Industrial Internet of Things (IIoT)-enabled Decentralized Computation Offloading in Smart Factory

Abstract

<p>The development of the factory intelligence with Industrial Internet of Things (IIoT) poses a new challenge on embedded processor capability. This has led to the emergence of the Data-Massive and Latency-Sensitive Computing Tasks (DLCT) problem that urgently needs to be solved. Mobile Edge Computing (MEC) emerged as a transformative technology for enabling efficient and real-time computation in smart factory environments. In this paper, a new computation system model is proposed, the sub-tasks tasks are divided into data sets and instruction sets, further categorized into k types of sub-tasks. The introduction of a high-level cache in the Central Processing Unit (CPU) explores the impact of faster data access mechanisms compared to accessing data from the main memory. Additionally, a load balance algorithm is proposed for sub-task allocation, and the cache with load balance is tested to evaluate the performance of the proposed algorithm. The numerical study shows that 32% overall computing time decreased based on load balance in one time slot, the computation time is further reduced by introducing the cache mechanism.<br></p>