Please use this identifier to cite or link to this item:
作者: Jea, K.F.
Wang, J.Y.
關鍵字: Optimization;Gradient;Distributed computing;Parallel computing;Average waiting time;data allocation;multiple channels;data grids;broadcast;system;availability;optimization;replication;environment;algorithm
Project: International Journal of Innovative Computing Information and Control
期刊/報告no:: International Journal of Innovative Computing Information and Control, Volume 6, Issue 9, Page(s) 3887-3909.
In this paper, we consider an optimization problem that aims to minimize the average waiting time for distributed services with different processing complexities and access probabilities. It is motivated by the fact that there are many large-scale scientific projects and commercial applications (e.g., image processing in astronomy), and their waiting time needs to be lowered down in order to maintain customer satisfaction. We first demonstrate several useful properties of this problem by mapping it to the Euclidean space R(n). Utilizing them, we then develop a gradient-based method for dividing and distributing services to multiple machines. The theoretical analyses show that the proposed method converges linearly and the resultant average waiting time is near optimal. Finally, we present experimental results that confirm the convergence speed and solution quality of the proposed method. Using the proposed method, a Service provider requires only a little execution time to deploy his/her services on multiple machines and provides users with a near-optimal average waiting time for their service requests. The proposed method can be extended to other similar optimization problems (e.g., vehicle routing problem) and promisingly achieves the same near-optimal results.
ISSN: 1349-4198
Appears in Collections:資訊科學與工程學系所

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.