Distributed method invocations require dynamic scheduling algorithms and efficient resource projections to provide timeliness guarantees to application objects. In this paper we present a dynamic scheduling algorithm that examines the computation times, real times and resource requirements of the application tasks to determine a feasible schedule for the method invocations. The schedule is driv...

A distributed system is a collection of independent computers that appears to its users as a single coherent system. Depending on job arrival rate some of the computing nodes may become overloaded while some other nodes may sit idle. The imbalance in load distribution may affect overall system performance. So proper task scheduling algorithm is needed. Real time task scheduling algorithm must t...

In this paper, we present a bio-inspired job scheduling mechanism that enables the adaptation of largescale, naturally parallel and compute-intensive Monte Carlo tasks to clustered computational farms. Examples of such farms include large-scale computational grids, with heterogeneous and dynamic performance. The kernel of this scheduling mechanism is a swarm intelligent algorithm, which is insp...