Economic Models for Managing Cloud Services by Sajib Mistry Athman Bouguettaya & Hai Dong

Author:Sajib Mistry, Athman Bouguettaya & Hai Dong
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

(4.3)

(b) The operation cost is determined by the server power consumption. Although there are other costs, such as maintenance costs, employee salaries and insurance costs, we choose the power consumption cost for simplicity. ARIMA is a popular model to predict the dynamic price of power [51]. In Eq. 4.4, the autoregressive (AR) part depends on the p lagged values of the time-series of aggregated AG n and α i is the coefficient constant of the lag operator L. The moving average (MA) part depends on the q lagged values of the previous prediction errors (𝜖 t ) and θ i is the coefficient constant. d represents the number of times that the difference operation is performed to obtain the stationary time-series. The values of (p, d, q) are determined by the Box-Jenkins method for the price history [20].

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.