Joint multi-server cache sharing and delay-aware task scheduling for edge-cloud collaborative computing in intelligent manufacturing

The rapid advancement of intelligent manufacturing has led to an increasing demand for computing resources in industrial computing tasks. As a new computing paradigm, edge-cloud collaborative computing (E3C) fills the delay gap of traditional cloud computing for industrial computing tasks. Nevertheless, the E3C performance is heavily contingent upon task scheduling, which plays a pivotal role in influencing the effectiveness of E3C task execution. In this paper, we tackle the task scheduling problem by introducing a novel scheduling model and algorithm. Firstly, we establish a task scheduling optimization model to precisely carve the joint multi-server cache sharing and delay-aware task scheduling problem. We formulate the joint task scheduling model as a constrained combinatorial optimization problem and prove its NP-hardness. Simultaneously, given the heightened security requirements of manufacturing E3C compared to conventional E3C, we address the task security concerns during the scheduling process by incorporating task privacy levels and encryption techniques to safeguard the shared task caches in the established model. Secondly, to solve the near-optimal joint strategy composed of scheduling, caching and sharing strategies derived from the established model, we propose a scheduling algorithm based on the improved artificial bee colony algorithm. Finally, we conduct extensive experiments to verify our scheduling model and algorithm. Experimental results substantiate that our multi-server cache sharing mechanism can further decrease the task execution delay by 31.13% in comparison to the conventional task scheduling. Furthermore, the proposed scheduling algorithm demonstrates superior performance in terms of solution accuracy compared to existing algorithms.