Request Prediction in Cloud with a Cyclic Window Learning Algorithm

TL;DR

This paper introduces a cyclic window learning algorithm using MLE and LLR to accurately predict cloud request loads, enhancing resource scaling efficiency and energy savings in cloud systems.

Contribution

It presents a novel cyclic window learning algorithm for request prediction in cloud systems, utilizing MLE and LLR, validated with real Google cluster data.

Findings

Achieved accurate predictions of task, CPU, and memory requests.

Reduced energy consumption through better resource scaling.

Validated with real-world cloud trace data.

Abstract

Automatic resource scaling is one advantage of Cloud systems. Cloud systems are able to scale the number of physical machines depending on user requests. Therefore, accurate request prediction brings a great improvement in Cloud systems' performance. If we can make accurate requests prediction, the appropriate number of physical machines that can accommodate predicted amount of requests can be activated and Cloud systems will save more energy by preventing excessive activation of physical machines. Also, Cloud systems can implement advanced load distribution with accurate requests prediction. We propose an algorithm that predicts a probability distribution parameters of requests for each time interval. Maximum Likelihood Estimation (MLE) and Local Linear Regression (LLR) are used to implement this algorithm. An evaluation of the proposed algorithm is performed with the Google cluster-trace data. The prediction is implemented about the number of task arrivals, CPU requests, and memory requests. Then the accuracy of prediction is measured with Mean Absolute Percentage Error (MAPE).