ABSTRACT
Significant investments and technological advances have established highperformance computational (HPC) Grid infrastructures as dominant platforms for large-scale parallel/distributed computing in science and engineering. This chapter focuses on the CometCloud engine and its autonomic scheduler and autonomic application management framework. It describes the hybrid HPC-Grids/Clouds infrastructure and presents the scientific application workflow that is used to drive the research. The chapter investigates usage modes for the hybrid Grids/Clouds infrastructure and explores infrastructure and application adaptivity. Computing infrastructures are moving towards hybridization, integrating different types of resource classes such as public/private clouds and grids from distributed locations. The chapter presents the goal of the experiments to investigate how possible usage modes for hybrid HPC Grids-Cloud infrastructure can be supported by a simple policy-based autonomic scheduler. The approach and infrastructure used can manage the heterogeneity and dynamics in the behaviors and performance of the platforms.
