| 英文摘要: | This project is to plan, organize and host a community workshop to explore the potential for advancing "Modeling Research in the Cloud." The workshop, to be held in Spring 2017 at UCAR's facilities in Boulder, CO, will assemble researchers, educators, students, forecasters, and information technology (IT) professionals in academia, government and the private sector (including major cloud computing vendors). Specifically, the workshop participants will discuss the opportunities, challenges, and benefits of performing modeling research in a cloud computing environment and the means to and potential for enabling a paradigm shift in the conduct of weather and climate prediction and process studies. To achieve the workshop goals, about 30 participants from the aforementioned segments will be invited to the workshop.
Intellectual Merit:
Cloud computing represents a fundamental change in the way IT services are developed, deployed, operated, and paid for, placing science communities in general and weather and climate prediction community in particular in the middle of a major paradigm shift. The cloud appears to be a potential avenue for researchers to gain access to significant computing resources beyond the traditional supercomputing center for end-to-end modeling studies, democratizing access to high performance computing resources, vast amounts of storage, and unprecedented access to large volumes of data.
Historically, the modeling community has relied mostly on high performance computing facilities (e.g., NCAR-Wyoming Supercomputing facility and XSEDE) and local computing clusters to perform their predictions. With the maturity of and significant advances in cloud computing, it has recently emerged as an alternative new paradigm for hosting and delivering a broad array of services over the Internet. The purpose of the workshop is to facilitate an in-depth discussion of the myriad aspects and formulate approaches for integrating cloud computing capabilities into the weather and climate prediction landscape, to identify atmospheric sciences communities, based on their usage patterns, needs, and resources, who would best benefit and utilize it, and discuss the significance of such integration for advancing discoveries.
Broader Impacts:
The workshop will help the community to work toward a transformation in the conduct of atmospheric prediction studies, enabling researchers and educators to carry out their work in more innovative, efficient, and productive ways and push beyond the boundaries of their current knowledge and approaches. In addition to democratizing access to computing, the workshop has the potential for transforming the conduct of atmospheric prediction studies. Cloud computing affords additional opportunities to advance data, cyber and geospatial literacy of students, and in the process the development of the next generation workforce, by enabling authentic data- and cyber-enabled inquiries.
There is a compelling need for students to understand and take advantage of rapidly developing and evolving cyberinfrastructure, data-related technologies, and data-enabled modes of scientific inquiry. The workshop will engage graduate students, offering opportunities for enhancing their skills in data science, big data, cloud computing, and professional development. The results of the workshop will also be disseminated broadly through Unidata newsletters and presentations at scientific and computing conferences, including the AMS and AGU meetings. |