AMS Short Course: Integrating NWP System Components Using Container Technology and Cloud Services

- | AMS Annual Meeting, Boston

The AMS Short Course on Integrating NWP System Components Using Container Technology and Cloud Services was held on 12 January 2020 preceding the 100th AMS Annual Meeting in Boston, Massachusetts.  

Software systems require substantial set-up to get all the necessary code, including external libraries, compiled on a specific platform. Recently, the concept of software containers has been gaining popularity, because they allow for software systems to be bundled (including operating system, libraries, code, and executables) and provided directly to users, eliminating possible frustrations with up-front system setup. This course provided information on using software containers that have been established for community use, to quickly spin up an NWP forecast system [using the Weather Research and Forecasting (WRF) model initialized with WRF Pre-processing System (WPS) and the Gridpoint Statistical Interpolation (GSI) data assimilation system] that can then be post-processed [using the Unified Post Processor (UPP)], visualized [with NCAR Command Language (NCL)], and verified [using the Model Evaluation Tools (MET) and METviewer].  Ultimately, the established containers substantially reduce the spin-up time for setting up and compiling software systems, and promote greater efficiency in getting to the end goal of producing model output and statistical analyses.

The goal of this course was to raise awareness about tools and facilities available to the community for testing and evaluating Numerical Weather Prediction (NWP) innovations, including the emerging set of software tools in reusable containers and cloud computing resources.

While this course may have appealed to a wide-reaching audience, this information may be particularly useful to undergraduate and graduate students interested in learning more about NWP and to university faculty that may find software containers and cloud computing to be useful teaching tools to add to their course curriculum. In addition, researchers who wish to collaborate with others may find these tools useful for sharing a code base and replicating procedures, even if they are not working on the same platform.

The format of the course consisted of a general overview of the NWP system components available in Docker containers (e.g., WPS/WRF, GSI, UPP, NCL, MET, METviewer) followed by the opportunity for participants to conduct hands-on learning exercises by running an end-to-end NWP system through specific usage examples with Docker containers.