A Scheduling Method for Multiple Content Searching Using Dynamic Prediction of Optimum Amount of Idle Computational Resource in Large-Scale Networks
Abstract— This paper proposes a search scheme for multiple target content items with a priority, which is 1stored in a lot of nodes scattered in a communication network, using idle computational resources. In our previous study, our proposed scheme successfully reduced searching costs such as searching time and communicat ion load by defining the utilization rate of all computational resources as a parameter, and setting its appropriate value according to the searching environment. In this paper, we assume that the appropriate parameter is unknown, and we propose a scheme which dynamically predicts optimal parameter value from the costs of the initial search results, and continues to make a searching schedule with correcting the optimal parameter value. The effectiveness of the proposed scheme is evaluated by computer simulations.
Index Terms— content search, dynamic prediction, idle computation resource, grid computing
Nagoya Institute of Technology, JAPAN
Chiba Institute of Technology, JAPAN
Cite: Chikara Miura, Shinji Sugawara, "A Scheduling Method for Multiple Content Searching Using Dynamic Prediction of Optimum Amount of Idle Computational Resource in Large-Scale Networks," Proceedings of 2019 the 9th International Workshop on Computer Science and Engineering WCSE_2019_SPRING, pp. 132-141, Yangon, Myanmar, February 27-March 1, 2019.