The performance of parallel applications is fundamentally dependent on the agreement of the task execution on processors available in Distributed Real Time System (DRTS). For such type of large and multifaceted real-time parallel applications, DRTS is one of the best platforms for execution. The parallel application is divided into multiple tasks which is to be executed over the nodes available in DRTS. These tasks have to be processed in relatively short periods of time, or the computation in real-time fashion is required. Using efficient load sharing and clustering policy can improve the performance of the system. The task allocation algorithm, which is NP-hard in complexity, must minimize the inter-task communication which can be achieved by clustering, and on the other hand, the clusters must be allocated on the appropriate node in DRTS. The algorithm proposed in this paper has used static load sharing policy and an effective clustering mechanism. The proposed algorithm was illustrated using a practical example and simulated in MATLAB 7.11.0.

A Distributed Real Time System (DRTS) is a set of nodes connected by a real time communication network that interact with each other in order to accomplish a common task. Each node has computation facility and its own memory, while the communication network has a limited communication capacity. DRTS has come out as a great platform for high performance parallel applications (Kopetz, 1997). The high performance environments presented by parallel and distributed computing system are much capable to provide high capacity of processing (Richard et al., 1982; Chien-Chung and Wem- Hsiang, 1985; Siva Ram, 1992/93; and Anurag and Vikram, 2012). For realization of such capacitive system, efficient task allocation algorithms under appropriate load distribution schemes must be employed in a very efficient manner (Karim et al., 2002).

The parallel application is divided into multiple tasks and these tasks are to be allocated on the nodes of distributed system for execution. Parallel application’s performance over distributed system depends on the assignment of tasks on available nodes or processors or workstations on the DRTS. If the implementation of allocation strategy has not been made carefully, the CPU time cannot be used properly and it results in the downgrade of the performance of the parallel application over distributed system (Gamal Attiya and Yskandar, 2006).

Information Technology Journal, Distributed Real Time System (DRTS), Task Allocation, Cluster, Parallel Application, Load Sharing, NP-Hard.