際際滷

際際滷Share a Scribd company logo

OpenACC Monthly Highlights March 2019

Download as PPTX, PDF
OpenACC
OpenACC

Stay up-to-date on the latest news, events and resources for the OpenACC community. This months highlights covers an upcoming OpenACC webinar, complete schedule of GPU hackathons, recent research papers and more!

1 of 11
Download to read offline
March 2019 OPENACC MONTHLY 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 OVER 100 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 DONT MISS THESE UPCOMING EVENTS COMPLETE LIST OF EVENTS Event Call Closes Event Date MIT GPU Hackathon April 5, 2019 June 3-7, 2019 Universidad de los Andes GPU Hackathon (Columbia) April 19, 2019 June 17-21, 2019 Princeton GPU Hackathon April 26, 2019 June 24-28, 2019 GPU Bootcamp in partnership with NCHC (Taiwan) April 26, 2019 June 26, 2019 NERSC GPU Hackathon May 15, 2019 July 15-19, 2019 University of Sheffield (UK) June 16, 2019 August 19-23, 2019 Brookhaven GPU Hackathon May 31, 2019 September 23-27, 2019
5 WEBINAROPENACC: DIRECTIVES FOR THE NEXT-GENERATION WORKFORCE Thursday, April 18, 2019 | 10 - 11 AM PST REGISTER TODAY With evolving computer architectures, the next generation workforce must be knowledgeable and proficient at employing the best ways to achieve parallelism and performance portability. Attend this webinar to learn more about OpenACC, understand how migrating legacy code can benefit interdisciplinary scientific research, and discover the latest tools and resources to help you get started accelerating your code with OpenACC. Sunita Chandrasekaran Assistant Professor University of Delaware
6 SPEEDING RESEARCH COMPUTING 2nd Annual GPU Hackathon at Pawsey Supercomputing Centre LEARN MORE Six teams across Australia attended the 2nd Annual GPU Hackathon at the Pawsey Supercomputing Centre to adapt scientific code for accelerated computing. Representing scientific domains such as atomic collision, computational fluid dynamics, and astrophysics, the teams achieved performance increases of 30-180x for their codes, both full applications and kernels, over baseline single- core/multicore system. A big THANK YOU to the organizers and mentors!
7 NERSC, NVIDIA TO PARTNER ON COMPILER DEVELOPMENT FOR PERLMUTTER SYSTEM LEARN MORE The National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory (Berkeley Lab) has signed a contract with NVIDIA to enhance GPU compiler capabilities for Berkeley Labs next-generation Perlmutter supercomputer. Together with OpenACC, this OpenMP collaboration gives HPC developers more options for directives-based programming from a single compiler on GPUs and CPUs. Our joint effort on programming tools for the Perlmutter supercomputer highlights how NERSC and NVIDIA are simplifying migration and development of science and engineering applications to pre-exascale systems and beyond. Doug Miles, Senior Director, NVIDIA
8 RESOURCES Paper: GPU Acceleration of an Established Solar MHD Code using OpenACC R. M. Caplan, J. A. Linker, Z. Miki, C. Downs, T. T旦r旦k, V. S. Titov GPU accelerators have had a notable impact on high-performance computing across many disciplines. They provide high performance with low cost/power, and therefore have become a primary compute resource on many of the largest supercomputers. Here, we implement multi-GPU acceleration into our Solar MHD code (MAS) using OpenACC in a fully portable, single-source manner. Our preliminary implementation is focused on MAS running in a reduced physics "zero-beta" mode. While valuable on its own, our main goal is to pave the way for a full physics, thermodynamic MHD implementation. We describe the OpenACC implementation methodology and challenges. "Time-to-solution" performance results of a production-level flux rope eruption simulation on multi-CPU and multi-GPU systems are shown. We find that the GPU-accelerated MAS code has the ability to run "zero-beta" simulations on a single multi-GPU server at speeds previously requiring multiple CPU server-nodes of a supercomputer. READ PAPER
9 RESOURCES Paper: Acceleration of Three-Dimensional Tokamak Magnetohydrodynamical Code with Graphics Processing Unit and OpenACC Heterogeneous Parallel Programming H.W. Zhang, J. Zhu, Z.W. Ma, G. Y. Kan, X. Wang, W. Zhang In this research, the OpenACC heterogeneous parallel programming model is successfully applied to modification and acceleration of the three-dimensional Tokamak magnetohydrodynamical code (CLT). The implementation of the OpenACC in CLT, performance test, and benchmarking are introduced in this paper. Combination of OpenACC and MPI technologies conveniently implements the multiple-GPUs parallel programming in CLT. Significant speedup ratios are achieved on NVIDIA TITAN Xp and TITAN V GPUs, respectively, with very few modifications to the source code. Furthermore, the validity of the double precision calculations on the above-mentioned two graphics cards has also been strictly verified with m/n=2/1 resistive tearing mode instability in Tokamak. READ PAPER
10 RESOURCES Video Tutorial: OpenACC Selected Topics Michael Wolfe, NVIDIA/PGI Compilers and Tools These engaging video tutorials cover a wide range of topics including: pool allocator using OpenACC, managed memory msing OpenACC, implicit deep copy, true deep copy, the future of data management, multicore execution, and pinned memory. VIEW NOW
WWW.OPENACC.ORG Learn more at

