Automatic negotiation of Service Level Agreements (SLAs) is a promising way to stipulate contracts in the Cloud market, where the high dynamicity of customers' requirements and providers' resources availability make it very difficult to statically define Quality of Service (QoS) level and pricing. To achieve high satisfaction levels for both parties, the negotiation decisions about stipulation conditions (or rejection) of contracts should be guided both by an overall strategic business policy and by dynamic information. In this paper, we propose to exploit capacity planning to support bilateral negotiation processes with the aim of optimizing the overall utility for service providers, by avoiding contracts that could incur in SLAs violations, keeping, at the same time, competitive service prices. In particular, the proposed technique exploits a heuristic algorithm to automatically evaluate a non-additive utility function and the acceptable region, taking into account QoS, resources availability, costs and penalties. The technique is compared with static approaches by using some simulations.
Exploiting Capacity Planning of Cloud Providers to Limit SLA Violations
Zimeo E.
2013-01-01
Abstract
Automatic negotiation of Service Level Agreements (SLAs) is a promising way to stipulate contracts in the Cloud market, where the high dynamicity of customers' requirements and providers' resources availability make it very difficult to statically define Quality of Service (QoS) level and pricing. To achieve high satisfaction levels for both parties, the negotiation decisions about stipulation conditions (or rejection) of contracts should be guided both by an overall strategic business policy and by dynamic information. In this paper, we propose to exploit capacity planning to support bilateral negotiation processes with the aim of optimizing the overall utility for service providers, by avoiding contracts that could incur in SLAs violations, keeping, at the same time, competitive service prices. In particular, the proposed technique exploits a heuristic algorithm to automatically evaluate a non-additive utility function and the acceptable region, taking into account QoS, resources availability, costs and penalties. The technique is compared with static approaches by using some simulations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.