Uma metodologia para definição de modelos de cobrança em ambientes de nuvens computacionais

Abstract

Nowadays cloud computing is considered a new concept for the provision of computational infrastructure. In this environment, clients can access and use software, platform or infrastructure as a service to reduce their costs, compared to the traditional values of datacenter model. To make this feasible, customers are charged by the use of computer resources by means of a charging model. These models determine how the use of any product or service will be paid by a client to a service provider. However, the definition of a charging model involves the execution of a series of processes so that is possible to identify the best parameters of the model for each provider. Therefore, it is necessary to clearly define the type of service that the provider adopts and the products you want to offer to the market. Moreover, it is necessary to verify the sales strategies adopted by the provider and the cost of operation. These difficulties were also found by PUCRS High Performance Laboratory (LAD-PUCRS). To facilitate this process, this paper presents a methodology to define and optimize charging models for providers of cloud computing. The method was applied in the LAD, which resulted in several improvements to the charging models used by the laboratory. Furthermore, the application of the methodology in this case study was evaluated and contributed to its improvement.