一类图像并行处理的理论最优化调度方法

在图像并行处理应用中，有很大一部分并行算法是属于迭代同步的数据并行算法。这类分布式应用的子任务间需要某种形式的同步，因而要有协作调度的算法保证子任务基本上同时开始，并以同样速度执行。该文提出了一种以并行程序最短执行时间为目标的数据划分和任务协作调度算法。与其它类似算法相比，这个算法的最大特点是考虑了通信开销，因而更加符合实际的应用，更加有效。

Theoretical Optimal Scheduling of a Class of Image Parallel Processing

A large and useful subset of image processing applications can be characterized using data parallel iterative synchronous algorithms. The classes of distributed applications require some form of synchronization among the subtasks, hence the need for coscheduling to guarantee that subtasks start at the same time and execute at the same pace on a group of workstations. A partitioning and coscheduling algorithm with minimum task's finish time is presented. One important rational characteristic of this algorithm is it has considered the communication overhead.