An incentive-based distributed mechanism for scheduling divisible loads in tree networks

Journal Article · Thu Mar 01 04:00:00 EST 2012 · Journal of Parallel and Distributed Computing

DOI:https://doi.org/10.1016/j.jpdc.2011.11.008· OSTI ID:1439043

Carroll, T. E.; Grosu, D.

The underlying assumption of Divisible Load Scheduling (DLS) theory is that the pro-cessors composing the network are obedient, i.e., they do not “cheat” the scheduling algorithm. This assumption is unrealistic if the processors are owned by autonomous, self-interested organizations that have no a priori motivation for cooperation and they will manipulate the algorithm if it is beneficial to do so. In this paper, we address this issue by designing a distributed mechanism for scheduling divisible loads in tree net-works, called DLS-T, which provides incentives to processors for reporting their true processing capacity and executing their assigned load at full processing capacity. We prove that the DLS-T mechanism computes the optimal allocation in an ex post Nash equilibrium. Finally, we simulate and study the mechanism under various network structures and processor parameters.

• Research Organization: Pacific Northwest National Laboratory (PNNL), Richland, WA (US)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-76RL01830
• OSTI ID: 1439043
• Report Number(s): PNNL-SA-84721
• Journal Information: Journal of Parallel and Distributed Computing, Journal Name: Journal of Parallel and Distributed Computing Journal Issue: 3 Vol. 72; ISSN 0743-7315
• Publisher: Elsevier
• Country of Publication: United States
• Language: English

Related Subjects

Algorithmic mechanism design
Game theory
Parallel Computing
Scheduling