Digital Twin-Empowered Service Provisioning Framework for Cloud-Assisted Internet of Things

The rapid growth of Internet of Things (IoT) devices has led to an exponential increase in data generation, causing significant challenges for service providers to manage data efficiently. Edge cloud collaboration has emerged as a solution to process large volumes of data generated by IoT networks so that users can offload computation services to the edge or cloud. Although cloud infrastructure is capable of handling spikes in user demand, the need for efficient provisioning of cloud resources becomes more critical for latency-sensitive services during peak workloads. In this regard, in this paper, we propose a digital twin-assisted service provisioning framework for cloud-oriented services in IoT. By integrating the digital twin with cloud-assisted IoT, we aim to provide dynamic allocation of cloud resources with data-driven provisioning decisions according to Quality of Service (QoS) requirements. We then develop a service provisioning algorithm that schedules cloud resources to minimize service completion time and processing time in the cloud. Finally, we evaluate the performance results of the proposed framework and show that the performance of the proposed algorithm outperforms conventional provisioning approaches.