A near-Zero run-time energy overhead within a computation outsourcing framework for energy management in mobile devices

In order to support run-time computation outsourcing to save energy, a certain amount of overhead has to be incurred to facilitate appropriate communication. As computation outsourcing is a methodology for saving energy on mobile devices, the amount of overhead incurred must be kept to a minimum. In this work, we support our compile-time methodology to facilitate the outsourcing of intensive computation with a run-time monitoring system that consumes an extremely low amount of energy (near-Zero). This near-Zero overhead resulted from analyzing the code at compile-time rather than run-time execution and profiling. The compile-time strategy utilized in our work analyzes the code at multiple levels of abstraction (High, Medium, and Low). The result of the analysis, which takes advantage of a real-time systems technique that calculates the maximum number of loop iterations, hence giving us a worst-case execution time for each loop within the benchmark application, allows for a fine-grain analysis of our benchmark. Resulting from analyzing the code, a client/server version of the applications is produced. As a result of producing this client/server version, certain runtime support has to take place on both the machine executing the client (the mobile device) as the machine executing the server. Our experimental results as performed on a Sharp Zaurus, utilizing Wi-Fi as a means of communication, showed tremendous energy saving while incurring a near-Zero run-time overhead. © 2008 IEEE