A Review and Taxonomy on Data Driven Regression Models for Estimating Future Cloud Workloads

A Review and Taxonomy on Data Driven Regression Models for Estimating Future Cloud Workloads

Dinesh Tulasidas Bharote1, Prof. Pallavi Bagde2

Department of CSE,SDBCE, Indore1,2

Abstract—Cloud Computing has long become asought after fields in computer science. Several applications which need high computational complexity but cannot be performed on conventional hardware prefer to leverage cloud based platforms. Hence with increasing traffic and load on cloud servers or cloud based platforms, there seems to be a natural need for cloud workload prediction so as to estimate and manage cloud based resources. Since cloud data is large and complex at the same time, hence it is necessary to use artificial intelligence based techniques for the estimation of cloud workload so as to improve upon the accuracy of conventional techniques.  This paper presents a review on the contemporary techniques for cloud workload prediction. The performance evaluation parameters have also been discussed.. It is expected that the paper would present with a headway for further research in cloud workload prediction.

Keywords—Cloud Workload Prediction, Aftificial Intelligence, Machine Learning, Artificial Neural Network (ANN), Mean Absolute Percentage error, Mean Square Error.