Energy Saving and Scavenging in Stand-Alone and Large Scale Distributed Systems

Abstract: This thesis focuses on energy management techniques for distributed systems such as handheld mobile devices, sensor nodes, and data center servers. My work specifically focuses on the heterogeneity in system hardware components and workloads. It includes energy management solutions for unregulated or battery-less embedded systems; and data center servers with heterogeneous workloads, machines, and processor wear-states. This thesis describes four major contributions:
(1) This thesis describes a battery test and energy delivery system design process to maintain battery life in embedded systems without voltage regulators.
(2) In battery-less sensor nodes, this thesis demonstrated a routing protocol to maintain reliable transmission through the sensor network.
(3) This thesis has characterized typical workloads and developed two models to capture the heterogeneity of data center tasks and machines. These models allow users to predict task finish time on individual machines. I then integrated these two models into a task scheduler based on the Hadoop framework for MapReduce tasks, and used this scheduler for server machine energy minimization using task concentration.
(4) In addition to saving server energy consumption, this thesis describe a method of reducing data center cooling energy by maintaining optimal server processor temperature setpoints through a task assignment algorithm. This algorithm considers the reliability impact of processor wear-states. I recorded processor wear-states through automatic timing slack tests on a cluster of machines with varying core temperatures, voltages, and frequencies. I used these optimal temperature setpoints in task scheduling algorithm that saves both server and cooling energy.