Resource monitoring and state prediction for mobile cloud platform

Date

2017-04-24

Authors

Yang, Xuesong

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Mobile Cloud Computing (MCC) is a novel technology that combines mobile device with cloud computing. However, it has some inherent shortcomings compared with traditional cloud computing. For example, there will be unstable communications, intermittent connections, and limitation of power supply. Also unpredictably resources can join and leave cloud environment which makes mobile cloud operation and management more and more complex. In order to obtain stable resource participation of mobile devices, a combined push-pull method [11] has been proposed for collecting and analyzing dynamic information of mobile devices. At monitored nodes, a periodically push model is used to timely send the loading resources to the monitoring components. At monitoring nodes, a pull model is triggered by the change degree. If the change degree is greater than the threshold or the update loading information is missed during the time interval, the monitoring component immediately sends a query to monitored node for the latest loading information. The experimental results show that the model can effectively improve the monitoring performance for mobile environment and reduce communication overhead.

Description

Keywords

Resource Monitoring, State Prediction, Mobile Cloud Computing, Push model, Pull model, Push-Pull Hybrid Model, Change Degree, User Tolerant Degree, Exponentially Weighted Moving Average

Citation