Outlier detection based fault-detection algorithm for cloud computing

Abstract

Outlier detection is becoming a recent area of research focus in data mining. In Cloud Computing, we use all resources as a service and these services should be very efficient, robust and corrective. Here we present FDACC (Fault Detection Algorithm for Cloud Computing) for faulty services using outlier detection method that can help to detect accurate and novel faulty services without any knowledge. FDACC implemented on this framework that has three components- cloud nodes, FDS (Fault Detection System) and End User. FDS is connected with end user and n number of nodes in cloud. FDS is able to detect all non-working services as well as detect faulty services and machines in cloud environment.