LSTM-based Service Migration for Pervasive Cloud Computing

Service migration in pervasive cloud computing is important for leveraging cloud resources to execute mobile applications effectively and efficiently. This paper proposes a LSTM (long and short-term memory model) based service migration approach for pervasive cloud computing, i.e., LSTM4PCC, which supports an accurate prediction of cloud resources. LSTM4PCC makes a prediction for cloud resource availability with a LSTM network and establishes a service migration mechanism in order to optimize service executions. We evaluate LSTM4PCC and compare it with the ARIMA (AutoRegressive Integrated Moving Average) approach in terms of prediction accuracy. The results show that LSTM4PCC performs better than ARIMA.

Authors:
Jing Haifeng, Zhang Yafei, Zhou Jiehan, Zhang Weishan, Liu Xin, Min Guizhi, Zhang Zhanmin

Publication type:
A4 Article in conference proceedings

Place of publication:
Proceedings IEEE 2018 International Congress on Cybermatics – 2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)

Keywords:
LSTM, Machine learning, Pervasive Cloud Computing, Service Migration

Published:

Full citation:
H. Jing et al., “LSTM-Based Service Migration for Pervasive Cloud Computing,” 2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), Halifax, NS, Canada, 2018, pp. 1835-1840. doi: 10.1109/Cybermatics_2018.2018.00305

DOI:
https://doi.org/10.1109/Cybermatics_2018.2018.00305

Read the publication here:
http://urn.fi/urn:nbn:fi-fe202003238887