Michael Wehner, a senior staff scientist at the Energy Department‘s Lawrence Berkeley National Laboratory computational research division, has discussed the laboratory’s work to produce large volumes of weather simulation data in an interview with Information Week published Friday.
William Jackson writes climate researchers at the Berkeley lab have generated more than 400 terabytes of modeling data over the last two years with the facility’s Edison and Hopper supercomputers.
The big data produced from the weather simulation projects have created challenges for the lab with respect to data storage and analysis, Wehner told the publication.
He said one challenge Berkeley wants to address involves the input/output area that he believes factors into the supercomputers’ lack of ability to collect data for analysis, according to the report.
Wehner also told Information Week the research lab also faces barriers when it tries to share big data with other researchers for collaboration.
Jackson writes the DOE focuses on data management and data indexing efforts such as Federated Earth System Grid for opening hundreds of petabytes of simulation data to the climate research community and Toolkit for Extreme Climate Analysis program for climate data visualizations.