Design and implementation of clusters monitor system based on Android

Abstract

With the development of high performance computing (HPC), clusters play a more and more important role in a lot of scientific fields such as graphics, biology, physics, and climatology etc. Therefore, monitoring becomes more critical for sufficient utilization of clusters considering the limited availability of HPC resources. However, the traditional clusters monitoring systems have a common disadvantage of poor mobility which limits the efficiency of cluster management. For this reason, the mobile clusters monitoring system designed on Android platform is presented in this paper that will make it possible to monitor the whole cluster anywhere and anytime to allow administrators to manage, diagnose, and troubleshoot cluster issues more accurately and promptly. The monitoring system is developed on the popular monitoring tools, Ganglia and Nagios, to collect required information from server and display on client - mobile phone. The remote monitoring system also provides the communication module and the other applications such as instant messaging, global positioning and shift changing notice. This system makes clusters monitoring more flexible, efficient, and convenient. Most importantly, the method described in this paper can provide a possibility to customize and extend functions based on user requirements. © 2013 Springer-Verlag.