This
1-day workshop will focus on the challenges of developing software to support
scientific research. The targeted participants are CSU scientists who need to
develop relatively complex software either as a research product or as a tool
to support research. The software can range from web applications built to
access and visualize large data sets, to complex modeling and simulation
software that run on distributed computing platforms. Discussions and
presentations at the workshop will cover at least the following topics:
CSU's
ISTeC (Information Science & Technology Center) is a university-wide
organization for promoting, facilitating, and enhancing CSU's research,
education, and outreach activities pertaining to the design and innovative
application of computer, communication, and information systems. For more
information please see istec.colostate.edu.
The
following are the aims of the workshop:
The
following are some of the outcomes we anticipate:
The
workshop will be structured into four sessions. In the first three sessions,
presenters from the CSU community outside of Computer Science will present
ongoing and planned software development work. Each presenter will have a
maximum of 15 minutes to present their work. Each session will end with a least
a 30 minute discussion on the work presented in the session. In the fourth
session, we will present a brief summary of the challenges and problems
identified during the workshop. Computer Science faculty members will also
present some of their experience with tackling the challenges and problems, and
propose some specific areas of collaboration. This will be followed by an open
discussion.
If
you would like to make a presentation at the workshop or have questions about
the workshop, please contact Robert
France:
Email: france AT cs.colostate.edu
Tel: 970-491-6356
Search
for Rockwell Hall on the CSU campus map.
There
is no registration cost.
Lunch, morning and afternoon snacks will be provided courtesy of ISTeC. If you
plan to attend the workshop please register with Sharon VanGorder before July
8, 2008:
Email: vangord AT cs.colostate.edu
Tel: 970-491-5862
Fax: 970-491-2466
Please provide Sharon with the following registration information:
9:00-9:15AM Opening Remarks (Robert France)
9:15-10:00AM Session 1
(Moderator: Jim Bieman)
Presenters
Mark
van der Woerd, Biochemistry & Molecular Biology (SAXS project)
Chris
Wilcox, Department of Computer Science (SAXS project)
Saul
Lozano, Department of Microbiology, Immunology, and Pathology (Dengue DSS
project)
10:00-10:30AM Open Discussion
10:30-11:00AM Morning
Coffee/Tea Break
11:00-11:30AM Session 2
(Moderator: Sudipto Ghosh)
Presenters
Olaf
David, Dept. of Civil Engineering (OMS)
Peter
Clarke, Florida International University (Computer-Based Communication)
11:30-NOON Open Discussion
NOON-1:00PM Lunch (box
lunches provided)
If
weather permits, we’ll eat out in the College of Business courtyard, else
we will eat in the Rockwell Hall Conference Room – Room 119.
1:00-2:00PM Session 3 (Moderator: Chuck Anderson)
Presenters
Jim
Graham, Natural Resource Ecology Laboratory (Spatial analysis)
Anton
Betten, Department of Mathematics (Large scale computations)
Crystal
Redman, Department of Computer Science (Simulation tool)
2:00-2:30PM Open Discussion
2:30-3:00PM Afternoon
Coffee/Tea Break
3:00-4:00PM Session 4
Presentations by Computer Science Faculty (Moderator: Robert
France)
Presenters
Asa
Ben-Hur
Sanjay
Rajopadyhe
Kathy
Kwinn/Debra Bartlett
SE
group
4:00-4:30PM Open Discussion
and closing remarks
Below
are the presentations that we have information on thus far:
Topic:
SAXS (Small
Angle X-ray Scattering) Project
Presenter: Mark van der Woerd, Biochemistry & Molecular Biology
Title: Software Engineering on the SAXS Project
Abstract: The SAXS software project is a collaboration between
the Biochemistry and Molecular Biology, Mathematics, Statistics, and Computer
Science departments at Colorado State University. The project purpose is to
develop software to help identify and refine the structure and function of
proteins. An experimental technique called Small Angle X-Ray Scattering (SAXS)
is used in the investigation. We are currently developing a software framework
and several programs related to the SAXS investigation. We have completed a
'Scatter' program to compute the theoretical intensity curve from a rigid
molecular model, and a preliminary 'Topology' program that models flexible
molecular structure. The programs are run from a simple user interface which we
distribute along with the programs under an open source license. Significant
performance gains have been made from the inception of the project in early 2008.
The computer science research areas associated with the project are software
engineering and parallel computing.
