Negotiation-Based Flexible SLA Establishment with SLA-driven Resource Allocation in Cloud Computing

Abstract

As various consumers tend to use personalized Cloud services, Service Level Agreements (SLAs) emerge as a key aspect in Cloud and Utility computing. The objectives of this doctoral research are 1) to support a flexible establishment of SLAs that enhances the utility of SLAs for both providers and consumers, and 2) to manage Cloud resources to prevent SLA violations. Because consumers and providers may be independent bodies, some mechanisms are necessary to resolve different preferences when they establish a SLA. Thus, we designed a Cloud SLA negotiation mechanism for interactive and flexible SLA establishment. The novelty of this SLA negotiation mechanism is that it can support advanced multi-issue negotiation that includes time slot and price negotiations. In addition, to prevent SLA violations, we provided a SLA-driven resource allocation scheme that selects a proper data center among globally distributed centers operated by a provider. Empirical results showed that the proposed SLA negotiation mechanism supports faster agreements and achieves higher utilities. Also, the proposed SLA-driven resource allocation scheme performs better in terms of SLA violations and the provider's profits.