At-scale capability computing, including data-intensive and compute-intensive approaches, is enabling science, engineering, and a rapidly growing list of nontraditional disciplines to confront extraordinary new research challenges. U.S. computational and data scientists are relying on these resources to tackle problems including predicting the structure of complex biological systems, designing materials with specified properties atom by atom, predicting climate and ecosystem changes, and improving complex systems from chemical plants to airplanes. The transformational research enabled by current National Science Foundation-funded national resources only sets the stage for the next leap in scientific research and engineering practice. Meeting the future needs of the computational science community will require sustained investment in new national resources capable of sustained extreme scale performance.
The two Brainstorm HPCD Workshops planned for 2015 will develop community-based inputs for the future grand challenges and open problems that drive the needs of the computational science and engineering research community for high-end computing and data analytics resources, specifically the community needs and requirements for NSF provisioned resources. The goals of these workshops are:
- Provide an effective mechanism to facilitate discussions and identification of the needs of the NSF-sponsored science and engineering research community for advanced capacity through capability computing, including data-intensive and compute-intensive.
- Proactively encourage the inclusion of issues related to education and involvement of industry
- Disseminate the results (final report, position papers, presentations) to the large computational science and engineering community
In addition, the final report will be made available to the National Research Council (NRC). Last year, the National Science Foundation (NSF) requested the NRC to carry out a study examining anticipated priorities and associated trade-offs for advanced computing in support of NSF-sponsored science and engineering research. The study encompasses advanced computing activities and programs throughout NSF. The NRC established the Committee on Future Directions for NSF Advanced Computing Infrastructure to Support U.S. Science in 2017-2020, co-chaired by William Gropp and Robert Harrison. The committee published its interim report in fall 2014 and is expected to publish a final report in summer 2015. The NRC committee is seeking further comments and input on a set of topical areas as they formulate their final recommendations, and in particular for nine topic areas in the body of the report. The workshop report and position papers will provide such input to the NRC study, especially in the area of high-end, at-scale computation and big data.