Analysis and Design of Metrics for Autonomic Computing

Abstract-- Today the computer software/systems have become extremely complex. The increased need to distribute data, applications and system resources across geographical boundaries has led to complexity of the software applications. This situation results in overall increase in maintenance efforts and cost. There is a need to have systems which are self-aware, self-protecting, self-healing, self-optimizing, and self- configuring. This would help in cheaper and faster maintenance. Such computing systems and infrastructure would take care of managing themselves. In an autonomic environment the IT infrastructure and its components are self-managing leading to reduced cost of owning and operating computer systems. The present paper does a study of existing quality characteristics and corresponding metrics, and based on analysis proposes a new set of characteristics as well as metrics to design a robust autonomic computing system. Keywords-Autonomic computing, self-aware, self-protecting, self-healing, self-optimizing, and self-configuring.