Dynamic Cost-Aware Routing of Web Requests

摘要: Work within next generation networks considers additional network convergence possibilities and the integration of new services to web. This trend responds ongoing growth end-user demand for that can be delivered anytime, anywhere, on any web-capable device, traffic generated by applications, e.g., Internet Things. To support massive enormous user base number devices with reliability high quality, web run from redundant servers. As servers need regularly deployed at different geographical locations, energy costs have become a source major concern operators. We propose cost aware method routing requests across replicated distributed exploit spatial temporal variations both electricity prices server network. The relies learning automaton makes per-request decisions, which computed much faster than regular global optimization methods. Using simulation testbed measurements, we show reductions are achievable minimal impact performance compared standard algorithms.