摘要： 对于高性能并行计算机而言，如何找到一种好的计算和数据划分，对数据和计算进行合理划分，增加数据本地化来减少处理器间的通信是提高其并行性能的关键。该文讨论了一种线性的自动进行无数据重组的计算和数据划分算法。
关键词:
并行编译,
数据划分,
计算划分,
循环级并行
Abstract: For high-performance parallel computer, finding a good decomposition of computation and data, minimizing communication by increasing the locality of data references is an important optimization for achieving high performance. This paper describes a linear decomposition algorithm that automatically finds computation and data decomposition and no data reorganization communication.
Key words:
Parallel compiler,
Data decomposition,
Computation decomposition,
Loop level parallel
董春丽;韩 林;赵荣彩. 并行编译中一种线性数据和计算划分算法[J]. 计算机工程, 2006, 32(24): 26-28.
DONG Chunli; HAN Lin; ZHAO Rongcai. Linear Computation and Data Decomposition Algorithm in Parallel Compilation[J]. Computer Engineering, 2006, 32(24): 26-28.