Research On Online Algorithms For Trading IaaS Cloud Services

Nowadays,there are more and more users turning to public cloud platforms for acquiring Infrastructure-as-a-Service(IaaS)offerings.Simultaneously,cost management becomes a vital concern to cloud users,for the reason of an increasing IaaS costs' proportion of enterprises' IT investments.Therefore,it is important to research on purchasing strategies and selling strategies of IaaS cloud services.Its purpose is to develop a more economical purchasing strategy and a more reasonable selling strategy according to their own usage.To address these issues,we propose some online algorithms for trading IaaS cloud services in accordance with the pricing policies published by most public cloud providers in this paper:1.Regarding the purchasing strategy,we put forward a new online reserving algorithm for instances in accordance with the pricing policies used in most public IaaS offerings.Through theoretical analysis,we prove that the cost of the proposed algorithm A? in this paper is not greater than 2-1/? times of the optimal offline algorithm,where ?>1 is a critical point in the online reservation algorithm proposed in this paper.Via extensive experimental simulations,we demonstrated that the online algorithm is much more cost effective for cloud users.2.In terms of the selling strategy,combined with the selling rules in Amazon EC2 marketplace,we design some deterministic online selling algorithms and a random online selling algorithm.We prove theoretically that our deterministic algorithm can achieve a ratio of 2-a-a/4 in managing instance purchasing cost,where a is the entitled discount due to reservation and a is the selling discount specified by the user who sells its reservations.Besides,our random online algorithm can guarantee a bounded competitive ratio of 2T/?,whose value is specific to the type of reserved instances.Via extensive experiments,we verify the proposed online selling algorithms can save more expenses.In this paper,we propose online purchasing algorithms and selling algorithms using competitive analysis.Through experimental evaluations based on real-world applications traces,we show that our algorithms can achieve guaranteed competitive ratios and cost savings in Amazon EC2 marketplace.