Using time series databases in edge computing architecture

DOI: 10.31673/2412-9070.2023.013841

Abstract

A vast amount of data is constantly being generated from a large number of sensors and monitoring systems. With the rapid development of IoT technologies in recent years, there are increasingly more IoT systems that generate a large number of queries to support the operation of smart homes, smart cities, smart farms, smart agriculture, smart factories, smart manufacturing, smart industry, autopilots, and many others. In order to reduce the number of queries to the main cloud, this article describes load balancing solutions for edge computing nodes, which will also reduce the delay in sending and processing. The article also describes the features of time series databases for their use in the IoT Edge Cloud infrastructure. IoT-based systems typically require a large amount of data. A large volume of data is collected from numerous sensors, and the streams of this data are constant. Therefore, storing and analyzing data are the main challenges that need to be addressed. In recent years, big data analytics have been widely used to analyze data collected from IoT systems to improve fault detection and diagnosis. Due to the intensive nature of processing big data, analysis is usually done in the cloud. However, sending data to the cloud can result in significant communication delays. Additionally, many fault detection and diagnosis tasks need to be performed in real-time. In response to this balancing of time and resource constraints, a new computing infrastructure has been developed that combines IoT, Edge, and Cloud. Pre-computations necessary for timely decision-making can be done on nearby Edge gateways and local networks. Integrated analytics can be left in the cloud, while more complex and/or accurate analysis results can be sent back to Edge devices and IoT devices to improve decision-making processes or support changing trends.

Keywords: IoT; Edge Cloud; TSDB; edge computing; time series databases; decentralization of calculations.

References
1. Кулкарні А., Буз Р., Тот А. What Is Time-Series Data? [Електронний ресурс]. URL: https://www.timescale.com/blog/time-seriesdata/.
2. Перрі Л. IoT and Edge Computing for Architects. Birmingham: Packt Publishing Ltd., 2020. 632 с.