TY - JOUR
T1 - A computation offloading method over big data for IoT-enabled cloud-edge computing
AU - Xu, Xiaolong
AU - Liu, Qingxiang
AU - Luo, Yun
AU - Peng, Kai
AU - Zhang, Xuyun
AU - Meng, Shunmei
AU - Qi, Lianyong
PY - 2019/6
Y1 - 2019/6
N2 - The Internet of mobile things is a burgeoning technique that generates, stores and processes big real-time data to render rich services for mobile users. In order to mitigate conflicts between the resource limitation of mobile devices and users’ demands of decreasing processing latency as well as prolonging battery life, it spurs a popular wave of offloading mobile applications for execution to centralized and decentralized data centers, such as cloud and edge servers. Due to the complexity and difference of mobile big data, arbitrarily offloading the mobile applications poses a remarkable challenge to optimizing the execution time and the energy consumption for mobile devices, despite the improved performance of Internet of Things (IoT) in cloud-edge computing. To address this challenge, we propose a computation offloading method, named COM, for IoT-enabled cloud-edge computing. Specifically, a system model is investigated, including the execution time and energy consumption for mobile devices. Then dynamic schedules of data/control-constrained computing tasks are confirmed. In addition, NSGA-III (non-dominated sorting genetic algorithm III) is employed to address the multi-objective optimization problem of task offloading in cloud-edge computing. Finally, systematic experiments and comprehensive simulations are conducted to corroborate the efficiency of our proposed method.
AB - The Internet of mobile things is a burgeoning technique that generates, stores and processes big real-time data to render rich services for mobile users. In order to mitigate conflicts between the resource limitation of mobile devices and users’ demands of decreasing processing latency as well as prolonging battery life, it spurs a popular wave of offloading mobile applications for execution to centralized and decentralized data centers, such as cloud and edge servers. Due to the complexity and difference of mobile big data, arbitrarily offloading the mobile applications poses a remarkable challenge to optimizing the execution time and the energy consumption for mobile devices, despite the improved performance of Internet of Things (IoT) in cloud-edge computing. To address this challenge, we propose a computation offloading method, named COM, for IoT-enabled cloud-edge computing. Specifically, a system model is investigated, including the execution time and energy consumption for mobile devices. Then dynamic schedules of data/control-constrained computing tasks are confirmed. In addition, NSGA-III (non-dominated sorting genetic algorithm III) is employed to address the multi-objective optimization problem of task offloading in cloud-edge computing. Finally, systematic experiments and comprehensive simulations are conducted to corroborate the efficiency of our proposed method.
KW - Big data
KW - Cloud-edge computing
KW - Computation offloading
KW - Energy consumption
KW - IoT
UR - http://www.scopus.com/inward/record.url?scp=85060695141&partnerID=8YFLogxK
U2 - 10.1016/j.future.2018.12.055
DO - 10.1016/j.future.2018.12.055
M3 - Article
AN - SCOPUS:85060695141
SN - 0167-739X
VL - 95
SP - 522
EP - 533
JO - Future Generation Computer Systems
JF - Future Generation Computer Systems
ER -