A data-value-driven adaptation framework for energy efficiency for data intensive applications in clouds

The emerging of cloud computing and Big Data has been presenting to the world both grand opportunities and challenges. However, the increasing trend in energy consumption in clouds due to the fast growing quantity of data to be transmitted and processed has made cloud computing, together with Big Data phenomenon, becoming the dominant contributor in energy consumption, and consequently in CO₂ emission. In this paper, we propose an adaptation framework for data-intensive applications aiming to improve energy efficiency. The adaptation mechanism is driven by the data value extracted from datasets or data streams of the applications. Our main contribution lies in the proposal of treating large amount of data according to their value, i.e., their level of importance.