SAM++: Porting the E3SM-MMF cloud resolving model using a C++ portability library. (March 2022)

• Record Type: Journal Article
• Authors: Lyngaas, Isaac; Norman, Matthew; Kim, Youngsung
• Is Part Of: International journal of high performance computing applications. Volume 36:Number 2(2022)
• Page Start: 214
• Page End: 230
• Publication Date: 2022-03
• Subjects: Accelerator Computing -- C++ -- C++/Fortran Interoperability -- Cloud Resolving Model -- Portability; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/10943420211044495
• Languages: English
• ISSNs: 1094-3420

Abstract

In this work, we demonstrate the process for porting the cloud resolving model (CRM) used in the Energy Exascale Earth System Model Multi-Scale Modeling Framework (E3SM-MMF) from its original Fortran code base to C++ code using a portability library. This porting process is performed using the Yet Another Kernel Library (YAKL), a simplified C++ portability library that specializes in Fortran porting. In particular, we detail our step-by-step approach for porting the System for Atmospheric Modeling (SAM), the CRM used in E3SM-MMF, using a hybrid Fortran/C++ framework that allows for systematic reproduction and correctness testing of gradually ported YAKL C++ code. Additionally, analysis is done on the performance of the ported code using OLCF's Summit supercomputer.