Solving the Peak Load Problem of Online Charging System using Dynamic Reserve

In Online Charging System (OCS), to do charging and billing for many millions of customers, the availability of the system is important. At the peak load moment, a huge number of messages are sent to OCS simultaneously, causing performance degradation OCS. In this paper, we introduce a method using Dynamic Reserve to solve the peak load problems and improve the system's performance. The method uses a rule-based approach and combines both system status and business rules to bring really good performance. It improved the performance when reduced by 60% transactions per second (tps) and credit control request messages (CCR). Because of the rule-based approach, the method is reliable and is applied to real systems.