.png){kind=icon}
返回
Application-aware computation offloading in edge computing networks
.png){kind=icon}
PRE
AI
DOI：10.1016/j.future.2023.04.009.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Edge computing involves distributive computation resources deployed at the network edge, unlike cloud computing, which has central computation resources in data centers. Edge computing is a complement of cloud computing because edge computing effectively reduces the computing response delay by processing computation tasks and data near terminals. Considering the dramatic increase of terminals connected to networks and data generated by terminals, computation tasks from different applications may require significantly different services with different computation requirements, storage requirements, and response delay requirements. Application-aware computation offloading and resource allocation in edge computation can provide efficient and guaranteed computation services to terminals. In this paper, an application-aware computation offloading and resource allocation problem is investigated in edge computing networks, where computation tasks from different applications have different requirements. A non-convex optimization problem of energy consumption minimization is formulated, where terminals, edge nodes, and a cloud are considered. We convert the original non-convex optimization problem into a lower-bound convex problem and an upper-bound convex problem. Then, an algorithm based on the branch-and-bound method is proposed to force the lower-and upper-bound solutions to approach the optimal solution. Finally, the performance of the algorithm is analyzed where the gap to the optimal solution is provided. Numerical results show that the proposed algorithm can provide guaranteed services for tasks of different application types, with improvements over application-unaware algorithms.(c) 2023 Elsevier B.V. All rights reserved.
Keyword:
Edge computing
Computation offloading
Application-aware
Non-convex
Branch-and-bound
Stochastic gain
期刊
F
IF:
6.1
论文数:
6.9K
被引数:
2.4W
机构
引用论文
.png){kind=icon}
Constrained Multiobjective Optimization for IoT-Enabled Computation Offloading in Collaborative Edge and Cloud Computing协同边缘和云计算中基于IoT的计算卸载的约束多目标优化
PREAI
分享收藏分享收藏The Extended Cloud: Review and Analysis of Mobile Edge Computing and Fog From a Security and Resilience Perspective扩展云: 从安全和弹性角度回顾和分析移动边缘计算和雾
OAAI
OAAI分享收藏
查看更多内容