期刊
出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.nima.2010.12.173
关键词
X-ray photon correlation spectroscopy; Parallel computing; Cluster architecture
With the development of third generation synchrotron radiation sources, X-ray photon correlation spectroscopy has emerged as a powerful technique for characterizing equilibrium and non-equilibrium dynamics in complex materials at nanometer length scales over a wide range of time-scales (0.001-1000 s). Moreover, the development of powerful new direct detection CCD cameras has allowed investigation of faster dynamical processes. A consequence of these technical improvements is the need to reduce a very large amount of area detector data within a short time. This problem can be solved by utilizing a large number of processors (32-64) in the cluster architecture to improve the efficiency of the calculations by 1-2 orders of magnitude [1] (Tieman et al., this issue). However, to make such a data analysis system operational, powerful and user-friendly control software needs to be developed. As a part of the effort to maintain a high data acquisition and reduction rate, we have developed a Matlab-based software that acts as an interface between the user and the high performance computing (HPC) cluster. (C) 2011 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据