CAREER: Architectural Optimization of High Performance Microprocessor-Based Systems

职业：基于高性能微处理器的系统的架构优化

• 批准号: 9701915
• 负责人: David Albonesi
• 金额: $ 21万
• 依托单位: University of Rochester
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1997
• 资助国家: 美国
• 起止时间: 1997-07-15 至 2001-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9701915&HistoricalAwards=false
• 关键词: CAREER; Architectural; Optimization; Performance; Microprocessor

The primiary purpose of this project is to develop architectural optimization tools with dramatically improved efficiency and accuracy. Tools will be developed in three inter-related categories. The first is a comprehensive semi-automated methodology for simultaneously exploring the microprocessor and system level design spaces. The second is a hardware emulator coupled to software simulation. The third parallelizes the search for an optimum using a network of workstations. The end result, called ROADS, will have dramatically improved efficiency and accuracy over existing methods. ROADS will form the centerpiece of a new course sequence that will integrate the topics of computer architecture and implementation. Multilevel simulation will allow students to address a much wider scope of design problems than the simulators used in most current curricula. The result will be a new generation of engineers better prepared to both design and implement new computing architectures.

这个项目的主要目的是开发具有显著提高效率和准确性的架构优化工具。将在三个相互关联的类别中开发工具。第一个是同时探索微处理器和系统级设计空间的综合半自动化方法。第二种是硬件仿真与软件仿真相结合。第三种是使用工作站网络并行搜索最优。最终的结果被称为ROADS，将大大提高现有方法的效率和准确性。道路将形成一个新的课程序列的核心，将集成计算机体系结构和实现的主题。多层模拟将允许学生解决比目前大多数课程中使用的模拟器更广泛的设计问题。结果将是新一代工程师更好地准备设计和实现新的计算架构。