The advent of Big Data technology in agriculture has precipitated a paradigm shift in the manner by which farmers oversee the operations of their farms. The paper delineates the influence of Big Data on enhancing agricultural productivity, highlighting the global tendency to leverage data from satellites, sensors, and drones for yield prediction, disease prevention, and resource optimisation. This article examines the various Big Data architectures employed in the agricultural sector and their respective applications. In light of the specific challenges posed by limited data, a lack of digital infrastructure, and reliance on traditional agricultural knowledge, our objective is to design a Big Data architecture tailored to the agricultural sector in Burkina Faso. The proposed architecture is based on a centralised data lake, a processing layer using machine learning algorithms, and a straightforward visualisation interface, facilitating the utilisation of the information by farmers. This architecture is designed to be accessible to small-scale farmers, taking into account both the lack of digital infrastructure and the literacy levels of producers.