Pre-task Assignment with Decentralized Resource Management Strategy for Heterogeneous Distributed System

摘要

Heterogeneous Distributed Computing (HDC) systems, which exploit the aggregate power of a network of workstations (WSs) and Personal Computers (PCs) are an inexpensive alternative to the dedicated parallel super computing systems. As these systems are widely available in academic and industrial environments, it is becoming popular to use these computing resources to solve time consuming applications. The main problem with such type of cluster computing environment is the continuous change in the performance of individual WSs/PCs which requires an efficient task partitioning, scheduling, and load balancing to get better performance. In this paper, we present Adaptive Decentralized resources management strategy along with Pre-task Scheduling (ADPS) strategy for a heterogeneous distributed raytracing system. The DPS strategy has following main features: i) the master pre-task assignment to each node/worker, this attempt to reduce the inter-process communication time and load imbalance which occurred in traditional Runtime Task Scheduling (RTS) strategies, ii) It adaptively sizes the sub-tasks size for workers according to their performance. Hi) Performances of the RTS and ADPS strategies are evaluated on manager/master and workers model of HDC system. The measured results of our proposed (ADPS) strategy have shown a significant improvement in speedup over RTS strategy.