RESOURCE USAGE PREDICTION BASED ON ARIMA-ARCH MODEL FOR VIRTUALIZED SERVER SYSTEM

Abstract

Performance degradation is unavoidable in server systems and this is because of factors such
as shrinkage of system resources, data corruption, and numerical error accumulation. The resource shrinkage
leads to the system failure due to the error propagation. Thus the resource prediction is useful to the
administrator of the system so that an accidental outage can be avoided. It has been observed in past that
most of the failures occur due to the exhaustion of free physical memory, so here free physical memory of a
server consolidation setup is observed. It is also found that most of the studies in this direction were using the
measurement-based approach with time series models for prediction. This paper reviews the effectiveness of
such models and it examines whether volatility is present in the data or not. It checks whether Gauss-Markov
assumptions about homoscedasticity holds good for the ordinary least square estimators of such models or
not. This paper applies a combination of AutoRegressive Integrated Moving Average - AutoRegressive
Conditional Heteroskedastic (ARIMA-ARCH) model to predict resource usage. Experimental results
demonstrate that the goodness of fit of the ARIMA-ARCH Model has improved when compared to the linear
ARIMA model.