Response time analysis of fault tolerant systems using primary site approach with LCFS repair service

The primary site approach uses redundancy for achieving fault-tolerance against node failures. The service to be made fault-tolerant is replicated in many sites, out of which one is selected as the primary site. The other sites on hot standby are backups. When a primary site fails, one of the backup sites takes over as the primary site while the failed primary site joins the repair queue. Two types of repair services are considered, namely the delayed repair service and the immediate repair service. We analyze the average response time of the system when the repair server uses the last-come, first-served (LCFS) service discipline to service the failed sites. We compare the derived response time with the average response time of a system with a repair server that uses the first-come, first-served (FCFS) service discipline