The ARP team is preparing to restart after the project was put on pause in December 2022. This article outlines the steps that we are taking in order to restart this project.
Background
The Africa Rainfall Project (ARP) aims to simulate rainstorms in sub-Saharan Africa to improve regional weather forecasts. The goal is to run a weather simulation at a high resolution (1 km) for the whole region for a period of one year. Providing accurate weather predictions is crucial for self-sufficiency of the local farming community. By comparing results obtained through computing rainfall data from various sources using the World Community Grid, scientists can create increasingly more accurate forecasts, improve future simulations and in turn weather predictions. To read more about this project, see our March 2022 research update.
In December 2022, the project was put on pause due to capacity limitations of their storage system. To read more about the details of this storage limitation, please refer to ARP update from March 2023. As we shared in a July 2023 update, part of this challenge was that the ARP team’s storage support from SURF, the cooperative association of Dutch educational and research institutions, came to an end. Fortunately, Amazon Web Services (AWS) agreed to host the data, and thus began the process of moving the data. Since the move is complete, the ARP team has been taking steps to make the data more broadly accessible through the AWS platform, but this work is still in the early stages. Currently, the focus is on finalizing the preparation of new work units and generating the data needed to restart ARP.
ARP Restart Update
We have started to work on resuming sending out ARP1 work units to volunteers. Our research partners at TU Delft have already sent us most of the required scripts and documentation, which we have reviewed and implemented locally. Currently, we are awaiting a determination on whether we will receive the rest of the required data and application code directly, or whether we will need to reproduce these components from documentation provided by the TU Delft team. We have assessed that both approaches are feasible, and wanted to let volunteers know that we are on a defined path to restarting ARP1 in the coming weeks.
The timeline of a possible restart of this project depends on the information from the ARP team. If the code and data required is not fully available to us, we will need to build out the missing components of the pipeline for generating new ARP1 work and confirm that we are able to reproduce the same inputs as we previously received from TU Delft servers to validate the new pipeline. Regardless, there will be no change and no action required by volunteers to facilitate the process and when ARP1 restarts work unit distribution will be the same as before.
Figure 1. An example of a unit. Each unit comprises three nested domains (1-way) with respectively 9 km, 3 km, and 1 km resolution.
Thanks to the ARP team for their continued partnership with the WCG. If you have any questions or comments, please leave them in this thread for us to answer.