基于因果关系复合的结构化并行程序设计

Causal Order Composition for Structured Parallel Programming

基于有关因果关系复合的ＣｏｍｍｕｎｉｃａｔｉｏｎＣｌｏｓｅｄＬａｙｅｒ理论，提出了一种结构化并行程序设计方法，以支持不同并行平台的“通用”并行程序设计方法．因果关系复合的范式定理的证明在理论上保证了采用“ＳＥＱＯＦＰＡＲ”形式进行结构化并行秩序设计的可行性；良好的代数性质也使得在这一模型中可采用代数方法的程序变换来进行并行程序设计、优化和验证．与其他并行程序设计模型和方法相比，提供了一种非常自然地综合描述数据并行和控制并行的途径；特别是基于代数变换的程序变换技术，提供了一种既独立于具体的体系结构又可高效实现的可能性．

Basing on the principles of communication closed layer, this paper proposes a structured parallel programming paradigm, to support general parallel programming across different platforms. The proof of Normal Formal of causal order provides us with the theoretical foundation for parallel programming in 'SEQ OF PAR' style. The well-founded algebra of CCL makes it possible to design, optimize and verify parallel programs through algebraic transformation approach. Thus, the frame work provides a natural integration of task and data parallelism and maintains portability and efficiency.