Acquisition, Sharing, and Processing of Large Data Sets for Strain Imaging: An Example of an Indented Ni 3 Al/Mo Composite N. S. MCINTYRE, 1,4 R. I. BARABASH, 2 J. QIN, 1 M. KUNZ, 3 N. TAMURA, 3 and H. BEI 2 1.—The University of Western Ontario, London, ON N6A 5B7, Canada. 2.—Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA. 3.—Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA. 4.—e-mail: smcintyr@uwo.ca The local effects of stress from a mechanical indentation have been studied on a Ni 3 Al single crystal containing submicron inclusions of molybdenum fibers. X-ray microdiffraction (PXM) was used to measure elastic and plastic defor- mations near the indents. An analysis of freshly acquired massive sets of PXM data has been carried out over the Science Studio network using parallel processing software called FOXMAS. This network and the FOXMAS software have greatly improved the efficiency of the data processing task. The analysis was successfully applied to study lattice orientation distribution and strain tensor components for both the Ni 3 Al and the Mo phases, particularly around eight indents patterned at the longitudinal section of the alloy. INTRODUCTION The subtle and localized structural properties in real crystal systems can now be addressed using highly coherent x-ray beams produced by synchro- tron radiation. Polychromatic x-ray microdiffraction (PXM) 1–5 uses a submicron beam of polychromatic (‘‘white’’) synchrotron x-rays to produce a Laue dif- fraction pattern for each microscopic region irradi- ated by the beam. Each x-ray pattern can now be more accurately assessed than is possible with related electron beam techniques like electron back- scattering diffraction. For example, PXM provides deviatoric strain tensors 1 that provide information on primary residual elastic strain directions. 2 Fur- thermore, the shapes of the diffraction spots can be analyzed to indicate the extent and direction of slip systems, 3 dislocation density, and the presence of dislocation walls. 3,4 Local grain misorientation can also be obtained from PXM, thus providing the pos- sibility of individual separate assessments of both residual elastic and plastic deformation in the crystal. This diffraction information is obtained for each microscopic region of the crystal and is usually displayed as maps that show changes in many of the crystal properties as a function of position. Frequently, thousands of diffraction images are collected in the course of a few hours of experi- mentation. An analysis of PXM data is a complex process. First, the approximate position of the most intense spots in each of the diffraction images must be found. Then, the geometric center of the spot is determined as well as its shape. Finally, the spot centers are indexed according to the crystallo- graphic data that are provided by the user. Earlier computer programs were developed by Larson and co-workers 1,5 and Tamura et al., 2 but there are now several versions, each with some common elements, but each is designed to meet the particular needs of the originating scientist. The analysis process is thus exacting and lengthy due to requirements for mathematical convergence at each stage of the cal- culation. The analysis is further complicated by the large number of diffraction images to provide microstructural information through mapping. Typically, the analysis of the thousands of PXM images takes a number of days using a modern desktop computer. Not only is this a gross inconve- nience to any protracted multisample experiment, but also it is made even more impractical by the desirability to refine the analysis process by varia- tion of some fitting parameters. Furthermore, the extended PXM analysis sequence makes it impos- sible to ensure that data being collected during an experiment are of useful quality. Therefore, the development of faster parallel computational facili- ties for PXM data are important. JOM, Vol. 65, No. 1, 2013 DOI: 10.1007/s11837-012-0496-9 Ó 2012 TMS (Published online November 8, 2012) 29