Abstract—Cloud computing is a transformative technology in distributed computing. As a result of the increasing demand for accessing data through the Internet by various endpoints, such as laptops, smartphones, and mobile and wearable devices, cloud services continue to grow and thrive. However, the extensive use of cloud-based services increases the risks of failure, compromising the service’s reliability and availability. To manage failures, cloud computing systems must be fault-tolerant. Although there are many techniques for fault tolerance in cloud computing, they lack a holistic solution for multiple layers of the cloud infrastructure. Additionally, as the complexity of the cloud infrastructure increases, there is an urge to provide an intelligent strategy to deal with failures. In this investigation, we introduce an Intelligent Fault-Tolerant Strategy (IFTS) that will respond to cloud failures efficiently. The proposed IFTS provided a solution for different cloud infrastructure layers. The strategy was simulated using the CloudSim simulator, and the main results of our experiments are as follows; host selection process in IFTS provided a better makespan by 11.57% than random selection and by 22.05% than the worst-fit process, with a 0% failure rate. IFTS cloudlet to Virtual Machine (VM) mapping process managed to decrease the failure rate by 28.08% in random mapping and 15.46% in available VM mapping to 5.91%, which is a huge difference and improvement. The results for application-level Fault Tolerant (FT) show that IFTS makespan is improved by 27.52% when using replication and by 20.83% when using checkpointing.
 
Keywords—cloud computing, Fault Tolerant (FT), host reliability, Virtual Machine (VM) allocation, replication, checkpointing


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).