Are Service Oriented Architectures the Only Valid Architectural Approach for the Transformation to Network Centric Warfare

摘要

The use of a Service Oriented Architecture (SOA) as the dominant single architectural design paradigm of Network Centric Warfare (NCW) introduces architectural infrastructure stability risk levels which may be unacceptable in C4ISR mission frameworks. Service reliability, service performance, and service availability represent a key set of requirements which must be satisfactorily implemented or mission risk will accelerate. The purpose of this research is to demonstrate that a complex adaptive architecture solution to the reliability, performance, and availability requirements imposed by NCW transformation provides a richer and more stable approach to DoD legacy capability exploitation than does a pure SOA approach. Using the Network Centric C4ISR Architecture Quality of Service (QoS) rating scale developed for this research paper as a behavioral classification methodology, the analysis concludes that a complex adaptive architectural model composed of at least Event Driven Components, Service Oriented Components, MOMS Components and GRID Components may present a more risk tolerant solution in terms of satisfactory implementation of NCW reliability, performance, and availability requirements. In simpler terms, a stand-alone SOA will be insufficient in terms of providing infrastructure stability. I propose that a highly available, disaster recoverable, GRID model (overlain with availability and performance monitoring agents) be implemented in order to sufficiently cover the reliability, performance, and availability issues needed for combat missions.