High-Performance and System Integration for Symbolic and Numeric Computation

符号和数值计算的高性能和系统集成

• 批准号: 9201800
• 负责人: Paul Wang
• 金额: $ 12.6万
• 依托单位: Kent State University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-01-01 至 1996-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9201800&HistoricalAwards=false
• 关键词: Performance; System; Integration; Symbolic; Numeric

The first goal of this project is to develop and parallel algorithms for symbolic polynomial factorization. The past work on ``pfactor'' is extended, and parallel linear and quadratic p-lifting techniques are investigated. The second goal is to investigate the integration of numeric, symbolic, and graphic techniques. A distributed paradigm is devised to interface loosely coupled scientific and mathematical computation components. The vision is to develop scientific computation systems consisting of multiple autonomous compute servers running on different hosts. Each compute server in such a system supplies a single or a set of related services such as polynomial factorization, solution of linear or nonlinear equations, automatic code generation from symbolic programs, or graphic display of mathematical functions, just to name a few services.

这个项目的第一个目标是开发符号多项式因式分解的并行算法.推广了过去关于“p因子”的工作，研究了并行线性和二次p-提升技术。第二个目标是研究数字、符号和图形技术的集成。设计了一个分布式范例来连接松散耦合的科学和数学计算组件。我们的愿景是开发由运行在不同主机上的多个自主计算服务器组成的科学计算系统。这样的系统中的每个计算服务器提供单个或一组相关服务，例如多项式因式分解、线性或非线性方程的求解、从符号程序自动生成代码、或数学函数的图形显示，仅举几个服务。