An Adaptable Approach to Fault Tolerance in Cloud Computing

Existing fault tolerance approaches in the cloud are broadly based on replication and checkpointing. Each of these has its advantages limitations. This paper presents an adaptable method for determining which two will be appropriate successful execution a task given conditions. The proposed classifies failure risk host machines available their history. Subsequently, fuzzy logic is used to determine approach by considering host's risk, user-defined task's priority, level resource redundancy. Setting priority provides user with control solicit desired while availability resources reflects provider's capability offer tolerance. Simulation experiments have verified that proactive selection fault-tolerance increases number tasks complete successfully.