NeTS: Small: Fine-grained Measurement of Performance Metrics in the Internet of Things

North Carolina State University, Raleigh NC

Investigators

Abstract

The Internet of Things (IoT) enables an exchange of data that was not previously available, such as temporal and spatial distribution of soil moisture, power consumption of electrical appliances in smart buildings, and integrity of concrete structures. Each IoT application scenario requires unique guarantees on certain performance metrics (such as latency, loss, power consumption etc.) from the IoT infrastructure deployed in that scenario. A small deterioration in these performance metrics can cause violations of service level agreements and result in revenue and functional losses. In order to ensure that the IoT infrastructure delivers the guarantees on these performance metrics, the first step is to measure these metrics. In this project, the PI proposes to develop a framework to measure IoT performance metrics, which is required by the IoT network managers to reactively perform troubleshooting and to proactively locate and update potential future bottlenecks. The successful completion of this project will greatly benefit our society. First, the research conducted in this project will result in tools that will enable network managers to perform fine-grained monitoring of the health of their IoT networks. Second, the PI will make the traces collected during this project publicly available for research. Third, the proposed project will bridge several different communities such as mobile networking, measurements, wireless sensor networks, and systems; and foster interaction and communication among them. Fourth, the educational side of the project will integrate the research findings into the undergraduate and graduate curricula at North Carolina State University in addition to several other broader impacts and outreach activities. The performance metrics that this project will measure include both quality of service metrics such as latency, loss, and throughput and resource utilization metrics such as power consumption, storage utilization, radio on-time etc. The framework will measure these metrics passively, i.e., without interrupting the regular operations of the IoT network. In developing such a framework for measuring IoT performance metrics, the PI proposes a three phased approach: a recording phase, during which IoT devices record information about the performance metrics of interest; a transportation phase, during which IoT devices transport the recorded information to a server that has ample storage (such as a data center); and a querying phase, during which a network manager or an automated monitoring service queries the server for the values of the performance metrics of interest.