PROJECT TITLE:

Online Resource Scheduling under Concave Pricing for Cloud Computing - 2016

ABSTRACT:

With the booming cloud computing business, computational resources are readily and elastically available to the purchasers. In order to attract customers with numerous demands, most Infrastructure-as-a-service (IaaS) cloud service suppliers offer several pricing ways such as pay as you go, pay less per unit when you employ a lot of (so referred to as volume discount), and pay even less when you reserve. The numerous pricing schemes among completely different IaaS service providers or maybe in the same provider kind a complex economic landscape that nurtures the market of cloud brokers. By strategically scheduling multiple customers' resource requests, a cloud broker will totally exploit the discounts offered by cloud service suppliers. In this paper, we target how a broker can facilitate a group of consumers to totally utilize the quantity discount pricing strategy offered by cloud service suppliers through cost-economical online resource scheduling. We have a tendency to gift a randomized online stack-centric scheduling algorithm (ROSA) and theoretically prove the lower certain of its competitive ratio. Three special cases of the offline concave cost scheduling downside and the corresponding optimal algorithms are introduced. Our simulation shows that ROSA achieves a competitive ratio shut to the theoretical lower bound underneath the special cases. Trace-driven simulation using Google cluster data demonstrates that ROSA is superior to the traditional online scheduling algorithms in terms of value saving.