More Related Content

OpenACC Monthly Highlights March 2019

  • 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 OVER 100 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 DONT MISS THESE UPCOMING EVENTS COMPLETE LIST OF EVENTS Event Call Closes Event Date MIT GPU Hackathon April 5, 2019 June 3-7, 2019 Universidad de los Andes GPU Hackathon (Columbia) April 19, 2019 June 17-21, 2019 Princeton GPU Hackathon April 26, 2019 June 24-28, 2019 GPU Bootcamp in partnership with NCHC (Taiwan) April 26, 2019 June 26, 2019 NERSC GPU Hackathon May 15, 2019 July 15-19, 2019 University of Sheffield (UK) June 16, 2019 August 19-23, 2019 Brookhaven GPU Hackathon May 31, 2019 September 23-27, 2019
  • 5. 5 WEBINAROPENACC: DIRECTIVES FOR THE NEXT-GENERATION WORKFORCE Thursday, April 18, 2019 | 10 - 11 AM PST REGISTER TODAY With evolving computer architectures, the next generation workforce must be knowledgeable and proficient at employing the best ways to achieve parallelism and performance portability. Attend this webinar to learn more about OpenACC, understand how migrating legacy code can benefit interdisciplinary scientific research, and discover the latest tools and resources to help you get started accelerating your code with OpenACC. Sunita Chandrasekaran Assistant Professor University of Delaware
  • 6. 6 SPEEDING RESEARCH COMPUTING 2nd Annual GPU Hackathon at Pawsey Supercomputing Centre LEARN MORE Six teams across Australia attended the 2nd Annual GPU Hackathon at the Pawsey Supercomputing Centre to adapt scientific code for accelerated computing. Representing scientific domains such as atomic collision, computational fluid dynamics, and astrophysics, the teams achieved performance increases of 30-180x for their codes, both full applications and kernels, over baseline single- core/multicore system. A big THANK YOU to the organizers and mentors!
  • 7. 7 NERSC, NVIDIA TO PARTNER ON COMPILER DEVELOPMENT FOR PERLMUTTER SYSTEM LEARN MORE The National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory (Berkeley Lab) has signed a contract with NVIDIA to enhance GPU compiler capabilities for Berkeley Labs next-generation Perlmutter supercomputer. Together with OpenACC, this OpenMP collaboration gives HPC developers more options for directives-based programming from a single compiler on GPUs and CPUs. Our joint effort on programming tools for the Perlmutter supercomputer highlights how NERSC and NVIDIA are simplifying migration and development of science and engineering applications to pre-exascale systems and beyond. Doug Miles, Senior Director, NVIDIA
  • 8. 8 RESOURCES Paper: GPU Acceleration of an Established Solar MHD Code using OpenACC R. M. Caplan, J. A. Linker, Z. Miki, C. Downs, T. T旦r旦k, V. S. Titov GPU accelerators have had a notable impact on high-performance computing across many disciplines. They provide high performance with low cost/power, and therefore have become a primary compute resource on many of the largest supercomputers. Here, we implement multi-GPU acceleration into our Solar MHD code (MAS) using OpenACC in a fully portable, single-source manner. Our preliminary implementation is focused on MAS running in a reduced physics "zero-beta" mode. While valuable on its own, our main goal is to pave the way for a full physics, thermodynamic MHD implementation. We describe the OpenACC implementation methodology and challenges. "Time-to-solution" performance results of a production-level flux rope eruption simulation on multi-CPU and multi-GPU systems are shown. We find that the GPU-accelerated MAS code has the ability to run "zero-beta" simulations on a single multi-GPU server at speeds previously requiring multiple CPU server-nodes of a supercomputer. READ PAPER
  • 9. 9 RESOURCES Paper: Acceleration of Three-Dimensional Tokamak Magnetohydrodynamical Code with Graphics Processing Unit and OpenACC Heterogeneous Parallel Programming H.W. Zhang, J. Zhu, Z.W. Ma, G. Y. Kan, X. Wang, W. Zhang In this research, the OpenACC heterogeneous parallel programming model is successfully applied to modification and acceleration of the three-dimensional Tokamak magnetohydrodynamical code (CLT). The implementation of the OpenACC in CLT, performance test, and benchmarking are introduced in this paper. Combination of OpenACC and MPI technologies conveniently implements the multiple-GPUs parallel programming in CLT. Significant speedup ratios are achieved on NVIDIA TITAN Xp and TITAN V GPUs, respectively, with very few modifications to the source code. Furthermore, the validity of the double precision calculations on the above-mentioned two graphics cards has also been strictly verified with m/n=2/1 resistive tearing mode instability in Tokamak. READ PAPER
  • 10. 10 RESOURCES Video Tutorial: OpenACC Selected Topics Michael Wolfe, NVIDIA/PGI Compilers and Tools These engaging video tutorials cover a wide range of topics including: pool allocator using OpenACC, managed memory msing OpenACC, implicit deep copy, true deep copy, the future of data management, multicore execution, and pinned memory. VIEW NOW