Purdue University

Jingjing Liang, an assistant professor of quantitative forest ecology in the department of forestry and natural resources, recently led a team that produced the first ever estimate of the total number of tree species in the world – and they did it with the help of Research Computing’s Bell community cluster supercomputer.

In order to build their model, Liang and his collaborators needed to collect and compile a huge database of data from all over the world. Because this was such a large dataset – more than one million sample plots from around the world – it required the computational power of a supercomputer to analyze.

Using the high memory nodes of Bell (and the Snyder cluster before it), Liang’s team arrived at their estimate of 73,000 tree species worldwide – 9,000 more species than have been documented.

The resulting paper was published in the Proceedings of the National Academy of Sciences (PNAS).

“It’s great to have these world-class HPC systems at Purdue,” says Liang. “Without these systems, we would not have been able to finish this study.”

Liang is also an avid user of Scholar, Purdue’s community cluster for classroom instruction. His graduate class “Big Data, AI and Forests” learns to use the statistical analysis software R on Scholar. He has also recently introduced Scholar into his undergraduate classes.

Liang appreciates that the Scholar cluster lets the students log into the remote desktop from anywhere and run the same type of software in a consistent matter.

“It makes my teaching methods and the classroom experience for the students a lot better,” he says.

To learn more about how researchers can use Bell, Scholar or other Research Computing resources, contact rcac-help@purdue.edu.

Writer: Adrienne Miller, science and technology writer, Research Computing, mill2027@purdue.edu.