RT DL Tasks Distribution for Sensitive Data Protection and Resource Optimization
Farah Zeidan, Mostafa ElHayani, Hassan Soubra · 2023
With the rapid growth of data-intensive applications and the proliferation of the Internet of Things (IoT) devices, traditional cloud architectures face significant challenges in enabling real-time decision-making. Centralized cloud infrastructures have limitations in handling the increasing volume of data generated at network ends. In response, edge computing has emerged as a promising paradigm that brings computation closer to the data source, offering advantages over traditional cloud architectures. This paper presents an edge computing system that incorporates a load balancer to determine whether real-time tasks should be offloaded to the edge or to the cloud. To enhance security, a data retention measure is introduced to ensure sensitive data remains within the edge device, irrespective of other load-balancing conditions. This approach minimizes the risk of attacks and unauthorized access by containing sensitive data within the edge device. Furthermore, the impact on system performance is evaluated by monitoring CPU utilization before and after task execution. This research contributes secure and efficient task offloading strategies in edge computing, with implications for real-time processing and data privacy sensitive applications.