Presenter: Chris Wilcox, Dept. of Computer Science
Title:
Dengue
Decision Support System
Presenters: Lars Eisen, Saul Lozano, Department of Microbiology,
Immunology, and Pathology
Title: Collaborative scientific simulation model development using
the Object Modeling System (OMS)
Abstract: The Object Modeling System (OMS) is a component based modeling
framework that uses an open-source software approach to enable the scientific
community to collaboratively address the many complex issues associated with
the design, development, and application of distributed hydrological and
environmental models. The modular modeling framework concept 1) facilitates
using the resources of a larger science community to work collaboratively on
development and application of a variety of modeling approaches to
multi-disciplinary problems; 2) provides a framework in which to objectively
compare alternative modeling approaches; and 3) provides a means of sharing the
latest modeling advances.
Presenter: Olaf David, Dept. of Civil Engineering, CSU
Title: Computer-Based Communication in Distributed Collaborative
Environments
Abstract: The advances and pervasiveness in the technology to support
electronic communication has resulted in the need for a new approach to model
and rapidly realize user-centric communication-intensive applications. One
approach to solving this problem is using an executable domain specific
modeling language (DSML) that supports the easy creation of communication
models and the rapid realization of these models. Such a DSML and the
supporting execution environment would be extremely beneficial in today's
collaborative scientific environment since the number of geographically
distributed research teams that need to collaborate in several domains
continues to increase. I will present a technology we are developing at Florida
International University that uses a DSML - Communication Modeling Language
(CML), which supports the creation and rapid realization of models for
communication-intensive applications. We are currently developing a prototype
of the Communication Virtual Machine (CVM) platform to model and realize
communication-intensive applications.
Presenter: Peter Clarke, School of Computing and Information Sciences,
College of Engineering and Computing, Florida International University, Miami,
Florida
Title: Building a web-based system for spatial analysis of species
potential distributions
Abstract: With the latest technology it is possible to provide an online
system that allows users to perform complex scientific analysis with little
training and is easy to use. Traditionally performing spatial analysis requires
collecting field data, purchasing remotely sensed data, purchasing and learning
statistical packages, and setting up high-performance computers to provide the
required performance. This process can take years and cost tens of thousands of
dollars in software and hardware. Today's technology provides the opportunity
to make spatial analysis available to virtually anyone with an Internet
connection. This makes scientific data and analysis available to a wider
audience including research scientists without the specific resources required.
Presenter: Jim Graham, Natural Resource Ecology Laboratory
Title: Large Scale Computations in Discrete Mathematics
Abstract:
The speaker is concerned with investigating and analyzing discrete
mathematical objects like optimal linear codes and incidence structures.
The talk will be centered around the following three questions:
What are we trying to compute and how?
What are the issues with developing software?
What are the issues with doing the actual computation?
Presenter: Anton Betten, Department of Mathematics
Title: MASON: A Multi-Purpose Simulation Framework
Abstract: Since 2002, the Evolutionary Computation Lab and the Center
for Social Complexity at George Mason University has been developing and
expanding MASON, a simulation framework written in Java that is used for
theMulti-Agent Simulation of Neighborhoods (or Networks). MASON has been
leveraged to create simulators beyond intelligent agent research. A few of the
domains modeled using MASON have ranged from Cooperative Observation, Keep-Away
Soccer, and Traffic Control to 3D Models of the Solar System. MASON provides a
mech- anism for discrete event simulation. Domain specifics like the actors and
environment in the simulator are easy to create and customize making MASON a
simple starting point for basic simulations. We will provide a tutorial that
outlines the basic knowledge needed to begin creating simulators with MASON.
Presenter: Crystal Redman, Dept. of Computer Science
Title: Machine learning with PyML
Abstract: PyML is an open-source machine learning environment which has
been available on sourceforge for around 6 years (see http://pyml.sf.net). In
this talk I will share my rationale and experiences in sharing software which
was developed for research purposes.
Presenters: Asa Ben-Hur, Department of Computer Science
Title: Software version control
Abstract: A version control system is a great asset to the development
and maintenance of complex software. This talk will outline advantages of using
version control and also some of the associated costs. It will mention some
heavily used tools, discuss criteria for picking a tool, and look at how
version control can be integrated with other tools commonly used in software
development.
Presenters: Kathy Kwinn, Debra Bartlett, Department of Computer Science