The Performance, Portability, and Productivity in HPC (P3HPC) Forum is an opportunity to present on ideas and progress toward the goal of performance portability across current and future high performance computing platforms.
The two primary goals of this meeting are to:
- Share best practices and ideas between application developers, hardware architects, and software architects, and to focus on the issue of achieving high performance across platforms without greatly sacrificing portability, productivity, and maintainability
- Identify major challenges toward the goal of performance, portability, and productivity, and work with vendors and tool providers on determining implementations and solutions
We are particularly interested in research that addresses the complexities of real-life applications and/or realistic workloads, the composability challenges arising from the use of bespoke solutions, and the desire to “future-proof” applications in the long term.
This meeting is open to the following:
- Application developers preparing codes for current and future platforms
- Vendors who provide (or are potential providers of) HPC platforms and/or enabling technologies
- Solution-providers who are developing software tools aimed at helping application teams approach the challenges of performance portability and productivity
Since this meeting is an open forum, all accepted presentations must be free of NDA material with appropriate review and release from your organization for public posting. We plan to have a combination of technical talks, panel/breakout sessions, working lunches, and time for informal interactions with colleagues.
The deadline to register is August 24, 2020.
This meeting launched in 2016, recognizing the immense challenges of porting and optimizing large applications to the advanced architecture systems planned for deployment within the DOE’s National Nuclear Security Administration (NNSA) and Office of Science (SC) laboratories. The meeting has continued in the subsequent years with an ongoing and expanded focus on the problems of performance portability across the larger HPC community.