第五章并行程序设计的一般过程.pptVIP

第五章 并行程序设计的 一般方法 ;Creating a Parallel Program;几个术语（ Terminology）;Identifying portions of the work that can be Identifying portions of the work that can be performed concurrently performed concurrently Mapping the concurrent pieces of work onto multiple processes running in parallel Distributing the input, output, and intermediate associated with the problem data Managing accesses to data shared by multiple processors. Synchronizing the processors at various stages of the parallel program execution.;Steps for Creating a Parallel Program;并行算法设计一般过程 ;划分;Decomposition基本概念;Exploit potential concurrency Minimize inter-process communication Exploit potential concurrency, limit sizes of processes;进程间通信开销;Assignment;划分技术 ;Recursive Decomposition ; Example 1 ;Example 2; 1. procedure SERIAL_MIN (A, n) 2. begin 3. min = A[0]; 4. for i := 1 to n ? 1 do 5. if (A[i] min) min := A[i]; 6. endfor; 7. return min; 8. end SERIAL_MIN; 1. procedure RECURSIVE_MIN (A, n) 2. begin 3. if ( n = 1 ) then 4. min := A [0] ; 5. else 6. lmin := RECURSIVE_MIN ( A, n/2 ); 7. rmin := RECURSIVE_MIN ( (A[n/2]), n - n/2 ); 8. if (lmin rmin) then 9. min := lmin; 10. else 11. min := rmin; 12. endelse; 13. endelse; 14. return min; 15. end RECURSIVE_MIN;例1：;树的仿真模型如下图所示，设Tree0为一条长为L0的直线段，在Tree0的中点和三等分点以一定角度延伸出L0/2或L0/3长的线段，从而得到Tree1。将Tree1的各分支做同样的变换得到Tree2，以此类推，直到得到一棵完整的树Tn，其中n为树的仿真精度，即迭代次数。;树的仿真任务图;Data Decomposition ;Output Data Decomposition ;Example ;Input Data Partitioning ;Example ;Intermediate Data Partitioning ;例: 数值计算中,矩阵运算经常会用到划分 按行划分 主要划分方法 按列划分 划分为子矩阵;Perimeter to Area comm-to-comp ratio (area to volume in 3-d) Depends on n,p: decreases with n, increases with p;在地图上求最短路径;例：用牛顿法求f(x)=0的根。;Master; 例：交通信息服务网格 地图查询服务 按功能划分 路况展示服务 出行方案服务; commonly used in search space exploration unlike the data decomposition, the search space is not known beforehand computation can terminate