Performance Oriented Scheduling and Allocation Technique in Edge-Fog-Cloud Collaborative Environment

Nowadays, smart devices are becoming more prevalent in industrialized countries generating requests that require computational processing. Recently, collaborative edge-fog-cloud computing networks have been developed to allocate users’ requests to computing resources. However, scheduling these requests while accounting for user requirements and limited resources remains a challenge. This study proposes a structured planning at multiple levels in a collaborative edge-fog-cloud environment to allocate and schedule requests, aiming to reduce network latency. So, an Integer Programming (IP) formulation is developed to minimize network latency for users. Some network limitations are considered in the model, such as network logic for directing requests to computational resources, meeting deadline and nodes capacity constraints. Additionally, constraints related to processing allowable workload volume are integrated into the model. This strategy changes the workload distribution among the edge, fog, and cloud layers to approximately 24%, 27%, and 47%, respectively, creating a more balanced workload distribution and reducing workload traffic. Other results indicate that simply increasing the computational capacity of the fog nodes does not always improve network performance. This suggests the need for a more analytical approach, considering additional factors simultaneously in the underlying network. These outcomes underscore the efficiency and practical significance of the proposed model in a collaborative edge-fog-cloud computing landscape. The findings can help cloud service enterprises in providing efficient services for addressing the request scheduling and allocation challenges in edge-fog-cloud networks.