CAREER: Assessment of Open-Source Software for High-Performance Computing

摘要：本项目将为软件工程专业的本科生和研究生实施一个经验-软件工程综合研究项目和一个经验-教育方法的范例。这将通过专注于评估用于高性能计算的开源软件质量的研究，以及通过开发相关课程和教学模块来实现。高性能计算是许多对现代社会至关重要的科学和工程领域的首选平台。这类现成的软件包包括实现数学模型的工具，以及管理并行分布式计算的中间件。软件开发的开源方法在这个领域正变得越来越有吸引力。该项目的研究活动将采用实证分析方法，找到评估用于高性能计算的开源软件质量的方法，使科学和工程用户能够做出更好的明智选择。选定的软件产品的案例研究将验证获得这种质量评估的实用性和有效性。美国的软件工程教育在很大程度上缺乏经验分析方法的指导。将针对这一需要编写一系列教材。研究结果和教学材料将通过出版物和nsf赞助的基于经验的软件工程中心（CeBASE）进行传播。

CCR-0132673CAREER: Assessment of Open-Source Software for High-Performance Computing Edward B. Allen, P.I.ABSTRACT:This project will implement an integrated program of empiricalsoftware-engineering research and an exemplar of education inempirical methods for undergraduate and graduate students insoftware engineering. This will be achieved through researchfocused on assessing the quality of open-source software forhigh-performance computing, and through development of relevantcourses and teaching modules.High-performance computing is the platform of choice in many fieldsof science and engineering that are critical to modern society.Off-the-shelf software packages in this category include tools toimplement mathematical models, and middleware to manage paralleland distributed computation. The open-source approach of softwaredevelopment is becoming increasingly attractive in this field.Research activities of this program will employ empirical analysismethods to find ways to assess the quality of open-source softwarefor high-performance computing, enabling scientific and engineeringusers to make better informed choices. Case studies of selectedsoftware products will validate the practicality and usefulness ofobtaining such quality assessments.Software-engineering education in America is largely lackinginstruction in empirical analysis methods. A spectrum of teachingmaterials targeted toward this need will be developed. Researchresults and instructional materials will be disseminated throughpublications and the NSF-sponsored Center for Empirically BasedSoftware Engineering (CeBASE).