scholarly journals A Game-Theoretic Framework for Resource Sharing in Clouds

Author(s):  
Faheem Zafari ◽  
Kin K. Leung ◽  
Don Towsley ◽  
Prithwish Basu ◽  
Ananthram Swami
Water Policy ◽  
2016 ◽  
Vol 19 (3) ◽  
pp. 479-495 ◽  
Author(s):  
Dagmawi Mulugeta Degefu ◽  
Weijun He ◽  
Jian Hua Zhao

Designing a feasible and stable water sharing mechanism for transboundary river basins is a big challenge. The stochastic and uncertain characteristics of water flow in these rivers is among the main reasons which make the formation of cooperative coalitions with feasible water allocations and self-enforceable allocation agreements difficult. When the water in these river basins is scarce the task becomes even more challenging. This article focuses on the application of stochastic game theoretic extension of the bankruptcy concept to transboundary water resource sharing under water scarce and uncertain conditions. Among the water allocation vectors obtained from stochastic bankruptcy rules only the ones from the stochastic constrained equal awards rule were self-enforcing under uncertainty. Furthermore, the authors also proposed an allocation rule that can be used under a stochastic setting. The proposed rule provides water allocations that are self-enforcing in the absence of uncertainty. Generally, the application of the stochastic bankruptcy approach could be a source of important strategic information which can serve for the sustainable sharing and management of these vital sources of fresh water, particularly during water scarcity.


2014 ◽  
Vol 2014 ◽  
pp. 1-14 ◽  
Author(s):  
Xin Xu ◽  
Huiqun Yu

On-demand resource management is a key characteristic of cloud computing. Cloud providers should support the computational resource sharing in a fair way to ensure that no user gets much better resources than others. Another goal is to improve the resource utilization by minimizing the resource fragmentation when mapping virtual machines to physical servers. The focus of this paper is the proposal of a game theoretic resources allocation algorithm that considers the fairness among users and the resources utilization for both. The experiments with an FUGA implementation on an 8-node server cluster show the optimality of this algorithm in keeping fairness by comparing with the evaluation of the Hadoop scheduler. The simulations based on Google workload trace demonstrate that the algorithm is able to reduce resource wastage and achieve a better resource utilization rate than other allocation mechanisms.


Author(s):  
Faheem Zafari ◽  
Kin K. Leung ◽  
Don Towsley ◽  
Prithwish Basu ◽  
Ananthram Swami ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document