An Optimized Parallel Computing Paradigm for Mobile Grids Based on DSPOM

Rajagopal, Aghila; Mohamed, M. A. Maluk

doi:10.4304/jetwi.6.1.119-132

Cited by 1 publication

(1 citation statement)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cluster is normally used to improve computing speed and reliability of one single computer. Through load balancing [2,3] and parallel computing [4] technology, the cluster achieves higher efficiency, so it tends to be widely used in railway information system [5], command control [6], road and bridge tunnel health monitoring [7], etc. Because of its special targets and tasks, the seismic profession needs to deploy seismic monitoring nodes in various regions.…”

Section: Introductionmentioning

confidence: 99%

Research on the Key Technology of Seismic Observation Equipment Cluster Monitoring

Wang¹,

Yong²,

Zhao³

et al. 2015

TOAUTOCJ

View full text Add to dashboard Cite

This paper proposes a reactive monitoring system model, which is suitable for seismic observation equipment cluster. By analyzing two key factors affecting the performance of the system under our proposed model, we pose a priority queue based (PQB) polling mechanism as well as the volatility-removed (VR) fault alarm mechanism for the cluster node monitoring system. Our novel PQB polling mechanism initializes priorities of cluster nodes according to their importance and then puts cluster nodes into the polling queue according to their priority order. Nodes with same priority follow FCFS algorithm, thus solving the problems that important cluster nodes be monitored first. A formula judging whether cluster node is in the fluctuant state is given by VR fault alarm mechanism, which improves the accuracy of the monitoring results. Finally, we build an experimental platform to analyze the performance of PQB polling and VR warning mechanisms. Results show that PQB polling mechanism increases the number of polling nodes than that of FCFS polling mechanism; meanwhile, VR warning mechanism is more close to the actual fault results than normal alarm mechanism. Furthermore, PQB and VR mechanisms have no significant impact on server performance.

show abstract

Section: Introductionmentioning

confidence: 99%