allenpope
commented
4 years ago Added the following paragraph in my suggestion: NSF OPP has also supported the organization of two, week-long Community Early System Model Polar Modeling Workshops at the National Center for Atmospheric Research Mesa Laboratory in Boulder, CO. The workshops focused on advancing polar science using HPC and earth system modeling (http://www.cesm.ucar.edu/events/2019/pmw/ \& http://www.cesm.ucar.edu/events/2019/pmw/); they involved both lectures and practical sessions and were considered highly successful by the participants. These workshops addressed many of the above concerns, but there is no current plan to continue them into the future. Similar workshops with similar structure and different foci may be successful ways to break down barriers to using HPC in polar research.
With funding from National Science Foundation’s Office of Polar Programs, we (Marika, Dave, Alice, Myself) have organized two 1-week long CESM Polar Modeling Workshops in Boulder, Colorado at the National Center for Atmospheric Research Mesa Laboratory. The workshops have focused on advancing polar science using high performance computing and earth system modeling. They involved both lectures and practical sessions. Both were highly successful. I think these workshops directly address some of the concerns that you have raised as barriers. But at present there is no plan to continue them into the future. As such, I think they might be worth mentioning in your report, and more generally to the community as an example of a successful way to break down barriers to using high performance computing in polar research.
http://www.cesm.ucar.edu/events/2019/pmw/ http://www.cesm.ucar.edu/events/workshops/2018PMWS/
The 2018 workshop was funded by an NSF Grant for which I was the PI https://www.nsf.gov/awardsearch/showAward?AWD_ID=1745839
The 2019 workshop was funded by NCAR/NSF OPP through a direct supplement to Marika Holland (NCAR Senior Scientist).