Cloud and distributed architectures for data management in agriculture 4.0 : Review and future trends.

[Display omitted] • Cloud architectures used in Agriculture 4.0. • Distributed Architectures and Cloud Computing complements. • Strategies of association between Edge, Fog, Cloud. • New architectural and computing trends. The Agriculture 4.0, also called Smart Agriculture or Smart Farming, is at the origin of the production of a huge amount of data that must be collected, stored, and processed in a very short time. Processing this massive quantity of data needs to use specific infrastructure that use adapted IoT architectures. Our review offers a comparative panorama of Central Cloud, Distributed Cloud Architectures, Collaborative Computing Strategies, and new trends used in the context of Agriculture 4.0. In this review, we try to answer 4 research questions: (1) Which storage and processing architectures are best suited to Agriculture 4.0 applications and respond to its peculiarities? (2) Can generic architectures meet the needs of Agriculture 4.0 application cases? (3) What are the horizontal development possibilities that allow the transition from research to industrialization? (4) What are the vertical valuations possibilities to move from algorithms trained in the cloud to embedded or stand-alone products? For this, we compare architectures with 8 criteria (User Proximity, Latency & Jitter, Network stability, high throughput, Reliability, Scalability, Cost Effectiveness, Maintainability), and analyze the advantages and disadvantages of each of them. [ABSTRACT FROM AUTHOR]