Authors: Bing Tang
Addresses: School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan, Hunan, 411201, China
Abstract: Market mechanism and economic model have been proven to be efficient to distributed grid resource management. The economy-based deadline and budget constrained scheduling algorithm with cost-optimisation and time-optimisation reveals the problem of load unbalanced. A new dynamic resource price-adjusting (RPA) strategy in grid environment is presented. On the basis of price fluctuating methodology and load balance evaluation method, a market-driven guided grid resource scheduling algorithm is proposed, which considers about RPA-cost optimisation and RPA-time optimisation. New algorithm is compared with traditional deadline and budget constrained algorithm in scenarios of a single user and multiple users, and the experimental comparison results show that the new algorithm is feasible which improves the performance of grid in aspect of not only increasing the task accomplishment ratio, but also solving the problem of load unbalanced.
Keywords: grid resource scheduling; quality of service; QoS; economic models; dynamic pricing; price adjustment; modelling; price fluctuation; load balancing; cost optimisation; time optimisation; economy-based deadlines; budget constrained scheduling; economics; grid computing.
International Journal of Information and Communication Technology, 2012 Vol.4 No.2/3/4, pp.132 - 153
Available online: 29 Aug 2012 *Full-text access for editors Access for subscribers Purchase this article Comment on this article