Scheduling precedence graphs in systems with interprocessor communication times
SIAM Journal on Computing
Scheduling parallel program tasks onto arbitrary target machines
Journal of Parallel and Distributed Computing - Special issue: software tools for parallel programming and visualization
IEEE Transactions on Parallel and Distributed Systems
On Exploiting Task Duplication in Parallel Program Scheduling
IEEE Transactions on Parallel and Distributed Systems
Static scheduling algorithms for allocating directed task graphs to multiprocessors
ACM Computing Surveys (CSUR)
Performance-Effective and Low-Complexity Task Scheduling for Heterogeneous Computing
IEEE Transactions on Parallel and Distributed Systems
IEEE Transactions on Parallel and Distributed Systems
Performance Modeling and Prediction of Nondedicated Network Computing
IEEE Transactions on Computers
Hypertool: A Programming Aid for Message-Passing Systems
IEEE Transactions on Parallel and Distributed Systems
IEEE Transactions on Parallel and Distributed Systems
DSC: Scheduling Parallel Tasks on an Unbounded Number of Processors
IEEE Transactions on Parallel and Distributed Systems
Hi-index | 0.00 |
For the different properties between dedicated grid and non-dedicated grid, two algorithms, named by DSADG (a Dynamic Scheduling Algorithm for Dedicated grid) and SARCNG (a Scheduling Algorithm with Reliable Communications for Non-dedicated grid), are presented. The former, based on the idea of dynamic critical path, proposes the method to compute the difference index of task executing on different grid nodes. Compared with other 9 algorithms, DSADG shows lower time complexity, smaller make-span and less grid nodes used in the scheduling process. Based on DSADG and the theories of queue and random process, the latter, focusing on the distinguishable features like heterogeneousness, reliability and availability, presents a function that scales the task scheduling reliability and grid node availability. Test results show SARCNG can not only ensure a smaller make-span, but also choose grid nodes with higher reliability.