Proactive machine learning-based solution for advanced manageability of multi-persona mobile computing. (December 2019)

• Record Type: Journal Article
• Authors: Tout, Hanine; Kara, Nadjia; Talhi, Chamseddine; Mourad, Azzam
• Is Part Of: Computers & electrical engineering. Volume 80(2019)
• Publication Date: 2019-12
• Subjects: Mobile device -- Multi-persona mobile computing -- Mobile cloud computing -- Offloading -- Optimization -- Dynamic programming -- Machine learning -- Artificial intelligence; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2019.106497
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Latest mobile virtualization techniques have opened the door for multi-persona mobility to overcome security and privacy concerns of bring-your-own devices practice. Multi-persona allows a physical device to co-host multiple virtual phones with impenetrable walls among them. However, physical resources should be always enough to support virtual instances and applications needs without performance degradation or system crash. Though computation offloading can augment devices resources, yet some applications are not offloadable. Additionally, idle applications and virtual environments impose high overhead on the device. Through machine learning, this work predicts future context and resource needs of currently running virtual environments and potential future active ones. It provides advanced manageability strategies, formulated in an optimization model, which appropriately turn off applications and switch off virtual environments to release device resources when needed. A dynamic programming algorithm is advocated to find the adequate strategies. Extensive experiments conducted demonstrate the efficiency of our proposition.