Managing data in large-scale climate modelling project

Posted on

Climate modelling experiments on modern-day supercomputers generate a huge volume of data. The team of scientists (climate, computational and data scientists) and professionals who could be interested to deal with such data are often spread across geographically distributed centers and research institutions, making it difficult to distribute the data for research collaborations.

The Energy Exascale Earth System Modeling Project is a large multi-laboratory and multi-institution initiative  involving over 100 scientists, software developers and management personnel from eight national laboratories of the US Department of Energy (DOE), several academic institutions and private industry partners. Its main objective is to develop and utilize the ultra-high resolution state-of-the-art earth system model, called the Energy Exascale Earth System Model (E3SM) to address the mission-specific climate change and energy research priorities of the United States and to optimize the use of DOE’s next-generation computational facilities.

On July 31, 2018, Computational Climate Scientist Valentine Anantharaj from the Oak Ridge National Laboratory illustrated the Project during a CMCC webinar.  
In this talk, Valentine Anantharaj described how (a) they plan and schedule the modeling across various supercomputing facilities; (b) gather and manage the data; (c) analyze and publish the results; and (d) preserve the provenance of their activities. He also discussed some of the lessons learned, especially the challenges involved in orchestrating the simulations and workflow across major computing and data facilities.

Watch the video:

Presenter: Valentine Anantharaj – Oak Ridge Computing Faculty, Oak Ridge National Laboratory
Moderator: Sandro Fiore – CMCC Foundation – Euro-Mediterranean Center on Climate Change (ASC Division)

Start typing and press Enter to search

Shopping Cart