A Grid Framework for Computational Mechanics Applications MICHAEL M. RESCH, NATALIA CURRLE-LINDE, UWE KÜSTER Höchstleistungsrechenzentrum Stuttgart (HLRS) University of Stuttgart Nobelstrasse 19, 70569 Stuttgart GERMANY BENEDETTO RISIO RECOM Services Nobelstrasse 15, 70569 Stuttgart GERMANY Abstract Currently, numerical simulation using automated parameter studies is a key tool in discovering functional optima in complex systems. In future, such studies of complex systems will be important for the purpose of steering simulations. One example is the optimum design and steering of high power furnaces of power plants. Grid technology makes it possible to carry out sophisticated simulations. However, the large scale of such studies requires organized support for the submission, monitoring, and termination of jobs, as well as mechanisms for the collection of results, and the dynamic generation of new parameter sets in order to intelligently approach an optimum. In this paper we present a Grid framework consisting of GriCoL (Grid Concurrent Language), which we propose as a simple and efficient language for the description of complex Grid applications, along with SEGL (Science Experimental Grid Laboratory), the problem solving environment within which GriCoL works. We apply this framework to a computational mechanics application with industrial applicability, the simulation of high power furnaces at a power plant. Key Words: Grid, GriCoL, SEGL, numerical simulation, dynamic parameter studies, computational mechanics 1 Introduction During the last 20 years the numerical simulation of engineering problems has become a fundamental tool for research and development. In the past, numerical simulations were limited to a few specified parameter settings because expensive computing time did not allow for more. Today, enormous computer resources, which can be provided by the Grid [1], enable the simulation of complete ranges of multi-dimensional parameter spaces in order to predict an operational optimum for a given system. We have used parameterized simulations in many disciplines. Examples are drug design by molecular dynamics, statistical crash simulation of cars, airfoil design by varying airfoil parameters, power plant simulation by varying burners and fuel quality. The mechanism proposed here offers a unified framework for such large-scale optimization problems in design and engineering. The framework levers the resources of the Grid using GriCoL (Grid Concurrent Language), a language for describing complex modeling experiments, utilized within its problem solving environment SEGL (Science Experimental Grid Laboratory). 1.1 Existing Tools for Parameter Investigation Studies There are some efforts in implementing such tools e.g. Nimrod [2], Ilab [3] or SkyFlow [4]. These tools are able to generate parameter sweeps and jobs, running them in a distributed computer environment (Grid) and collecting the data. ILab also allows the calculation of multi-parametric models in independent separate tasks in a complicated workflow for multiple stages. However, none of these tools is able to perform the task dynamically by generating new parameter sets by an automated optimization strategy as is needed for handling complex parameter problems. In addition to the above mentioned environments, tools like Condor [5], UNICORE [6] or AppLeS [7] (Application-Level Scheduler) can be used to launch pre-existing parameter studies using distributed resources. These, however, give no special support for dynamic parameter studies. Complex parameter studies can be facilitated by allowing the system to dynamically select parameter sets on the basis of previous intermediate results. This dynamic parameterization capability requires an iterative, self-steering approach. Possible strategies for the dynamic selection of parameter sets include genetic Proceedings of the 2nd WSEAS Int. Conference on Applied and Theoretical Mechanics, Venice, Italy, November 20-22, 2006 360