Robust Resource Allocation Algorithms in Real-Time Distributed System

Real time distributed systems (RTDS) operate in an environment that un- dergoes unpredictable changes causing certain system performance features to degrade. Such unpredictability can be the result of sudden machine failures, in- creases in system load, or errors caused by inaccurate initial estimation. These systems need robustness to guarantee limited degradation despite uctuations in the behavior of its component parts or environment. The robustness gives an idea of the stability of the solution with regards to another performance metric such as schedule length, load balance of an application, queue waiting time of batch scheduler, etc. Designing robust systems for such environments involves determining resource allocations that can account for uncertainty in estimated system parameters. Robustness of the system can be achieved with proper resource allocation.