SoC | Main / Soc2007

CSCS Soc 2007.

For the Summer of Code 2007, CSCS would like to find students interested in working on one of the following projects. We anticipate we need 1 student for the first project, and 1-3 students for the second one.

If you have questions about these projects, please contact the mentor for these projects, Rick.Riolo@gmail.com, Director of the CSCS Computing Lab.

If you have a project you think are related to complex systems research, please feel free to contact us, or just submit a proposal via SoC to CSCS. If it is a project we feel we can mentor and one that is related to complex systems research, we will consider it.

Note: Some information you should include in your application:

Information about the skills you bring to the project. (classes taken, programming langauges you know, past projects, etc.)
What skills you would like to gain or improve while working on this project.
What interests you about the project(s), including how the project(s) fit in with your other interests and goals.

Project #1: Complete GridSweeper

GridSweeper is a tool for running batch simulations with different parameter settings on grid computing systems. The core architecture was implemented for Google Summer of Code 2006, including code for generating many kinds of parameter sweeps, transferring data files via FTP, and submitting jobs to the grid via DRMAA. For the software to be usable, several pieces need to be implemented and tested:

A command-line interface for submitting and monitoring jobs.
A plug-in to support running arbitrary command-line tools.
A plug-in to interface with models built with Repast
A graphical user interface for submitting and monitoring jobs.

For wider distribution, more complete documentation and packaging will also be needed.

For more information about GridSweeper development last year, and the state of the project now, see Soc2006.

Candidates for working on the GridSweeper project should have OOP experience, preferably with Java and/or Objective-C. Experience designing and implementing GUI's would be useful as well.

Project #2: The I-SPOC project

Increasingly, scientists from across disciplines rely on computational models to build deeper understanding and better intuitions about the behavior and dynamics of complex systems. The widespread use of computational models has not been paralleled by adjustments in the traditional formats and venues for scientific communication, i.e., papers in hardcopy journals and books.

The overall goal of this project is to build computational and social infrastructure to support the use of a new form of scientific communication called a SPOC (Scientific Paper with Open Communication). A SPOC combines a standard academic paper with open source computational models written in any publicly accessible computer language. SPOCs will (i) link computational results with the models that produce them, allowing independent verification and validation (ii) create incentives for cleaner, more transparent code and for the sharing of code (iii) enable others to extend and improve existing computational models and to verify model robustness (iv) bring computational models to life allowing faculty, students, and other scholars to see dynamic phenomena emerge and (v) have an enormous effect on the teaching of science.

The CSCS SoC-2007 project includes two main tasks:

Building foundational infrastructure, called I-SPOC, which involves implementing software to create a web-accessible repository to store SPOCs and user-friendly interfaces for creators and users of SPOCs; and
The creation of a set of some initial SPOCs, both to test the infrastructure (including user interfaces) and to serve as examples for others to emulate. Creating the initial set of SPOCs will involve implementing some SPOCs for "classic" computational models of complex systems, as well as some creating some SPOCs chosen to exercise particular features of the I-SPOC infrastructure/website. The models to be implemented (and in some cases, extended) will be based on simple computational models of various social, biological and physical complex systems. Some examples of the kinds of models to be implemented include those discussed in the books listed here and the models described here.

For a more detailed description of SPOCs and the I-SPOC project, please see this pdf document.

As described above, part of the overall I-SPOC project involves building the infrastructure to support the web-based I-SPOC repository. Thus candidates interested in building the I-SPOC should have comprehensive experience with at least one content management system (e.g. Nuke variant, wiki variant, drupal). It also would be useful to have strong web application development skills (PHP/SQL/HTML), as well as some experience designing and implementing user interfaces to web-based databases.

The second part of the I-SPOC project involves implementing computational models of complex systems, most often as "agent/individual based" models. Therefor candidates for this part of the project should foremost be interested in computational modeling in general, and agent/individual based modeling in particular. Candidates should be skilled in OOP, preferably in Java. Experience using (and writing analysis programs in) one or more packages such as Mathematica, Matlab, R, etc, would also be useful.