A non-intrusive minimum prociess synchronous checkpointing protocol for mobile distributed systems
Parveen Kumar, Lalit Kumar, Ruby Chauhan, Vishal Gupta · 2005
Mobile computing raises many new issues, such as lack of stable storage, low bandwidth of wireless channels, high mobility and limited battery life. These issues make traditional checkpointing algorithms unsuitable for checkpointing mobile distributed systems. Minimum process coordinated checkpointing is a good approach to introducing fault tolerance in a distributed system transparently. This approach is domino-free and requires at most two checkpoints of each process on stable storage, and forces only interacting processes to checkpoint. Sometimes, it also requires piggybacking of information onto normal messages, blocking of the underlying computation or taking some useless checkpoints. In this paper, we propose a non-intrusive minimum process synchronous checkpointing protocol for mobile distributed systems, where only the minimum number of tentative checkpoints is taken. We also optimize the number of useless forced (mutable) checkpoints and message overheads as compared to Cao et al. (2001).