Research

Distributed Computing

For my Ph.D., I am working with Dr. Douglas Thain as a member of the Cooperative Computing Lab. Currently, my main research focus is investigating methods of combining various distributed system tools and services into a single framework or programming model. The goal of such a software framework is to simplify the use of distributed abstractions and services so that end users may construct scalable distributed workflows.

Storage

As part of my research, I develop and maintain a few filesystem adapters for the Parrot middleware application along with various other storage service software.

  • bxgrid-query: Script that provides command line access to the BXGrid web services.

  • parrot-bxgrid: This module provides transparent read-only access to any ROARS filesystem. Currently it is used extensively by our internal BXGrid biometrics website to provide fault-tolerant and robust access to biometric data.

  • parrot-hdfs: This module provides read-write access to the Hadoop Distributed Filesystem and has been testing using a variety execution engines such as Condor and WorkQueue.

Computation

In addition to my work on storage services, I also experiment with various computational abstractions and frameworks.

  • Weaver: This high level distributed computing workflow framework in Python designed to simplify and facilitate the use various computational and storage abstractions provided by the CCL.

  • python-workqueue: Python WorkQueue bindings to enable the development of master/worker type applications.

  • Starch: Tool for creating executable standalone application archives.

Operations

I also contribute to the maintenance and development of various CCL and BXGrid operations.

Publications

Journals

  1. Irena Lanc, Peter Bui, Douglas Thain, and Scott Emrich. Adapting Bioinformatics Applications for Heterogeneous Systems: A Case Study. Submitted to Concurrency and Computation: Practice and Experience. October, 2011.
  2. Peter Bui, Li Yu, Andrew Thrasher, Rory Carmichael, Irena Lanc, Patrick Donnelly, and Douglas Thain. Scripting distributed scientific workflows using Weaver, Concurrency and Computation: Practice and Experience. November, 2011.

Conferences

  1. Patrick Donnelly, Peter Bui, and Douglas Thain. Attaching Cloud Storage to a Campus Grid Using Parrot, Chirp, and Hadoop. IEEE International Conference on Cloud Computing Technology and Science, pages 488-495, November, 2010.

Workshops

  1. Peter Bui, Dinesh Rajan, Badi Abdul-Wahid, Jesus Izaguirre, Douglas Thain. Work Queue + Python: A Framework For Scalable Scientific Ensemble Applications. Workshop on Python for High Performance and Scientific Computing at SC 2011. November, 2011.
  2. Irena Lanc, Peter Bui, Douglas Thain, and Scott Emrich. Adapting Bioinformatics Applications for Heterogeneous Systems: A Case Study. The Second International Workshop on Emerging Computational Methods for the Life Sciences, pages 7-14. June, 2011.
  3. Andrew Thrasher, Rory Carmichael, Peter Bui, Li Yu, Douglas Thain, and Scott Emrich. Taming Complex Bioinformatics Workflows with Weaver, Makeflow, and Starch. Workshop on Workflows in Support of Large Scale Science, pages 1-6. November, 2010.
  4. Hoang Bui, Peter Bui, Patrick Flynn, and Douglas Thain. ROARS: A Scalable Repository for Data Intensive Scientific Computing. The Third International Workshop on Data Intensive Distributed Computing at ACM HPDC 2010. June, 2010.
  5. Peter Bui, Li Yu, and Douglas Thain. Weaver: Integrating Distributed Computing Abstractions into Scientific Workflows using Python. Challenges of Large Applications in Distributed Environments at ACM HPDC 2010. June, 2010.

Book Chapters

  1. Douglas Thain, Michael Albrecht, Hoang Bui, Peter Bui, Rory Carmichael, Scott Emrich, and Patrick Flynn. Data Intensive Computing with Clustered Chirp Servers. Data Intensive Distributed Computing: Challenges and Solutions for Large-scale Information Management, IGI Global, Chapter 7.

Technical Reports

  1. Michael Albrecht, Patrick Donnelly, Peter Bui, and Douglas Thain. Makeflow: A Portable Abstraction for Cluster, Cloud, and Grid Computing, Department of Computer Science and Engineering, University of Notre Dame, TR-2011-02.

Presentations

  1. Weaver: Simple Distributed Scientific Programming. Poster, CSE Student Research Symposium, University of Notre Dame. November 2010.
  2. Work Queue: A Scalable Master/Worker Framework. Tutorial, University of Notre Dame. June 29, 2010.
  3. Weaver: Simple Distributed Scientific Programming. Poster, Cyberinfrastructure Days, University of Notre Dame. April 2010.
  4. Weaving Abstractions into Workflows. Condor Week, University of Wisconsin. April 2010.

Image Registration

For my Master's thesis, I worked with Dr. Jay Brockman on accelerating image registration using GPGPU technology. Specifically, I profiled and analyzed a 2-D image registration algorithm and ported it to the CUDA framework to achieve significant speedup.

Publications

  1. Peter Bui and Jay Brockman. Performance Analysis of Accelerated Image Registration using GPGPU. The 2nd Workshop on General Purpose Processing on Graphics Processing Units, pages 38-45. March, 2009.

Thesis

  1. Peter Bui. AIR: Accelerated Image Registration. Master's Thesis, University of Notre Dame. November, 2009.

Service

  • Referee. CIGPU 2010. Computational Intelligence on Consumer Games and Graphics Hardware.