Alumna Recognized for Early Career Excellence

An alumna of the Department of Energy Computational Science Graduate Fellowship (DOE CSGF) has added to her many awards with recognition from the Institute of Electrical and Electronics Engineers Computer Society (IEEE-CS).

Amanda Randles, Duke University assistant professor of biomedical engineering (with secondary appointments in mathematics, computer science and mechanical engineering), is one of three 2017 winners of the IEEE-CS Technical Consortium on High Performance Computing (TCHPC) Award for Excellence for Early Career Researchers in High Performance Computing. The award recognizes individuals “who have made outstanding, influential, and potentially long-lasting contributions in the field of high performance computing” in the five years following their Ph.D., a release says.

Randles and co-winners Antonio J. Peña of the Barcelona Supercomputing Center and Shuaiwen Leon Song of Pacific Northwest National Laboratory will receive their awards in November at the SC17 supercomputing conference in Denver.

Randles was a DOE CSGF recipient from 2010 to 2013 and earned a doctorate in applied physics from Harvard University in 2013. Her main research uses high-performance computing to understand how the body’s blood vessels behave. The highly demanding simulations have twice been a finalist for the prestigious Association for Computing Machinery (ACM) Gordon Bell Prize for achievement in high-performance computing.

One of Randles’ latest projects (and a 2015 Bell Prize finalist) is HARVEY, a fully functional human arterial system simulation named for the 16th-century English doctor who first described blood circulation. The program modeled coronary artery geometries of a specific patient and scaled to 294,912 computer processors.

Randles continues refining her models and plans to add capillary beds, which deliver oxygen to tissues, and veins, which return oxygen-depleted blood to the heart and lungs. Such simulations will require even more computing power, so her group is working to port HARVEY to the next generation of machines.

Meanwhile, her lab also collaborates with clinicians, including pediatric cardiologists at Duke Health, to begin modeling treatment options for individual patients. Such simulations will help doctors make better recommendations.

Earlier this year, MIT Technology Review magazine named Randles one of “35 Innovators Under 35” for 2017. In 2014, she was awarded the NIH Early Independence Award to support the development of models of cancer migration in human vasculature. Randles is a co-inventor on 115 U.S. parallel computing patents and won the ACM/IEEE-CS George Michael High Performance Computing Fellowship in 2010 and 2012.

DEIXIS, the DOE CSGF annual, profiled Randles on pages 22-23 of the 2017 issue.