David Molnar

Using services and service compositions to enable the distributed execution of legacy simulation applications

In the field of natural and engineering science, computer simulations play an increasingly important role to explain or predict phenomena of the real world. Although the software landscape is crucial to support scientists in their every day work, we recognized during our work with scientific institutes that many simulation programs can be considered legacy monolithic applications. They are developed without adhering to known software engineering guidelines, lack an acceptable software ergonomics, run sequentially on single workstations and require tedious manual tasks. We are convinced that SOA concepts and the service composition technology can help to improve this situation. In this paper we report on the results of our work on the service- and service composition-based re-engineering of a legacy scientific application for the simulation of the ageing process in copper-alloyed. The underlying general concept for a distributed, service-based simulation infrastructure is also applicable to other scenarios. Core of the infrastructure is a resource manager that steers server work load and handles simulation data.

Atomistic multiscale simulations on the anisotropic tensile behaviour of copper-alloyed alpha-iron at different states of thermal ageing

The mechanical behaviour of steels is strongly related to their underlying atomistic structures which evolve during thermal treatment. Cu-alloyed α-Fe undergoes a change in material behaviour during the ageing process, especially at temperatures of above 300°C, where precipitates form on a large time-scale within the α-Fe matrix, yielding first a precipitation strengthening of the material. As the precipitates grow further in time, the material strength decreases again. This complex process is modelled with a multiscale approach, combining Kinetic Monte Carlo (KMC) with Molecular Dynamics (MD) simulations in a sequential way and exploiting the advantages of both methods while simultaneously circumventing their particular disadvantages. The formation of precipitates is modelled on a single-crystal lattice with a diffusion based KMC approach. Transferring selected precipitation states at different ageing times to MD simulations allows the performance of nano tensile tests and the analysis of failure initiation. The anisotropic tensile behaviour is investigated in the [100], [110] and [111] directions, showing monotonically decreasing tensile strengths and deformation strains. Hence precipitation strengthening is mainly due to dislocation–precipitate interactions which are non-existent at small tensile loadings in this scenario. At the point of ductile failure, dislocations are generated at the interfaces between precipitates and the Fe matrix. Straining in the [100] direction, they lie on {110} and {112} glide planes, as expected. With the method presented here, the changes of the anisotropic tensile moduli are related to different states of thermal ageing, i.e., to nucleation, growth and Ostwald ripening of Cu precipitates.

Plasmon hybridization enhances the transient absorption signal of a single nanoparticle

A tiny variation of a single metal nanoparticles dielectric properties has only a weak influence on the light field. We demonstrate, using antenna concepts, how plasmon hybridization helps to increase the particles influence.