A community of researchers, scientists, and staff at the University of Maryland are working to integrate and deploy computing resources, Grid middleware, specialized scientific application software and web services in a comprehensive Grid system for scientific analysis. We call this effort The Lattice Project. The lead researchers are Michael Cummings and Adam Bazinet.
A broad spectrum of life science research will benefit greatly from increased access to high performance computing resources. Several properties of modern life sciences research drive the need for high performance computing:
- Trend toward quantitative analysis in experimental biology. The nature of much life sciences research involves inference and hypothesis testing that require computation.
- The rapidly increasing size and number of analysis inputs including databases, samples, and experiments.
- Increasing complexity of analytical models and concomitant increase in possible solution space. Maximum likelihood calculations, Monte Carlo and other stochastic simulation methods increasingly involve more parameters. While such models can increase our understanding of the underlying natural processes under investigation, in many cases they lead to an exponential increase in the number of possible solutions.
Despite promising developments that individually address these issues, the computational environment in which many life scientists work remains substantively deficient, materially impeding progress in a number of fields. It is our goal to make sufficient computational and data resources available to an active community of researchers.
The Lattice Project is developing a community-based Grid system that integrates Grid middleware technologies and widely used life science applications. This system is based on a novel Grid architecture that encompasses resources from high-end clusters and multiprocessors to individual desktop computers. We are strongly committed to the principles of open source software development, and we intend to release all software as freely-available source code except in those very few cases where commercial software is used.