Fortran Testing and Refactoring Infrastructure

Ultimately, The Fortran testing and refactoring infrastructure can be used with all software engineering projects written in FORTAN 77, Fortran90, Fortran95 and Fortran2003. Consequently, there are a large number of non-Nasa related applications written in Fortran that could benefit from our Fortran testing and refactoring framework. The Fortran developer community at large will benefit by being able to use improved mechanisms that have been available for some time to the C and C++ communities to test and refactor Fortran codes which will result in more robust, scalable and extensible codes. More specifically, the proposing firm is fortunate to be co-located with the Climate and Global Dynamics Division (CGD) at the National Center for Atmospheric Research who is administrating the Community Earth System Model (CESM). CESM is a fully-coupled, global climate model that provides state-of-the-art computer simulations of the Earth's past, present, and future climate states supported by NSF and DOE. The firm is also leading the DOE SciDAC project, FACETS, to integrate fusion codes some of which are Fortran codes which would benefit from the proposed tools. NASTRAN variants, Computational Fluid Dynamics (CFD), combustion and geophysical data processing codes written in Fortran may benefit from our work regardless of whether they are commercial or academic ventures. We plan to work closely with the team developing the Photran Eclipse Plugin to have the broadest impact.

While process study missions (e.g. CALIPSO/CloudSat) are critical to improve underlying climate model physics, future mission observations (e.g. CLARREO) are critical to determine the impact of those climate model improvements on the accuracy of predicting future climate change. Since the proposed tools will make it easier for NASA scientists to develop and optimize climate modeling codes, the proposed project in a sense supports all climate-related current and future missions to ensure that studies address serious accuracy issues in climate change observation and prediction. Beyond mission-related research, the proposed project has applications for all modeling efforts supported by NASA High-End Computing and the Modeling, Analysis and Prediction programs. One of our specific early adopter targets is a coupled atmosphere-ocean model called modelE developed at the NASA Goddard Institute for Space Studies (GISS). Other potential NASA applications from GISS are four mesoscale dynamics models including the Goddard Cumulus Ensemble. From Goddard Space Flight Center, the GEOS system of models could apply the proposed tools. Fortran codes in other NASA disciplines such as the CFD Codes for Turbomachinery (Glenn Research Center) are also potential applications. Related climate code targets with fringe connections to NASA include the Community Atmospheric Model and the Weather Research and Forecasting Model. More »