Monitoring service using Markov Chain model in mobile grid environment

Abstract

Recently mobile devices become considered as grid resources as the technology of mobile devices and wireless communication network improves. But mobile devices have several problems such as instability of wireless communication, intermittent connection, limitation of power supply, and low communication bandwidth. These problems make it difficult to use mobile grid computing resources for stable job processing effectively. For stable resource participation of mobile devices, a monitoring scheme that collects and analyzes dynamic information of mobile devices, such as CPU, memory, storage, network and location, is required. But, if the time interval of monitoring is very short, overhead of collecting information increases. The scheme, however, cannot keep correct state information in dynamic environments if the interval is very long. This paper proposes a monitoring service scheme that adjusts the time interval of monitoring according to the state information predicted by Markov Chain model. © 2010 Springer-Verlag Berlin Heidelberg.