RIA: New Techniques for High-Level Synthesis

RIA：高级合成新技术

• 批准号: 9010406
• 负责人: John Nestor
• 金额: $ 6.7万
• 依托单位: Illinois Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1990
• 资助国家: 美国
• 起止时间: 1990-06-01 至 1992-11-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9010406&HistoricalAwards=false
• 关键词: RIA; New; Techniques; Level; Synthesis

Research is on high level synthesis tools that translate a program-like behavioral description into a design suitable for implementation. The tools must meet designer constraints, especially timing constraints, while producing high quality designs. The research is being done in three phases. First, algorithms are being developed to produce an initial schedule that meets timing constraints, thus providing a starting point for the synthesis process. Second, a new framework is being developed for synthesis and design exploration using simulated annealing. The framework represents scheduling and allocation decisions and provides a set of moves that will modify designs while maintaining all timing constraints. Third, it is planned to expand the framework to incorporate physical design information and to provide a bridge to logic synthesis tools.

研究高级综合工具，将类似程序的行为描述转换为适合实现的设计。这些工具必须满足设计者的约束，特别是时间约束，同时产生高质量的设计。这项研究分三个阶段进行。首先，正在开发算法，以产生满足时间限制的初始时间表，从而为合成过程提供一个起点。第二，正在开发一种新的框架，用于使用模拟退火法进行综合和设计探索。该框架代表了调度和分配决策，并提供了一组移动，这些移动将修改设计，同时保持所有时间约束。第三，计划扩大该框架，以纳入物理设计信息，并为逻辑综合工具提供桥梁。