An Approximation Algorithm for Multiprocessor Scheduling of Trees with Communication Delays

In this paper, we investigate multiprocessor scheduling of trees considering the communication delay. The objective of our study is to minimize the makespan of the scheduling. We propose an approximation algorithm for computing a scheduling of trees when each task and communication have unit execution times. We show that the approximation difference of the total completion time of our proposal algorithmfrom the optimum is at most [m \over 2], where m is the number of processing elements. We also show that our algorithm is optimal for m = 2.