Temporal Databases and Time-Series Analytics for IoT Applications

Abstract

The rapid proliferation of Internet of Things (IoT) devices has introduced new challenges in managing large volumes of time-stamped data. Traditional relational databases often fall short when handling the velocity, volume, and variety of time-series data generated by sensors and embedded systems. This paper investigates the role of temporal databases in storing, indexing, and querying time-series data for IoT applications. We evaluate early-generation time-series databases such as OpenTSDB, InfluxDB (pre-2016 releases), and RRDtool, focusing on their performance in terms of ingestion rates, data compression, downsampling, and support for temporal joins. Novel approaches to time-indexing, including time-partitioned tables and sliding window models, are discussed. Applications in smart cities, industrial monitoring, and environmental systems illustrate the practical significance of these databases in enabling real-time analytics, anomaly detection, and automated decision-making. The study concludes by outlining key design considerations for future IoT-centric temporal data management systems.