe, Asia and India. We launched www.gpuhackathons.org to support the growing program and bring resources right to your fingertips. 2019 A RECORD-BREAKING YEAR: GLOBAL HACKATHONS AND A NEW WEBSITE The team made tremendous progress during the week, and that momentum has carried the project forward ever since. In four days they managed to obtain a two-orders-of-magnitude increase in performance, and since then they've made further optimizations that gained them another order of magnitude. This speedup has opened up an entirely new class of problems in quasistatic global geophysics. Jeroen Tromp, Blair Professor of Geology and Professor of Applied & Computational Mathematics at Princeton University
  • 7. 7 WATCH VIDEO GPU Bootcamps are an exciting and unique way for scientists and researchers to learn the skills needed to start quickly accelerating scientific codes on GPUs. Recently launched, these 1- or 2-day events introduce available GPU libraries, programming models, and platforms where attendees learn the basics of GPU programming through extensive hands-on collaboration based on a real-life code using the OpenACC programming model. OpenACC held over 20 GPU Bootcamps worldwide in 2019. GPU BOOTCAMPS LAUNCHED IN 2019 DELIVERING HANDS-ON SKILLS BUILDING
  • 8. 8 OPENACC AT GTC 2019 MARCH 18-20 | SAN JOSE, CA USA VIEW SESSIONS From weather and climate to computational biology to supercomputing and HPC, OpenACC sessions featured at GTC19 helped to facilitate discussions, educate attendees and encourage networking and collaboration. Tutorials delivered hands-on skills building, a Birds of a Feather panel provided an open forum to discuss the specification, and the Meet the Experts sessions enabled one-on-one deep dives into using OpenACC to solve specific challenges. Additionally, our user community enjoyed an evening of comradery at the OpenACC User Group Reception.
  • 9. 9 A SNAPSHOT FROM PASC19 JUNE 12-14 | ZURICH, SWITZERLAND SEE INTERVIEW OpenACCs own Sunita Chandrasekaran, User Adoption Chair, was Technical Papers co-chair at PASC19, hosted by ETH Zurich. An interdisciplinary conference, and this year saw 36 papers submitted spanning multiple scientific domains. Best paper was awarded to Sam Hatfield from the University of Oxford for Accelerating High-Resolution Weather Models with Deep-Learning Hardware.
  • 10. 10 VIEW TALKS Hosted by RIKEN Center for Computational Science (RIKEN R-CCS) in Kobe, Japan, the 2019 OpenACC Annual Meeting brought together researchers and developers to discuss how to improve the specification, help accelerate scientific efforts using the OpenACC programming model, and grow the OpenACC organization and community. 100+ Attendees 2 Keynote Presentations 11 Curated Talks 1 GPU Bootcamp at RIKEN R-CCS OPENACC ANNUAL MEETING SEPTEMBER 2-5 | KOBE, JAPAN
  • 11. 11 READ BLOG Co-located with SC19, WACCPD has been one of the major forums to bring together programming model users, developers, and tools community to share knowledge and experiences to tackle emerging complex parallel computing systems. The workshop highlights the state-of-art through accepted papers, showcases all aspects of heterogeneous systems, and discusses innovative features, techniques and lessons learned. SC19: SIXTH WORKSHOP ON ACCELERATOR PROGRAMMING USING DIRECTIVES
  • 12. 12 TRAINING AND EDUCATION WATCH RECORDING ON-DEMAND WEBINAR: OpenACC: Directives For The Next-generation Workforce Presented by Sunita Chandrasekaran, University of Delaware Ever-evolving computer architectures are demanding that the next generation workforce be knowledgeable and proficient at employing the best ways to achieve parallelism and performance portability. Watch this webinar to learn more about OpenACC and discover the latest tools and resources to help you get started accelerating your code with OpenACC.
  • 13. 13 TRAINING AND EDUCATION SEE TUTORIALS VIDEO TUTORIAL: OpenACC Selected Topics Presented by Michael Wolfe, NVIDIA/PGI Compilers and Tools These engaging video tutorials cover a wide range of topics including: Pool allocator using OpenACC; Managed memory using OpenACC; Implicit deep copy; True deep copy; The future of data management; Multicore execution; and Pinned memory.
  • 14. 14 TRAINING AND EDUCATION VIEW COURSE ONLINE VIDEO COURSE: OpenACC Programming Presented by Appentra Explore Appentras new online video course. Presented as a series and broken into simple, easy-to- follow steps, these videos succinctly explain how to quickly get up and running on OpenACC code parallelization. Learn best practices for parallel programming using OpenACC, how to decompose codes into parallel patterns and a practical step-by-step process based on patterns for parallelizing any code.
  • 15. 15 STAY IN THE KNOW: JOIN THE OPENACC COMMUNITY JOIN TODAY The OpenACC specification is designed for, and by, users meaning that the OpenACC organization relies on our users active participation to shape the specification and to educate the scientific community on its use. Take an active role in influencing the future of both the OpenACC specification and the organization itself by becoming a member of the community.