Development of an automated gridded crop growth simulation support system using Kubernetes

Authors: KIM, JUNHWAN - Rural Development Administration - Korea; PARK, JIN YU - Seoul University; HYUN, SHINWOO - Seoul University; YOO, BYUNG HYUN - Seoul University; Fleisher, David; KIM, KWANG SOO - Seoul University

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/1/2021
Publication Date: 5/7/2021
Citation: Kim, J., Park, J., Hyun, S., Yoo, B., Fleisher, D.H., Kim, K. 2021. Development of an automated gridded crop growth simulation support system using Kubernetes. Computers and Electronics in Agriculture. 186. https://doi.org/10.1016/j.compag.2021.106187.
DOI: https://doi.org/10.1016/j.compag.2021.106187

Interpretive Summary: Scientists are using computers and software programs, such as mathematical crop and soil models, to study options to increase food security. These studies require tremendous amounts of data, computational time, and knowledge regarding information technology that scientists frequently do not have expertise in. Methodology that can significantly reduce labor and resource costs associated with conducting these types of studies was developed and tested. The new system, called GROWLERS-Kube, uses inexpensive computers, networking hardware, and software to make it easier for scientists to more efficiently conduct these types of studies. The system uses a 'distributed-computing' method that essentially optimizes all available computing processing speed. A test case was conducted to evaluate optimal rice planting date windows across the Korean countryside. The simulation, which required thousands of model runs to complete, was used to show that up to 30% time-savings can be achieved with this approach. This methodology can simplify the amount of effort required to explore various 'what-if' type questions on a landscape scale regarding agricultural system productivity. This research can thus help scientists and food policy planners evaluate options for increasing food production while maintaining environmental stewardship needs involved in addressing United States food security concerns.

Technical Abstract: Spatial simulations of crop growth under climate change have been performed using a cluster computer of which operation would require expertise in distributed computing. The objective of this study was to develop an orchestration aid system for concurrent gridded simulations of crop growth, which would support the design of climate change adaptation options on crop production. The design goal of the orchestration aid system to help a user build a set of virtualized cluster computers using a simple input file which would require little expertise in distributed computing rather than manual configuration.. The orchestration aid system, which was referred to as GROWLERS-kube, was developed to launch multiple sets of gridded simulations using pods or containers managed by Kubernetes. As a case study, GROWLER-kube was executed using 16 Raspberry Pi 4 computers to perform 120 sets of the gridded simulations under diverse crop management options, including varying planting date and cultivar, for the period from 2001-2010. The wall time for the given sets of the gridded simulation differed by configuration of virtualized cluster computers, such as the number of pods used for server and client nodes, although the total number of physical nodes were identical. For example, the wall time difference between virtualized cluster computer sets was about 28.9% when 15 worker nodes were used. In particular, the acceleration of the gridded simulations was at maximum using a large number of the virtualized cluster computers with a small number of nodes. These configuration results suggest that GROWLERS-kube would facilitate the spatial assessment of climate change impact on crop production without considerable effort and expertise in distributed computing, which would aid a researcher to focus on the design of adaptation strategies.