Better Interfaces and Better Tools for Hardware-Software Codesign

随着花在设计上的时间成为数字系统设计成本中越来越重要的一部分，通过协同设计进行并行硬件和软件开发变得越来越重要。此外，新兴的现场可编程逻辑技术允许数字系统协同设计人员在设计过程中轻松地在软件和硬件之间迁移系统功能。然而，目前存在的协同设计工具很少，在这些工具中，可用性和功能之间往往存在反比关系。它们缺乏适合设计人员工作方式的接口，或者它们不能与现有的芯片和程序开发框架集成。我们建议从斯佩尔曼学院、杰克逊州立大学和塔斯基吉大学招收五名学生，这些都是历史上的黑人学院和大学（HBCUs），并在他们攻读奥本大学计算机科学与工程博士学位的过程中为他们提供支持、指导和建议，研究硬件软件协同设计接口和工具开发方面的主题。通过引导学生调查软硬件协同设计工具和界面开发的主题，我们不仅可以显著增加协同设计专家的供应，还可以显著增加非裔美国计算机科学家的供应：语言开发、图形用户界面、功能自动迁移的算法、工具间框架和协同设计的正式方法。

As time spent on design becomes an increasingly significant fraction of the cost of digital system design, concurrent hardware and software development through codesign becomes increasingly important. Also, emerging field programmable logic technologies permit digital system codesigners to easily migrate system functionality between software and hardware during the design process. Yet few codesign tools exist currently, and among those there is often an inverse relation between usability and functionality. They lack interfaces that fit with designers' ways of working, or they do not integrate with existing frameworks for chip and program development. We propose to recruit five students from Spelman College, Jackson State University, and Tuskegee University, all Historically Black Colleges and Universities (HBCUs), and provide them with support, mentoring, and advising as they pursue courses of study leasing to the PhD degree in Computer Science and Engineering at Auburn University, investigating topics in hardware-software codesign interface and tool development. We can significantly increase not only the supply of codesign specialists but also the nation's supply of African American computer scientists by guiding students in investigations of topics in hardware-software codesign tool and interface development: language development, graphical user interfaces, algorithms for automatic migration of functionality, inter-tool frameworks, and formal methods for codesign.