Shaffer

``Collaborative Problem Solving Environments''
Prof. Clifford A. Shaffer
Department of Computer Science
Virginia Tech
Blacksburg, VA 24061
E-mail: shaffer@cs.vt.edu
URL: http://www.cs.vt.edu/~shaffer/

Long before digital libraries, code reuse, or the Internet became popular terms --- or even distinct topics of research --- the numerical analysis community created a digital library of reusable code available via the Internet. Netlib is still widely used today by scientists and engineers in a wide range of disciplines. Today, successor efforts such as NetSolve are improving the access of scientists and engineers to high-quality software. It is important to note that, while crucial to the scientific endeavor, these efforts have little to do with numerical analysis or high performance computing. Instead, they have everything to do with marshaling and distributing resources so that a wide community can make the most of its investment in high-quality software.

The scientific computing research community would achieve greater productivity if it can address a complex of process management issues that are endemic to scientific and engineering research. Our next goal should be to provide appropriate software support and process management standards that will allow scientists and engineers to make the most of their existing codes and computing resources.

The experience at Virginia Tech and elsewhere in dealing with large-scale engineering design and analysis problems has demonstrated the following critical problems:

It is difficult to integrate codes from multiple disciplines developed by a diverse group of people on multiple platforms located at widespread locations;
It is difficult to share software between potential collaborators in a multidisciplinary effort --- difficult even for a team to continue using a research code once the author has left the group;
Current tools for synchronous collaboration are inadequate.

Integrating codes from different disciplines raises both pragmatic and conceptual issues. Pragmatically, the issue is how best to support the interoperability of independently-conceived programs residing on diverse, geographically distributed computing platforms. Another pragmatic concern is that large, complicated codes now exist that cannot simply be discarded and rewritten for a new environment. However, interoperability is best achieved by adherence to common protocols of data interchange and the use of clearly identified interfaces. The notions of interfaces and protocols lead directly to the domains of object-oriented software and distributed computing. Thus, a key issue is how to unify legacy codes, tied to specific machine architectures, into an effective whole. Conceptually, the key issue is how to foster coordinated problem solving activities among multiple experts in different technical domains, and leverage existing codes and computer hardware resources connected by the Internet.

Many of these problems can be resolved through the use of collaborative problem solving environments (CPSE). A CPSE is a system that provides an integrated set of high level facilities to support groups engaged in solving problems from a proscribed domain. A CPSE allows users to define and modify problems, choose solution strategies, interact with and manage the appropriate hardware and software resources, visualize and analyze results, and record and coordinate problem solving tasks. Perhaps most significantly, users communicate with a CPSE at a higher level than the language of a particular operating system, programming language, or network protocol. Expert knowledge of the underlying hardware or software is not required.

Key elements to CPSEs include the following:

CPSEs should be based on aggressive use of new software engineering technology for code reuse and interoperability such as component frameworks (one example is JavaBeans). A properly designed component framework architecture for scientific computing would allow users to link together data generators such as simulations, data filters to convert data formats, and data visualizations.
An integral part of the CPSE must be support for access to distributed resources. For example, our own work uses JavaBeans acting as proxies for simulation codes and files on high-performance computers. These proxy beans are linked together on a local workstation in ways that allow scientists and engineers to combine simulation codes to produce new results, and to link these results in turn to visualization programs.
The CPSE must also support real-time (synchronous) collaboration among teams of scientists and engineers. Team members should be able to see and manipulate the workspace together despite being in separate places.
The CPSE must include process support in ways that encourage effective documentation of simulation and visualization codes, along with true interoperability. As mentioned above, a primary problem encountered by multidisciplinary groups is that existing codes are hard to use by newcomers. Within a CPSE, each component should carry with it effective support for understanding the conditions under which the component should be used, and how it should link to other components. It is equally important that the builders of new components be motivated to provide the necessary documentation.
The CPSE should support changing models and conducting multiple experiments. Basic tools such as effective version control for experiments, and automatically documenting the models and parameters used for a given run, would be helpful. So would automatic recompilation of the underlying simulations when the model is changed by selecting a new subroutine within a larger simulation code.
Finally, the CPSEs should be created with a proper understanding of how the application domain works. We still do not understand enough about how multidisciplinary teams work to be sure what the best software tools will be, and so we need to study such teams in operation. Participatory design techniques will be crucial as Computer Scientists work with the scientists and engineers who are the domain experts to create effective CPSEs.

URLs to Ongoing Work