Key technology research on decentralized Internet-based smart agriculture platform

Abstract

Agriculture is the fundamental industry for national development and the foothold of the national economy, and the Internet era has provided a new direction for the development of smart agriculture. This paper discusses the feasibility of the computing framework based on the Hadoop distributed computing framework under the decentralized Internet, from the HDFS file transfer system and MapReduce distributed computing method and verifies the feasibility of the distributed computing framework in the smart agriculture platform by conducting an example analysis for the sample data of tea growing environment in the tea plantation in the smart agriculture platform. From the data, the data transfer rate of the decentralized Internet-based Hadoop distributed computing framework in this paper is 43.95 Mb/s, and the average data upload time is 81.35 s. The optimal environment for tea growth in the tea garden is 74.21% and 15.32°C for soil humidity and temperature, 86.78% and 23.62°C for ambient humidity and temperature, respectively, and the average value of light was 6219.46lux, and the mean value of CO₂ concentration was 489.27mg/L, etc. This shows that the decentralized Internet-based Hadoop distributed computing framework can provide faster data transmission and time-saving data upload time for the smart agriculture platform and can also comprehensively analyze the growing environment of crops and then optimize the growing environment of crops in a targeted manner to improve crop yields and increase economic benefits. Of course, the application of decentralized Internet in smart agriculture platforms also provides a new research direction for developing Internet technology.