A Software Framework for Efficient Execution of Weather Research and Forcasting Models

The WRF model is simulation system and heavily dependent on computing, I/O, and network and running it on the fastest computers is a major challenge. Optimal and timely implementation is the goal of many meteorological organizations. The management and implementation of high-performance computing applications is mainly based on cluster and cloud computing infrastructures due to the highly coordinated, consistent, and parallel processing of large data in them. These infrastructures typically use virtualization to manage access to resources. The purpose of this research is to use Containers as a lightweight virtualization to implement the WRF model, which provides a very small overhead for resources in the cloud infrastructure. Model implementation scheduling enables organizations to perform forecasts periodically, sometimes several times a day, and increase model members to obtain better output. To calculate the runtime of the model at different geographic scales on a software and hardware platform a mathematical formula is proposed.