AQA2: An analytical QoS-assessment approach for service function chaining in cloud environment

Abstract

The increasing use of high-quality Cloud-based services makes Cloud providers provide their customers with a higher quality for every service. In which, Service Level Agreements (SLAs) between users and Cloud providers become a key factor which is carefully considered before running any service or application. Thus, Cloud providers always deal with the challenge of quality of service (QoS), especially in relation to how a service provider can prior recommend their users good services and ensure a negotiated QoS level for that services in their cloud infrastructure. This paper proposes a novel analytical method, named AQA2, to prior assess QoS of service function chains (SFC) in cloud environment before SFC deployment in order to achieve the most appropriate SLA between cloud providers and customers. We first consider two main types of service chain to analyze and represent the system model. By applying queuing theory, we transform the system model into queuing systems, and we then figure out mathematical models to represent QoS parameter measurements of a service chain. Finally, we propose the AQA2 scheme, conduct experiments and evaluate on OpenStack cloud platform. To our best knowledge, the AQA2 is the first analytical approach applied to prior assess QoS of service function chains in order to optimize SLAs negotiation in cloud environment.