Ohio State engineers aid in prioritizing and optimizing big data from simulations

A new tool under development by engineers at The Ohio State University aims to assist scientists in comprehending the vast amount of data generated by large-scale simulations, such as those used for jet turbine engine design and climate change research. The tool will solve two major challenges for computational scientists who generate big data sets: deciding what data is most essential and displaying it in the most meaningful way possible.

Han-Wei Shen, professor of computer science and engineering, received a $735,200 three-year grant from the National Science Foundation (NSF) to develop time-varying analysis algorithms that will enable researchers to extract and analyze the most significant pieces of data while a simulation is still running. Jen-Peng Chen, associate professor of mechanical and aerospace engineering, is the co-principal investigator on the project.

“Essentially we’re hoping to develop a software library that can be easily used by different types of simulations,” said Shen. “We’re going to try to be as generic as possible.”

The ultimate goal of the research is to create an open source software framework that can be applied to any large-scale simulation. The researchers estimate that a prototype will be available in two years, with the full framework available by the end of the three-year project.

The researchers are using Chen’s simulations of turbo machinery in aerodynamics to demonstrate the algorithms’ effectiveness. Core data analytics technologies are used to facilitate effective summarization, indexing and triage of the data.

Shen’s research is also supported by a $750,000 grant from the Department of Energy, which is part of a $25 million five-year award for the SciDAC Institute for Scalable Data Management, Analysis and Visualization.

Both the NSF and DOE awards Shen received are part of President Obama’s Big Data Research and Development Initiative.