.png){kind=icon}
.png){kind=icon}
Return
A preemptive task unloading scheme based on second optional unloading in cloud-fog collaborative networks
.png){kind=icon}
PRE
AI
DOI：10.1016/j.comcom.2025.108315.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Long-distance data transmission between Internet of Things (IoT) devices and remote cloud center often leads to unacceptable latency for certain tasks. Fog computing has emerged as a promising solution for low-latency tasks. Consequently, the concept of cloud-fog collaborative networks has garnered significant attention. However, existing research primarily focuses on heterogeneous tasks, overlooking the crucial aspect of considering both task priority and second unloading. To address this gap, this paper proposes a novel task unloading scheme that concurrently takes preemptive priority and second optional unloading into account. In this scheme, delay-sensitive tasks (DSTs) are given preemptive priority over delay-tolerant tasks (DTTs). Furthermore, some DTTs may undergo preprocessing in the fog layer to optimize resource utilization. Moreover, tasks encountering blocking or preemption in the fog layer can also be secondarily unloaded to the cloud layer. In this framework, we devise a four-dimensional Markov chain (4DMC) to model and analyze this process. Through numerical experiments, we assess performance indicators under various parameters. Ultimately, our proposed strategy is compared with the unloading scheme that does not incorporate second unloading through both numerical analysis and simulation validation. The results indicate that our scheme notably enhances the throughput of DTTs, albeit at a marginal performance trade-off.
Journal
IF:
4.3
Papers:
545
Citations:
1.1W