Skip to Main Content
In this paper we present an innovative decentralized dynamic resource scheduling solution of large scale application workflows onto distributed, heterogeneous grid environments. The presented resource allocation solution aims to optimize large-scale scientific application workflows by efficiently and effectively mapping and scheduling them onto grid resources. The proposed solution is part of a framework that aims to help scientists easily deploy large-scale scientific workflow applications from a wide-range of research fields. The decentralized architecture, together with the scheduling policies and fault management being proposed in this paper, are specifically designed to optimize the scheduling management for workflows in case of a wide range of grid environment. In order to demonstrate the validity and performance of the presented scheduling solution we propose using modeling and simulation techniques. For that we present the characteristics of a generic simulator that includes all the necessary components to help in evaluating a wide-range of scheduling procedures, such as the ones presented in this, in the context of large-scale heterogeneous grid environment.