| Abstract [eng] |
The incident management system is understood as a system that integrates together many systems like customer relationship management (CRM), network information system (NIS), payment system and field work management related system. It is obvious that the infrastructure architecture has great impact in the availability solution, but if the application design is based only on the fact that underlying architecture is good enough (it might be clustered, duplicated, backed up, etc.), then it is a great risk that application might fail under critical situation. The incident management system will more likely to survive critical situation, if: - Prime design principles are based on service oriented architecture (SOA) principles; - Integration between the systems is based on the asynchronous messaging pattern; - Possible latency and delay issues are included into availability evaluations; - Each software node is designed as autonomous as possible; - Client node software does not relies on the central storage and is able to exchange information with other client nodes; - Availability network is defined and network node measurement is in place; - Business procedures for handling critical situations are present and executable;. |
