Enabling Autonomic Grid Applications Dynamic Composition, Coordination and Interaction.pdfVIP

Enabling Autonomic Grid Applications: Dynamic Composition, Coordination and Interaction Zhen Li and Manish Parashar The Applied Software Systems Laboratory, Rutgers University, Piscataway NJ 08904, USA {zhljenny, parashar}@ Abstract. The increasing complexity, heterogeneity and dynamism of networks, systems and applications have made our computational and information infras- tructure brittle, unmanageable and insecure. This has necessitated the investiga- tion of an alternate paradigm for system and application design, which is based on strategies used by biological systems to deal with similar challenges of complex- ity, heterogeneity, and uncertainty, i.e. autonomic computing. Project AutoMate investigates conceptual models and implementation architectures to enable the development and execution of self-managing applications. It supports the defi- nition of autonomic elements, the development of autonomic applications as the dynamic and opportunistic composition of these autonomic elements, and the pol- icy, content and context driven execution and management of these applications. This paper introduces AutoMate architecture and describes the Rudder coordina- tion framework and its use in enabling autonomic behaviors. 1 Introduction The emergence of wide-area distributed and decentralized “Grid” environments, such as pervasive information systems, peer-to-peer systems, and distributed computational infrastructures, has enabled a new generation of applications that are based on seam- less access, aggregation and interactions. Examples include pervasive applications that leverage the pervasive information Grid to continuously manage, adapt, and optimize our living context, crisis management app