The cloud computing paradigm requires solutions supporting customers in the selection of services that satisfy their functional and non-functional requirements. These solu- tions must i) support the dynamic, multi-cloud nature of service provisioning, ii) manage scenarios where no total preference relation over service properties is available, and iii) prevent providers from misrepresenting or overstating their properties. In this paper we put forward the idea of modeling multi- cloud provisioning scenarios as procurement e-auctions (where the auctioneer is the customer and the bidders are service providers). We introduce a service selection process based on matching and ranking algorithms, and an e-auction mechanism that addresses the above requirements, encouraging trustwor- thy bids and therefore improving the truthfulness on the e- auction outcome. Finally we describe the implementation of a prototype used to evaluate the performance of our approach with respect to traditional query-based engines.
e-Auctions for Multi-Cloud Service Provisioning / M., Anisetti; C., Ardagna; Bonatti, PIERO ANDREA; E., Damiani; Faella, Marco; Galdi, Clemente; Sauro, Luigi. - (2014), pp. 35-42. (Intervento presentato al convegno IEEE International Conference on Services Computing (SCC 2014) tenutosi a Anchorage, AL, USA nel 2014-June--July) [10.1109/SCC.2014.14].
e-Auctions for Multi-Cloud Service Provisioning
BONATTI, PIERO ANDREA;FAELLA, MARCO;GALDI, CLEMENTE;SAURO, LUIGI
2014
Abstract
The cloud computing paradigm requires solutions supporting customers in the selection of services that satisfy their functional and non-functional requirements. These solu- tions must i) support the dynamic, multi-cloud nature of service provisioning, ii) manage scenarios where no total preference relation over service properties is available, and iii) prevent providers from misrepresenting or overstating their properties. In this paper we put forward the idea of modeling multi- cloud provisioning scenarios as procurement e-auctions (where the auctioneer is the customer and the bidders are service providers). We introduce a service selection process based on matching and ranking algorithms, and an e-auction mechanism that addresses the above requirements, encouraging trustwor- thy bids and therefore improving the truthfulness on the e- auction outcome. Finally we describe the implementation of a prototype used to evaluate the performance of our approach with respect to traditional query-based engines.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
