Optimal Service Pricing for a Cloud Cache Verena Kantere, Debabrata Dash, Gre ´gory Franc ¸ois, Sofia Kyriakopoulou, and Anastasia Ailamaki Abstract—Cloud applications that offer data management services are emerging. Such clouds support caching of data in order to provide quality query services. The users can query the cloud data, paying the price for the infrastructure they use. Cloud management necessitates an economy that manages the service of multiple users in an efficient, but also, resource-economic way that allows for cloud profit. Naturally, the maximization of cloud profit given some guarantees for user satisfaction presumes an appropriate price- demand model that enables optimal pricing of query services. The model should be plausible in that it reflects the correlation of cache structures involved in the queries. Optimal pricing is achieved based on a dynamic pricing scheme that adapts to time changes. This paper proposes a novel price-demand model designed for a cloud cache and a dynamic pricing scheme for queries executed in the cloud cache. The pricing solution employs a novel method that estimates the correlations of the cache services in an time-efficient manner. The experimental study shows the efficiency of the solution. Index Terms—Cloud data management, data services, cloud service pricing. Ç 1 INTRODUCTION T HE leading trend for service infrastructures in the IT domain is called cloud computing, a style of computing that allows users to access information services. Cloud providers trade their services on cloud resources for money. The quality of services that the users receive depends on the utilization of the resources. The operation cost of used resources is amortized through user payments. Cloud resources can be anything, from infrastructure (CPU, memory, bandwidth, network), to platforms and applica- tions deployed on the infrastructure. Cloud management necessitates an economy, and, there- fore, incorporation of economic concepts in the provision of cloud services. The goal of cloud economy is to optimize: 1) user satisfaction and 2) cloud profit. While the success of the cloud service depends on the optimization of both objectives, businesses typically prioritize profit. To maximize cloud profit we need a pricing scheme that guarantees user satisfaction while adapting to demand changes. Recently, cloud computing has found its way into the provision of web services [15], [18]. Information, as well as software is permanently stored in Internet servers and probably cached temporarily on the user side. Current businesses on cloud computing such as Amazon Web Services [14] and Microsoft Azure [19] have begun to offer data management services: the cloud enables the users to manage the data of back-end databases in a transparent manner. Applications that collect and query massive data, like those supported by CERN [17], need a caching service, which can be provided by the cloud [31]. The goal of such a cloud is to provide efficient querying on the back-end data at a low cost, while being economic- ally viable, and furthermore, profitable. Fig. 1 depicts the architecture of a cloud cache. Users pose queries to the cloud through a coordinator module, and are charged on- the-go in order to be served. The cloud caches data and builds data structures in order to accelerate query execu- tion. Service of queries is performed by executing them either in the cloud cache (if necessary data are already cached) or in a back-end database. Each cache structure (data or data structures) has an operating (i.e., a building and a maintenance) cost. A price over the operating cost for each structure can ensure profit for the cloud. In this work, we propose a novel scheme that achieves optimal pricing for the services of a cloud cache. 1.1 Setting the Price for Cloud Caching Services The cloud makes profit from selling its services at a price that is higher than the actual cost. Setting the right price for a service is a nontrivial problem, because when there is competition the demand for services grows inversely but not proportionally to the price. There are two major challenges when trying to define an optimal pricing scheme for the cloud caching service. The first is to define a simplified enough model of the price- demand dependency, to achieve a feasible pricing solution, but not oversimplified model that is not representative. For example, a static pricing scheme cannot be optimal if the demand for services has deterministic seasonal fluctuations. The second challenge is to define a pricing scheme that is adaptable to 1) modeling errors, 2) time-dependent model IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 23, NO. 9, SEPTEMBER 2011 1345 . V. Kantere is with the Department of Electrical Engineering and Information Technology, Faculty of Engineering and Technology, Cyprus University of Technology, Archbishop Kyprianos 31, Limassol Savings Co- operative Bank Building, PO Box 50329, Lemesos 3603, Cyprus. E-mail: verena.kantere@cut.ac.cy. . D. Dash is with the ArcSight, an HP Company, Avenue de Florissant 22, Renens 1020, Switzerland. E-mail: debabrata.dash@hp.com. . G. Franc ¸ois is with the  Ecole Polytechnique Fe´de´rale de Lausanne, EPFL STI IGM LA-CO, ME C2 401 (Baˆtiment ME), Station 9, Lausanne CH- 1015, Switzerland. E-mail: gregory.francois@epfl.ch. . S. Kyriakopoulou is with the  Ecole Polytechnique Fe´de´rale de Lausanne, Lausanne CH-1015, Switzerland. E-mail: sofia.kyriakopoulou@epfl.ch. . A. Ailamaki is with the  Ecole Polytechnique Fe´de´rale de Lausanne, EPFL IC IIF DIAS, BC 226 (Baˆtiment BC), Station 14, Lausanne CH-1015, Switzerland. E-mail: anastasia.ailamaki@epfl.ch. Manuscript received 15 Mar. 2010; revised 10 Nov. 2010; accepted 14 Jan. 2011; published online 25 Feb. 2011. Recommended for acceptance by D. Lomet. For information on obtaining reprints of this article, please send e-mail to: tkde@computer.org, and reference IEEECS Log Number TKDESI-2010-03-0151. Digital Object Identifier no. 10.1109/TKDE.2011.35. 1041-4347/11/$26.00 ß 2011 IEEE Published by the IEEE Computer Society