The Analysis of Time Series Forecasting on Resource Provision of Cloud-based Game Servers

Time Series Forecasting With Volatility Activation Function

“…Mouine [14] has analyzed time series forecasting models by using ARIMA, Prophet and LSTM in order to forecast the number of virtual machines which need cloud source monitoring data in her study conducted on description of a forecast model that is effective on game session workloads for which a certain time period is given. As a result of her study, she found LSTM is the most accurate model in forecasting the demand of virtual machines in terms of RMSE and MAE.…”