Exploring Parallelization of Nearest Neighbor Search and Clustering in High-Dimensional Space on Emerging Parallel Architectures with Applications in Computer Vision

University Of Arkansas, Fayetteville AR

Investigators

Abstract

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). This project will explore alternative parallelization techniques for very largescale data intensive applications on emerging new parallel architectures, including distributed memory computer clusters using multicore and GPGPU processors. The experimental environment includes an application and system for large scale image classification and orientation suitable for use in geographic information systems. A variety of evaluation hardware is expected to be available for the project. The ultimate goal of this research is to leverage large scale systems, especially storage, to provide services to end users in the form of geospatial data. The proposed research will leverage preliminary work by the PIs in the application of CUDA and MPI programming and will demonstrate the application of these technologies to very large scale storage, potentially holding multiple terabytes of aerial and satellite imagery.