PROJECT TITLE :

Dynamic Cloud Instance Acquisition via IaaS Cloud Brokerage

ABSTRACT:

Infrastructure-as-a-Service clouds supply diverse pricing options, including on-demand and reserved instances with varied discounts to draw in totally different cloud users. A sensible problem facing cloud users is how to attenuate their costs by selecting among different pricing options based on their own demands. In this paper, we tend to propose a new cloud brokerage service that reserves a massive pool of instances from cloud suppliers and serves users with worth discounts. The broker optimally exploits both pricing edges of long-term instance reservations and multiplexing gains. We tend to propose dynamic ways for the broker to create instance reservations with the target of minimizing its service cost. These methods leverage dynamic programming and approximation algorithms to rapidly handle massive volumes of demand. Our in depth simulations driven by large-scale Google cluster-usage traces have shown that vital worth discounts will be realized via the broker.