Title :
QoS management by competitive agent-based negotiation in distributed cloud services
Author :
Ravindran, Kaliappa
Author_Institution :
City Coll. & Grad. Center, City Univ. of New York CUNY, New York, NY, USA
Abstract :
Agent-level negotiations between an application and a cloud service provider (SP) face two complexities: i) inaccuracy of computational models in capturing the behavior of a cloud-based system made up of diverse components and resources; and ii) disparity in the goals and priorities of an application and the SP in exercising a cloud-based system. The paper suggests trial actions by an agent to learn about the system behavior and the opponent agent goals, as part of effective negotiation strategy. The paper identifies transaction-style primitives for agents to invoke trial actions on the system QoS space: searching, alternations, and improvements. A larger goal of optimal QoS control can be achieved with our primitives. A case study of CDN realized on a cloud is also described.
Keywords :
cloud computing; learning (artificial intelligence); quality of service; software agents; QoS management; SP; agent-based negotiation; automated learning; cloud service provider; quality of service; transaction-style primitives; Bandwidth; Conferences; Delays; Quality of service; Resource management; Servers;
Conference_Titel :
Cloud Networking (CloudNet), 2014 IEEE 3rd International Conference on
Conference_Location :
Luxembourg
DOI :
10.1109/CloudNet.2014.6968991