High availability and fault management in Objective Architecture systems

At the heart of many real-time and near real-time distributed computing systems, such as radar, navigational, missile defense, and command and control, you find mission critical requirements. This designation demands continuous system availability even in the face of system faults. Lives are at stake and mission success is at risk. Standards-based high availability and computing resource management solutions with well defined functional boundaries and interfaces are gaining traction. This is seen in such Navy examples as Aegis Modernization, Littoral Combat Ship, Common Processing System, and more recently within the PEO-IWS Objective Architecture DoD Information Technology Standards Registry (DISR). As implemented, Objective Architectures hold the promise of providing a path to meet current and future demands placed on these system designs. Examples of these demands include: increased functional density and complexity, resource consolidation, mixed criticality systems and the ever present size, weight and power concerns. The architectural approach must be both flexible and scalable to fit the broadest set of cases, minimizing or eliminating special-case solutions. There are a set of standards published by the Service Availability (SA) Forum that address these issues. SA Forum standards are currently being used in the programs mentioned above. We will also consider how SA Forum specifications can be used to support operational availability in legacy systems while laying the groundwork for evolution of the objective architecture. This paper reviews the above issues and the specific benefits brought about by a SA Forum-based solution.