AutoGrid: Towards an Autonomic Grid Middleware

Computer grids have drawn great attention of academic and enterprise communities, becoming an attractive alternative for the execution of applications that demand huge computational power, allowing the integration of computational resources spread through different administrative domains. However, grids exhibit high variation of resource availability, node instability, variations on load distribution, and heterogeneity of computational devices and network technology. Due to those characteristics, grid management and configuration is error-prone and almost impracticable to be performed solely by human beings. This paper describes AutoGrid, an autonomic grid middleware built using Adapta reconfiguration framework and runtime system. AutoGrid introduces self-managing capabilities to the Integrade grid middleware, such as: context-awareness, self- healing, self-optimization and self-configuration. This paper also presents insights and experiments that show the benefits towards an autonomic grid infrastructure.