We are studying the role emergent algorithmic strategies can play within
heterogeneous, high-performance grid-based environments.
We are especially interested in systems with reconfigurable hardware
components, fast interconnects, and CAD/PC workstation clusters.
We anticipate that such an environment can yield significant improvements
in performance for large-scale applications.
Of course many combinatorial problems relevant to resource sharing either
along a grid or within a cluster are NP-hard.
Thus we seek to determine how best to deal with this level of complexity.
A major goal is to explore the effectiveness of fixed-parameter tractability
as an algorithm design paradigm.
Among this project's foci are issues of trustworthiness, security,
reliability, hardware/software balancing, information sharing, and
computation/ communication tradeoffs.
