Platon — a Problem Solving Environment for Computational Steering of Evolutionary Optimisation on the Grid

PLATON is a Problem Solving Environment (PSE) for distributed numerical optimisation that allows the construction and steering of large-scale optimisation processes on a computational Grid. It is also an environment for coupling optimisation and simulation software. Evolutionary Optimisation is a good example of a problem which can efficiently designed to run in a Grid environment. After a short introduction to PLATON, we focus on the generic interface description used to describe the software components and couple them together, the tuple-space concept of distributed computing and the functionality of our object oriented generic communication template library, which uses standard low-level communication tools like PVM and MPI, but hides the complexity of their actual use.