An adaptive fault tolerance approach to enhance the execution of applications on multi-cluster grid configurations from mobile grid interfaces in wireless networks. (1st January 2011)

• Record Type: Journal Article
• Authors: Rossetto, Anubis G.M.; Rolim, Carlos Oberdan; Leithardt, Valderi R.Q.; Dantas, M.A.R.; Geyer, Cláudio F.R.
• Is Part Of: International journal of high performance systems architecture. Volume 3:Number 4(2011)
• Page Start: 202
• Page End: 215
• Publication Date: 2011-01-01
• Subjects: fault tolerance -- workflow -- mobile grid computing -- multi-cluster; Computer architecture -- Periodicals; Computer systems -- Periodicals; High performance computing -- Periodicals; 004.205
• DOI: 10.1504/IJHPSA.2011.045505
• Languages: English
• ISSNs: 1751-6528

Abstract

Mobile grid environments can be broadly divided into two different types of configurations. In the first case, devices are used as grid interfaces. In the second configuration, mobile devices are considered to be grid resources. However, the interaction between mobile devices and wired networks still has to overcome several difficulties before it can produce an operational mobile grid infrastructure, in either type of configuration. Examples of these challenges include the submission of distributed applications to execute tasks in a coordinated manner, frequent disconnections, restrictions on power consumption and the selection of suitable grid resources for application executions. In this paper, a research study is conducted involving an adaptive fault tolerance approach for grid interfaces in wireless networks to enhance the submission and execution of distributed applications. The approach examines submission, monitoring and management functions through the use of fault tolerance and workflow mechanisms, especially when disconnections occur and the mobile devices are executing an application. The empirical results show that the proposed approach can achieve a satisfactory level to support grid interfaces in a wireless network, and provide a more robust environment for application executions with less power.