No abstract

SUMMARYScheduling and load balancing have been one of the key issues in distributed systems due to their significant impact on system performance as well as system resource utilization. With the rapid development of Peer-to-Peer (P2P) network technology, traditional scheduling and load balancing techniques are facing new challenges because P2P-based systems require more dynamic mechanisms for task scheduling and load balancing among heterogeneous network computing nodes. In this paper, we propose a scheme for dynamic scheduling and load balancing in a P2P-based environment. Our scheme aims at ServiceOriented P2P-based distributed systems, however, it can be applied to traditional distributed architecture straightforwardly. Furthermore, we evaluate the performance of our scheme using simulation experiments in a cluster-based distributed computing environment. Indeed, our results show that our scheme can achieve significant system performance gain compared to commonly used random and round robin scheduling algorithms in P2P-based systems. Meanwhile, we use a real-time distributed simulation to evaluate our algorithm, which can provide more trustful results in terms of performance evaluation. Indeed, the ease of using a distributed real-time simulation, to measure the real-time performance of the algorithm can promote more flexible and efficient methods for the development of load balancing algorithms. This is especially the case for the hard-to-predict P2P-based distributed systems.

Dynamic load balancing is a key factor in achieving high performance for large scale distributed simulations on grid infrastructures. In a grid environment, the available resources and the simulation's computation and communication behavior may experience critical run-time imbalances. Consequently, an initial static partitioning should be combined with a dynamic load balancing scheme to ensure the high performance of the distributed simulation. In this paper, we propose a dynamic load balancing scheme for distributed simulations on a grid infrastructure. Our scheme is composed of an online network analyzing service coupled with monitoring agents and a run-time model repartitioning service. We present a hierarchical scalable adaptive JXTA service based scheme and use simulation experiments to demonstrate that our proposed scheme exhibits better performance in terms of simulation execution time. Furthermore, we extend our algorithm from a local intra-cluster algorithm to a global inter-cluster algorithm and we consider the proposed global design through a formalized Discrete Event System Specification (DEVS) model system