Premium
Optimization of power consumption and device availability based on point process modelling of the request sequence
Author(s) -
Durand JeanBaptiste,
Girard Stéphane,
Ciriza Victor,
Donini Laurent
Publication year - 2013
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/j.1467-9876.2012.01053.x
Subject(s) - computer science , sequence (biology) , consumption (sociology) , power consumption , point (geometry) , sleep mode , process (computing) , mode (computer interface) , power (physics) , hidden markov model , markov chain , constant (computer programming) , mathematical optimization , basis (linear algebra) , markov process , time point , real time computing , operations research , statistics , artificial intelligence , mathematics , machine learning , social science , genetics , physics , geometry , philosophy , aesthetics , quantum mechanics , sociology , biology , programming language , operating system
Summary. The paper addresses the optimal choice of the waiting period (or time out ) that a device should respect before entering sleep mode, to optimize a trade‐off between power consumption and device availability. The optimal time out is inferred by appropriate statistical modelling of the times between user requests. In our first approach, these times are assumed to be independent, and a constant optimal time out is inferred accordingly. In our second approach, some dependence is introduced through a hidden Markov chain, which also models specific activity states, like business hours or night periods. This model leads to a statistical framework for computing adaptive optimal time‐out values. Different strategies are assessed by using real data sets, on the basis of power consumption and device availability from the user's point of view.