ADSLANF: A negotiation framework for cloud management systems using a bulk negotiation behavioral learning approach

One of the major challenges in cloud computing is the development of a service-level agreement (SLA) negotiation framework using an intelligent third-party broker negotiation strategy. Current frameworks exploit various negotiation strategies using game theoretic, heuristic, and argumentation-based approaches for obtaining optimal negotiation with a better success rate (negotiation commitment). However, these approaches fail to optimize the negotiation round (NR), total negotiation time (TNT), and communication overhead (CO) involved in the negotiation strategy. To overcome these problems, certain researchers have exploited trade-off, concession, and behavioral learning strategies with varying degrees of sacrifices (reductions) in their concerned proposal generation. Such sacrifices can prevent negotiation break-off and optimize the negotiation strategy to an extent with fewer NRs, less TNT, and less CO. It maximizes the utility value and the success rate. To further optimize the negotiation strategy and prevent negotiation break-off, a bulk negotiation behavioral learning (BNBL) approach is proposed. This approach uses the reinforcement learning negotiation strategy to provide varying degrees of sacrifice for obtaining an optimal result. Hence, the proposed automated dynamic SLA negotiation framework (ADSLANF) using the BNBL approach will reduce the NRs, TNT, and CO. It also significantly maximizes the utility value and success rate (SLA commitment) among negotiation parties such as service consumers and service providers.