Fault tolerance aware scheduling technique for cloud computing environment using dynamic clustering algorithm

In cloud computing, resources are dynamically provisioned and delivered to users in a transparent manner automatically on-demand. Task execution failure is no longer accidental but a common characteristic of cloud computing environment. In recent times, a number of intelligent scheduling techniques have been used to address task scheduling issues in cloud without much attention to fault tolerance. In this research article, we proposed a dynamic clustering league championship algorithm (DCLCA) scheduling technique for fault tolerance awareness to address cloud task execution which would reflect on the current available resources and reduce the untimely failure of autonomous tasks. Experimental results show that our proposed technique produces remarkable fault reduction in task failure as measured in terms of failure rate. It also shows that the DCLCA outperformed the MTCT, MAXMIN, ant colony optimization and genetic algorithm-based NSGA-II by producing lower makespan with improvement of 57.8, 53.6, 24.3 and 13.4 % in the first scenario and 60.0, 38.9, 31.5 and 31.2 % in the second scenario, respectively. Considering the experimental results, DCLCA provides better quality fault tolerance aware scheduling that will help to improve the overall performance of the cloud environment.