Parallel Processing Implementation on Weather Monitoring System for Agriculture

Abstract

Weather monitoring and forecasting are very important in agricultural sectors. There are several data need to be collected in real-time to support weather monitoring and forecasting systems, such as temperature, humidity, air pressure, wind speed, wind direction, and rainfall. The purpose of this research to develop a real-time weather monitoring system using a parallel computation approach and analyze the computational performance (i.e., speed up and efficiency) using the ARIMA model. The developed system wireless has been implemented on sensor networks (WSN) platform using Arduino and Raspberry Pi devices and web-based platform for weather visualization and monitoring. The experimental data used in our research work is a set of weather data acquired and collected from January until March 2017 in Bogor area. The result of this research is that the speed up of the using eight processors computation three times faster than using a single processor, with the efficiency of 50%.