The Community for Technology Leaders
RSS Icon
Subscribe
Issue No.06 - June (1997 vol.8)
pp: 608-622
ABSTRACT
<p><b>Abstract</b>—Many partitioned scientific programs can be modeled as iterative executions of computational tasks and represented by iterative task graphs (ITGs). An ITG may or may not have dependence cycles. In this paper, we consider the symbolic scheduling of ITGs on distributed memory architectures with nonzero communication overhead and propose heuristic algorithms for scheduling both cyclic and acyclic ITGs without searching an entire iteration space. Our approach incorporates techniques of software pipelining, graph unfolding, directed acyclic graph (DAG) scheduling, and load balancing. We analyze the asymptotic optimality of the algorithms to show that the derived schedules are competitive to optimal solutions. We also study the sensitivity of scheduling performance on inaccurate weights. Finally, we present experimental results to demonstrate the effectiveness of the optimization techniques.</p>
INDEX TERMS
Scheduling, communication optimization, granularity, software pipelining, iterative task graphs, directed acyclic graphs.
CITATION
Tao Yang, Cong Fu, "Heuristic Algorithms for Scheduling Iterative Task Computations on Distributed Memory Machines", IEEE Transactions on Parallel & Distributed Systems, vol.8, no. 6, pp. 608-622, June 1997, doi:10.1109/71.595579
28 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool