An effective failure recovery mechanism for SIP/IMS services

There are many challenges to provide telecommunication services. The reliability is one of the most important challenges. Telecommunications systems require 99.999% reliability, which means that system only tolerant of downtime about 5 minutes per year. To achieve such high reliability, multiple servers and a shared database are deployed for a Session Initiation Protocol (SIP)/ IP Multimedia Subsystem (IMS) service. This paper investigates the effects of different database (DB) access mechanisms such as Write-Through (WT), Write-Back (WB), and the database-only (DB-only) mechanisms. Then this paper proposes an enhancement for the WB mechanism to improve the successful call setup probability. The analytic models are derived to analyze the performance of the proposed mechanism.