User demand prediction algorithm used in mobile device driver for cloud storage

Author

Mityakov, Alexander ; Petrov, Dmitry L. ; Kostenko, Maksim ; Butusov, Denis

Author_Institution

Fac. of Comput. Technol. & Inf., St. Petersburg Electrotech. Univ. ”LETI”, St. Petersburg, Russia

fYear

2011

fDate

23-25 Aug. 2011

Firstpage

788

Lastpage

793

Abstract

Access from a mobile device to data, stored in a cloud, leads to poor client quality experience. It is essential to reduce user download wait time for a requested file from a network to enhance client quality experience. The goal of the work is to develop a prediction algorithm, which should predict a user data request. The prediction algorithm is based on analysis of recent and current user activity of using a mobile device as well as a cloud. The basis of the algorithm is the Markov chain model. The efficiency criterion is minimization of user download wait time for a requested file from a cloud.

Keywords

Markov processes; cloud computing; device drivers; mobile computing; user modelling; Markov chain model; client quality experience; cloud storage; mobile device driver; user demand prediction algorithm; user download wait time; Algorithm design and analysis; Cloud computing; History; Kernel; Markov processes; Mobile handsets; Prediction algorithms; Markov models; cloud storage; user modeling;

fLanguage

English

Publisher

ieee

Conference_Titel

ITS Telecommunications (ITST), 2011 11th International Conference on

Conference_Location

St. Petersburg

Print_ISBN

978-1-61284-668-2

Type

conf

DOI

10.1109/ITST.2011.6060162

Filename

6060162