Towards Virus Scanning as a Service in Mobile Cloud Computing: Energy-Efficient Dispatching Policy under N-Version Protection (Preprint-2017)
To aim of this project is to scan the files and reduce the energy consumption of service engines and the missed detection error cost, while satisfying the queue stability for VSaaS.
We propose an online algorithm for file-scan dispatching, which can outperform the alternative algorithm in terms of energy efficiency.
Using those heterogeneous Detection engines in parallel to scan the files, the N-version protection can reduce failures of traditional antivirus software detecting malware due to vulnerabilities of an individual virus scanner. However, those work only focus on the security aspect that is beneficial for the clients, without considering the queuing delay and energy consumption in the cloud.
The proposed online dispatching algorithm and an existing is K-least delay algorithm, respectively. It is shown that the proposed dispatching algorithm can achieve much less time average penalty than the K-least delay algorithm, while having smaller time average queue length.
We plot the time average energy consumption and the time average queue length for the proposed dispatching algorithm and the K-least delay algorithm, respectively.
It is shown that the proposed dispatching algorithm achieves a much better tradeoff between the time average energy consumption and the time average detection error cost than the K-least delay algorithm.