摘要：Among a lot of rollback-recovery protocols developed for providing fault-tolerance for long-running distributed applications, sender-based message logging with checkpointing is one of the most lightweight fault-tolerance techniques to be capable of being applied in this field, significantly decreasing high failure-free overhead of synchronous logging by using message sender's volatile memory as storage for logging its corresponding message. However, attempting to apply this technique into large-scale and geographically distributed systems such as broker-based sensor networks, etc., the following problems should be addressed; reducing the number of control and data messages passing on core networks occurring during its fully message logging and recovery procedures. In this paper, we present a lightweight message logging protocol for distributed sensor networks to solve all of them by employing the current and future distributed systems' architectural features mentioned above.
