期刊
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
卷 216, 期 -, 页码 -出版社
ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
DOI: 10.1016/j.jnca.2023.103656
关键词
Edge computing; Task offloading; Cloud-edge-device collaboration; Dependent task; Service caching
Edge computing provides abundant computing and storage resources to meet the growing requirements of delay-sensitive mobile applications. This paper proposes an efficient dependent task offloading mechanism that can optimize the overall task completion time in scenarios where edge servers have limited service caches and computing power.
Edge computing provides abundant computing and storage resources at the edge of a network, aiming to meet the growing requirements of delay-sensitive mobile applications. To fully utilize the advantages of edge computing, it is essential to design an appropriate task offloading strategy to efficiently use edge resources. However, the existing studies of task offloading often ignored the dependencies between tasks and the limited service capabilities of edge servers, resulting in long completion times or even infeasible offloading decisions. Therefore, an efficient dependent task offloading mechanism that can optimize the overall task completion time is proposed in this paper. This mechanism is suitable in scenarios where edge servers have limited service caches and computing power. First, a collaborative offloading model that includes one cloud server, edge servers and local devices is designed. This model uses the cloud server as the offloading node to balance the workload among edge servers. Second, the research problem is formulated as a binary optimization function with the goal of minimizing the application completion time, and a dependent task offloading algorithm for a cloud-edge-device collaborative model is proposed. Finally, extensive experiments are performed to verify the effectiveness of the proposed model and algorithm. The experimental results show that compared with other algorithms, the proposed algorithm can reduce the application completion time by approximately 6% to 30%. In addition, the proposed dependent task offloading mechanism displays better adaptability and scalability for large-scale tasks.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据