Adaptive Checkpointing for Emergency Communication Systems

Abstract: The purpose of an emergency communication system is to be ready andavailable at all times during an emergency situation. This means that emergencysystems have specic usage characteristics, they experience long idleperiods followed by usage spikes. To achieve high availability it is importantto have a fault-tolerant solution. In this report, warm passive replication isin focus. When using warm passive replication, checkpointing is the procedureof transfering the current state from a primary server to its replicas. Inorder to utilize resources in a more eective manner compared to when usinga xed interval checkpointing method an adaptive checkpointing method isproposed. A simulation-based comparison is carried out using MATLABand Simulink to test both the proposed adaptive method and the xed intervalmethod. Two metrics, response time and time to recover, and fourparameters are used in the simulation. The results show that an adaptivemethod can increase eciency, but in order to make a good adaptive methodit is necessary to have specic information regarding system congurationand usage characteristics.