This is a quick snapshot of research I have done and/or am currently doing. I would like to thank the Department of Homeland Security for their support since September 2003, as well as the support of Sandia National Laboratories as described below. Please note that, for all research done since September 2003, the following applies:

This research was performed while on appointment as a U.S. Department of Homeland Security (DHS) Fellow under the DHS Scholarship and Fellowship Program, a program administered by the Oak Ridge Institute for Science and Education (ORISE) for DHS through an interagency agreement with the U.S Department of Energy (DOE). ORISE is managed by Oak Ridge Associated Universities under DOE contract number DE-AC05-00OR22750. All opinions expressed in this paper are the author's and do not necessarily reflect the policies and views of DHS, DOE, or ORISE.

My interests include/have included the following:

• Random graphs, dynamic networks, and models
• Algorithm analysis and asymptotics
• Parallel programming models and languages
• Coding theory
• Scheduling algorithms

Random Graphs, Dynamic Networks, and Models

I have been working on this project with Dr. Quentin Stout since at least January 2005; writ large, we are looking at certain models for random graph generation and evolution that may be useful models of certain observed phenomena. No external publications available as of yet.

Algorithm Analysis and Asymptotics

I have been working on this project with Dr. Kevin Compton since September 2005. Also appealing to the "writ large" classification as there are no publications yet, we are considering some variations on the asymptotic work of Flajolet and others.

Parallel Programming Models and Languages

From June 2003 to July 2006, I collaborated with researchers at Sandia National Laboratories in Albuquerque, New Mexico on an investigative research project in parallel programming models and programming environments. This work focused on an evaluation of alternate programming models and environments to the message passing model (MPI), used throughout Sandia applications programming.

As a side note, the project briefly had the services of Phil H. Smith, who in a short time helped develop a prototype of the BEC runtime that produced some useful performance numbers. Phil's assistance was also quite useful in helping shape the project into an industrial-strength software project.

Posters

• Supercomputing 2003

Presentations

• Sandia Student Symposium 2003
• Homeland Security Scholars and Fellows, Sandia National Laboratories, 2004
• Sandia Student Symposium 2004
• Supercomputing 2004, Student Days
• Supercomputing 2004, UPC Booth
• EECS 587, the University of Michigan, November 10, 2004

Publications

• An Evolutionary Path towards Virtual Shared Memory with Random Access
  Jonathan Leighton Brown, Sue Goudy, Mike Heroux, Shan Shan Huang, and Zhaofang Wen
  ACM Symposium on Parallelism in Algorithms and Architecture 2006 (SPAA'06)
  Accepted as an announcement.
• BEC: A Virtual Shared Memory Parallel Programming Environment
  Jonathan Leighton Brown, Sue Goudy, Mike Heroux, Shan Shan Huang, and Zhaofang Wen
  Sandia National Laboratories Tech Report 2006-0421, January 2006
  This tech report provides an overview of BEC, the "bundle-exchange-compute" programming model. It is similar to the very-popular Bulk Synchronous Parallel (BSP) model, and provides a support infrastructure for PRAM C. The vast majority of my work on this involved designing the runtime library; development of this is currently underway.
• Toward an Application Support Layer: Numerical Computation in Unified Parallel C
  Jonathan Leighton Brown and Zhaofang Wen
  Sixth International Conference on Parallel Processing and Applied Mathematics (PPAM'05), September 2005
  In Lecture Notes on Computer Science.
• PRAM C: A New Programming Environment for Fine-Grain and Coarse-Grain Parallelism
  Jonathan Leighton Brown and Zhaofang Wen
  Sandia National Laboratories Tech Report 2004-6171, November 2004
  This tech report is a design proposal for PRAM C, a new programming environment for parallel programming that attempts to capture a PRAM-style of programming. PRAM C differs from other attempts to allow for PRAM-style programming in that it is a software only solution to the problem of fine-grain communication. This work was done in part while on a CSRI summer internship at Sandia National Laboratories in June - August 2004. Work was in part supported by the Department of Homeland Security as part of a graduate fellowship.
• Unified Parallel C and the Computing Needs of Sandia National Laboratories
  Jonathan Leighton Brown and Zhaofang Wen
  Sandia National Laboratories Tech Report 2004-6172, November 2004
  This tech report summarizes work done in evaluating the UPC environment available on the Cray T3e/750. Performance of sorting and minimum spanning tree algorithm implementations in UPC was analyzed and compared to MPI implementations. Some suggestions for improvements to the UPC implementation are included. This work was done in part while on CSRI summer internships at Sandia National Laboratories in June - August 2003 and June - August 2004, during an independent study at the University of Michigan with Dr. Quentin Stout in September - December 2003, and while on contract to Sandia National Laboratories in January - May 2004. The report is based on the final report provided to Sandia National Laboratories on the contract.
• Experiences Implementing Sorting Algorithms in Unified Parallel C
  Ron Brightwell, Jonathan Leighton Brown, Quentin Stout, and Zhaofang Wen
  Fourth Annual UPC Developers Workshop, September 2004

Coding Theory Topics

Scheduling Algorithms