ABI: Development: Parallel Computing for Phylogenetics: Grid, Public and GPU Computing

University Of Maryland, College Park, College Park MD

Investigators

Abstract

The University of Maryland College Park is awarded a grant to support the development of cyberinfrastructure that will facilitate, accelerate, and improve the ability to infer evolutionary relationships among organisms through the use of several types of parallel computing methods. The central idea is to build upon and extend previous work at multiple different levels of parallel computing to better take advantage of a diversity of computing resources, particularly existing desktop processing capacity available through volunteer computing. The main product will be an enhanced grid computing system, customized for phylogenetic tree search and accessible through a user-friendly web interface, that links thousands of individual computers worldwide. The project will improve use of a second underutilized resource inside desktop computers, namely graphics cards and other types of processors not usually used for normal data processing, which can do many scientific calculations faster than the central processing unit (cpu). A free, open-source library for parallelized maximum likelihood calculations will be extended. Public computing provides a mechanism for meaningful public involvement in scientific research. Almost 17,000 people in 143 countries have volunteered use of their computers and have run evolutionary analyses through the current system. With additional project development effort, the number of volunteers is expected to increase. This model of computing not only increases the speed and quality of scientific research, but is also more energy efficient. Furthermore, the project provides opportunities to involve students and others in parallel and public computing research. Results of this project will be accessible via the PI's websites at http://openevolution.org and http://lattice.umiacs.umd.edu.