CRIF: Purchase of a Resource for Computational Chemistry

在化学研究仪器和设施-多用户仪器获取(CRIF)计划的支持下，华盛顿大学化学系将获得一个双处理器计算机服务器集群；一个文件服务器；用于交互使用的工作站；以及将安装在华盛顿大学科学并行计算中心所要求的硬件和机器上的一整套软件。该设备将加强若干领域的研究，包括a)并行实施贝叶斯概率理论用于核磁共振数据分析；b)溶胶-凝胶材料的多尺度模拟；c)固体核磁共振中的化学位移计算；d)支持分子半导体等结构掺杂实验的计算研究；e)分子电子电路元件的计算研究和支持表面化学分析；f)聚合物链结构和动力学的分子模拟，也支持实验研究。该设备将通过教授现代计算化学的理论和实践来加强化学系的本科生和研究生教育。使用具有全套化学软件的最先进的计算设施将鼓励研究生和本科生研究人员开发具有现代计算方法的专业知识，这将增强他们对化学概念的掌握，并对他们的实验研究计划做出协同贡献。其中两个主要用户将开发在核磁共振数据分析中非常有用的软件的集群优化并行版本。考虑到现代集群硬件的经济性，这些代码将得到广泛传播，并可能产生重大影响。最后，调查人员配置系统的经验以及所有讲习班和教程材料将在网上公布；这将帮助其他地点的化学家获得和使用类似的设施。

With support from the Chemistry Research Instrumentation and Facilities - Multiuser Instrument Acquisition (CRIF) Program, the Department of Chemistry at Washington University will acquire a cluster of dual-processor computer servers; a file server; workstations for interactive use; and a comprehensive suite of software to be installed both on the requested hardware plus on machines at the Washington University Center for Scientific Parallel Computing. This equipment will enhance research in a number of areas including a) parallel implementation of Bayesian probability theory applied to NMR data analysis; b) multi-scale modeling of sol-gel materials; c) chemical-shift computations in solid-state NMR; d) computational studies to support experiments on isostructural doping of molecular semiconductors; e) computational studies of molecular electronic circuit components and in support of surface chemical analyses; and f) molecular modeling of polymer chain structure and dynamics in the bulk and within domains of shell crosslinked nanoparticles, also in support of experimental studies. This equipment will enhance undergraduate and graduate education in the Chemistry Department through instruction in the theory and practice of modern computational chemistry. Access to a state-of-the-art computational facility with a complete range of chemistry software will encourage graduate and undergraduate researchers to develop expertise with modern computational methods, which will both enhance their mastery of chemical concepts and contribute synergistically to their experimental research programs. Two of the major users will develop cluster-optimized parallel versions of software of great utility in the analysis of NMR data. These codes will be broadly disseminated and could make major impacts, given the economy of modern cluster hardware. Lastly, the investigators' experiences configuring the system and all workshop and tutorial materials will be posted on the web; this will assist chemists at other sites in the acquisition and use of similar facilities.