Porting Multiparametric MATLAB Application for Image and Video Processing to Desktop Grid for High-Performance Distributed Computing

Abstract

Optical microscopy is usually used for structural characterization of materials in narrow ranges of magnification, small region of interest (ROI), and in static regime. But many crucial processes of damage initiation and propagation take place dynamically in the wide observable time domain from 10-3 s to 103 s and on the many scales from 10-8 m (solitary defects places) to 10-2 m (correlated linked network of defects). We used one of them to observe in real-time regime the dynamic behavior of the material under mechanical deformation in loading machine, record its evolution, and apply our multiscale image processing software (MultiscaleIVideoP). Our calculations include many parameters of physical process (process rate, magnification, illumination conditions, hardware filters, etc.) and image processing parameters (size distribution, anisotropy, localization, scaling parameters, etc.), hence the calculations are very slow. That is why we have the extreme need of more powerful computational resources. The GRID-version of the proposed application MultiscaleIVideoP would have a very wide range of potential users, because modern laboratories has commercial microscopes with digital output connection to PC and perform everyday tasks of complex static and dynamic morphology analysis: in biology, geology, chemistry, physics, materials science, etc.

Deploying this application on a Grid computing infrastructure, utilising hundreds of machines at the same time, allows harnessing sufficient computational power to undertake the simulations on a larger scale and in a much shorter timeframe. Running the simulations and analysing the results on the Grid provides the excessive computational power required.

Yuri Gordienko, Institut de Physique du Metal – Kiev – Ukraine