phone: I don’t answer
fax: send an email
office: 227 Malone Hall
mailing: 160 Malone Hall, 3400 N. Charles St, Baltimore, MD 21218
I’m most recently interested in the interaction between algorithms for machine learning and the memory hierarchy. This was the topic of my most recent class Advanced Topics in Data-Intensive Computing. We’ve had some initial papers in this direction. knor (K-means NUMA optimized routines) is the fastest k-means out there for R on GitHub and sometimes CRAN. Disa won the best presentation award for this work at HPDC. Also, Da’s paper FlashR: Parallelize and Scale R for Machine Learning using SSDs is a best-paper award candidate at the upcoming PPoPP.
Randal Burns is a Professor of Computer Science in the Whiting School of Engineering at Johns Hopkins University. His research interests lie in building the high-performance, scalable data systems that allow scientists to make discoveries through the exploration, mining, and statistical analysis of big data.
For the last five years, he has focused on data-systems for high-throughput neuroscience. He is a co-founder of NeuroData along with Joshua Vogelstein. NeuroData democratizes access to world-class data sets, including electron-microscopy connectomics, CLARITY, MRI, and array tomography data.
Randal is a core member of the JH Turbulence Database Group. He and his students have built Open Numerical Laboratories in which anyone can explore, mine, and analyze world-class turbulence simulations.
At Johns Hopkins, Randal is both a member of and on the steering committee of the Kavli Neuroscience Discovery Institute. He is a member of the Institute for Data-Intensive Science and Engineering.
Prior to arriving at JHU in 2002, Randal was Research Staff Member at IBM’s Almaden Research Center in San Jose. He earned his Ph.D. in 2000 and M.S. in 1997 from the Department of Computer Science at the University of California at Santa Cruz (Go Slugs!). He earned his B.S. degree from the Department of Geophysics at Stanford University (Go Cardinal!)