A Scientific Workflow Management System for orchestration of parallel components in a cloud of large-scale parallel processing services

摘要

HPC Shelf is a proposal of a cloud computing platform for development, deployment and execution of component-based HPC applications with large-scale parallel processing requirements. Through components, it addresses the challenge of dealing with heterogeneous software and hardware resources in a large-scale HPC infrastructure comprising multiple parallel computing platforms. This paper introduces SAFe, a framework for deriving applications in HPC Shelf. It makes the role of a Scientific Workflow Management System (SWfMS) through which applications may deploy and monitor the execution of large-scale parallel computing systems built of components. (C) 2018 Elsevier B.V. All rights reserved.