Analysis and Prediction of Amazon EC2 Spot Instance Prices

摘要

Amazon Elastic Compute Cloud (EC2) is a web service that provides secure, re-sizable compute capacity in the cloud. It is designed to make web-scale cloud computing for developers. There are three different ways of pricing for Amazon EC2 instances: On-demand, Reserved instances, and Spot instances. Spot instances provide users with access to unused Amazon EC2 capacity at high discounts relative to On-demand and reserved prices. The spot prices fluctuate based on the demand and supply of available unused capacity of EC2. When users request spot instances, they specify the maximum spot price they are willing to pay. Spot instances are launched when the spot price is lower than the price specified by user. It will continue to run until either user chooses to terminate it or the spot price exceeds the maximum price specified by users. The major challenge for researchers in this area is to predict the price for a particular instant of time which assures uninterrupted execution. In this paper, price of spot instances is analyzed and regression algorithm is used to predict useful information for cloud clients and cloud sellers who want to start selling spot instances. Efforts have been made to design a methodology based on regression algorithm to provide useful data to the customers of spot instances.