Minimum process communication induced checkpointing for mobile computing

Abstract

Mobile computing is a computing era in which all nodes in a network are mobile and self-configured nodes capable of communicating with one another via wireless links. Because of the limited availability of resources such as battery health, power, space, and energy with each node, the mobile nature of the nodes creates many challenges. The presence of these nodes in the system may cause faults, necessitating the development of an efficient fault tolerant system capable of operating even in the presence of faults. In this paper, a checkpointing-based recovery mechanism is proposed in order to make the system fault tolerable so that it can continue to operate even in the presence of a fault. The dynamic value of the timing interval was used in the proposed work to determine the best, average, and worst cases of taking checkpoints. This system operates in two phases, one coordinated and one uncoordinated, to reach the checkpoints that classify it as a quasi-synchronous system. The study’s performance was evaluated based on the number of checkpoints taken while varying the number of nodes, as well as the computational cost and latency.

Subject Classification:

• 6804

Keywords:

• Checkpointing
• Fault tolerance
• Mobile computing
• Recovery
• Fault