*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Austin, Texas | Posted: November 20, 2015
PhD Candidate Patrick Flick (CSE) won "Best Student Paper" at the Supercomputing '15 conference -- held this week in Austin, Texas -- where a contingent of more than 50 students, faculty, and researchers from Georgia Tech advanced the field of high-performance computing (HPC).
Flick, working with Professor Srinivas Aluru (CSE), created parallel algorithms for distributed-memory construction that are 110x times faster than the best method running on a sequential, single computer. Using the human genome as a racetrack to test his speed, Flick was able to index the entire human genome in just 7.3 seconds running on 1024 Intel Xeon cores.
It is believed to be the first algorithm and implementation that uses this approach for distributed-memory parallel systems, and an important one for analyzing complex biological data.
“Bioinformatics is an example of a scientific field that is extremely data intensive; speed matters and speed helps,” Flick said. “We are not aware of any other parallel suffix array or suffix tree construction algorithms which achieve speedups this high.”
Next, Flick is working on a journal paper that includes more improvements, additional techniques, and further showcases the algorithms on real applications. Flick also authored another paper at SC'15 with fellow graduate students Chirag Jain (CSE) and Tony Pan (CSE) about how to partition large graphs that arise in metagenomics -- another data-intensive application area.
PhD Candidates Dipanjan Sengupta (CS) and Kapil Agarwal (CS) also were nominated for best paper at SC'15 after developing a scalable framework (dubbed “GraphReduce”) to process large graphs that exceed a device’s GPU memory.
Other activity by Georgia Tech at SC'15 included:
For more about what happened at this year's conference, visit http://sc15.supercomputing.org/
See photos of Georgia Tech at SC'15 from throughout the week.