Publications

Bazinet, A. L., and M. P. Cummings. The Lattice Project: a Grid research and production environment combining multiple Grid computing models. In Weber, M. H. W. (Ed.) Distributed & Grid Computing - Science Made Transparent for Everyone. Principles, Applications and Supporting Communities. Tectum. To appear.

• Synopsis: The Lattice Project is the research that our laboratory does in Grid computing, and its original aim has been to fulfill the ever-increasing computational needs of life scientists and other researchers. Our Grid system has performed well over 100 CPU years of computation since it came online. It is based on a novel Grid architecture that encompasses resources from high-end clusters and multiprocessors to individual desktop computers, which has partly been achieved by combining the functionality of the Globus and BOINC toolkits. The Lattice Project continues to develop as the primary Grid computing solution at the University of Maryland.

Myers, D. S., A. L. Bazinet and M. P. Cummings. 2008. Expanding the reach of Grid computing: combining Globus- and BOINC-based systems. Pages 71-85. In E.-G. and A. Zomaya (Eds.) Grids for Bioinformatics and Computational Biology, Wiley Book Series on Parallel and Distributed Computing. John Wiley & Sons, New York.

• Synopsis: Grid computing systems can be broadly classified into two types. The first type are heavy-weight, feature-rich systems that tend to concern themselves primarily with providing access to large-scale, intra- and inter-institutional resources such as clusters and multiprocessors. Grid systems developed using the Globus Toolkit are examples of this class. The second general class of Grid computing systems is the Desktop Grid, in which cycles are scavenged from idle desktop computers. BOINC (a descendant of the SETI@home project) is an example of a public Desktop Grid, as it harnesses resources that exist outside of institutional control. Here, we describe middleware that allows Globus- and BOINC-based Grid systems to inter-operate. This middleware allows Globus-based computational Grids to incorporate a much wider range of resources. Additionally, by decreasing the startup cost for new Desktop Grid computing projects, it makes Desktop Grid a viable option for many kinds of applications. We have effectively used a combined Globus- and BOINC-based Grid system in computational biology research.

Bazinet, A. L., D. S. Myers, J. Fuetsch and M. P. Cummings. 2007. Grid Services Base Library: a high-level, procedural application program interface for writing Globus-based Grid services. Future Generation Computer Systems 22:517-522.

• Abstract: The Grid Services Base Library (GSBL) is a procedural application programming interface (API) that abstracts many of the high level functions performed by Globus Grid services, and dramatically lowers the barriers to writing Grid services. The library has extensively tested and used for computational biology research in a Globus Toolkit-based Grid system, in which nineteen bioinformatics Grid services written with this API are deployed.

Lee, S., T. D. Wang, N. Hashmi and M. P. Cummings. 2007. Bio-STEER: a Semantic Web workflow tool for Grid computing in the life sciences. Future Generation Computer Systems 23:497-509.

• Abstract: Life science research is becoming evermore computationally intensive. Hence, from a computational resource perspective, Grid computing provides a logical approach to meeting many of the computational needs of life science research. However, there are several barriers to the widespread use of Grid computing in life sciences. In this paper, we attempt to address one particular barrier: the difficulty of using Grid computing by life scientists. Life science research often involves connecting multiple applications together to form a workflow. This process of constructing a workflow is complex. When combined with the difficulty of using Grid services, composing a meaningful workflow using Grid services can present a challenge to life scientists. Our proposed solution is a Semantic Web-enabled computing environment, called Bio-STEER. In Bio-STEER, bioinformatics Grid services are mapped to Semantic Web services, described in OWL-S. We also defined an ontology in OWL to model bioinformatics applications. A graphical user interface helps to construct a scientific workflow by showing a list of services that are semantically sound; that is, the output of one service is semantically compatible with the input of the connecting service. Bio-STEER can help users take full advantage of Grid services through a user-friendly graphical user interface (GUI), which allows them to easily construct the workflows they need.

Cummings, M. P., J. C. Huskamp. 2005. Grid computing. EDUCAUSE Review 40:116-117.

• This paper describes the appropriateness of Grid computing for academic institutions.

Myers, D. S., and A. L. Bazinet. 2004. Intercepting Arbitrary Functions on Windows, UNIX, and Macintosh OS X Platforms. Technical Report CS-TR-4585, UMIACS-TR-2004-28, Center for Bioinformatics and Computational Biology, Institute for Advanced Computer Studies, University of Maryland.

• Abstract: It is often desirable to modify the behavior of existing code bases by wrapping or replacing functions. When editing the source code of those functions is a viable option, this can be a straight-forward process. When the source of the functions cannot be edited (e.g., if the functions are provided by the system C library), then alternative techniques are required. Here, we present such techniques for UNIX, Windows, and Macintosh OS X platforms. We have used these techniques to update bioinformatics applications to call the application program interface (API) provided by the Berkeley Open Infrastructure for Network Computing (BOINC), a distributed computing toolkit.

Myers, D. S., and M. P. Cummings. 2003. Necessity is the mother of invention: a simple Grid computing system using commodity tools. Journal of Parallel and Distributed Computing. 63:578-589.

• Abstract: Access to sufficient resources is a barrier to scientific progress for many researchers facing large computational problems. Gaining access to large-scale resources (i.e., university-wide or federally supported computer centers) can be difficult, given their limited availability, particular architectures, and request/review/approval cycles. Simultaneously, researchers often find themselves with access to workstations and older clusters overlooked by their owners in favor of newer hardware. Software to tie these resources into a coherent Grid, however, has been problematic. Here, we describe our experiences building a Grid computing system to conduct a large-scale simulation study using "borrowed" computing resources distributed over a wide area. Using standard software components, we have produced a Grid computing system capable of coupling several hundred processors spanning multiple continents and administrative domains. We believe that this system fills an important niche between a closely coupled local system and a heavyweight, highly customized wide area system.