TY - GEN
T1 - Mobility- And load-adaptive controller placement and assignment in LEO satellite networks
AU - Chen, Long
AU - Tang, Feilong
AU - Li, Xu
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/5/10
Y1 - 2021/5/10
N2 - Software-defined networking (SDN) based LEO satellite networks can make full use of satellite resources through flexible function configuration and efficient resource management of controllers. Consequently, controllers have to be carefully deployed based on dynamical topology and time-varying workload. However, existing work on controller placement and assignment is not applicable to LEO satellite networks with highly dynamic topology and randomly fluctuating load. In this paper, we first formulate the adaptive controller placement and assignment (ACPA) problem and prove its NP-hardness. Then, we propose the control relation graph (CRG) to quantitatively capture the control overhead in LEO satellite networks. Next, we propose the CRG-based controller placement and assignment (CCPA) algorithm with a bounded approximation ratio. Finally, using the predicted topology and estimated traffic load, a lookahead-based improvement algorithm is designed to further decrease the overall management costs. Extensive emulation results demonstrate that the CCPA algorithm outperforms related schemes in terms of response time and load balancing.
AB - Software-defined networking (SDN) based LEO satellite networks can make full use of satellite resources through flexible function configuration and efficient resource management of controllers. Consequently, controllers have to be carefully deployed based on dynamical topology and time-varying workload. However, existing work on controller placement and assignment is not applicable to LEO satellite networks with highly dynamic topology and randomly fluctuating load. In this paper, we first formulate the adaptive controller placement and assignment (ACPA) problem and prove its NP-hardness. Then, we propose the control relation graph (CRG) to quantitatively capture the control overhead in LEO satellite networks. Next, we propose the CRG-based controller placement and assignment (CCPA) algorithm with a bounded approximation ratio. Finally, using the predicted topology and estimated traffic load, a lookahead-based improvement algorithm is designed to further decrease the overall management costs. Extensive emulation results demonstrate that the CCPA algorithm outperforms related schemes in terms of response time and load balancing.
UR - https://www.scopus.com/pages/publications/85111911634
U2 - 10.1109/INFOCOM42981.2021.9488806
DO - 10.1109/INFOCOM42981.2021.9488806
M3 - 会议稿件
AN - SCOPUS:85111911634
T3 - Proceedings - IEEE INFOCOM
BT - INFOCOM 2021 - IEEE Conference on Computer Communications
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 40th IEEE Conference on Computer Communications, INFOCOM 2021
Y2 - 10 May 2021 through 13 May 2021
ER -