Acquisition of a Computer Cluster for the Center of Applied Mathematics and Statistics at NJIT

新泽西理工学院应用数学和统计中心购置计算机集群

• 批准号: 0420590
• 负责人: Daljit Ahluwalia
• 金额: $ 27.09万
• 依托单位: New Jersey Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-09-01 至 2008-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0420590&HistoricalAwards=false
• 关键词: Acquisition; Computer; Cluster; Center; Applied

The Department of Mathematical Sciences (DMS) and Center for Applied Mathematics and Statistics (CAMS) at the New Jersey Institute of Technology, under the auspices of the MRI program, acquire a 64-node Beowulf-class computer cluster for research in the mathematical sciences. Each compute-node of the cluster is composed of two 2.0 GHz, 64-bit processors (AMD Opteron CPU); the cluster includes a total of 256 GB of memory, mass storage devices, scientific software, and hardware for a high speed Myrinet network. The machine is dedicated to the support of research by faculty and graduate students in CAMS and DMS, and is used for projects which involve mathematical modeling and the development of computational techniques to study fundamental processes in physical science and biology. Examples of these projects include: the development of efficient molecular dynamics methods with applications to fluid flow in nano-devices and drug molecule/protein target interactions; simulations of large interacting systems of neurons in the visual cortex; investigations of granular systems; studies of mesoscopic models for surface processes in biology; simulations of surface evolution in crystalline materials; and improved numerical methods for studying aspects of electromagnetic wave propagation.The research activities are primarily involved with the mathematical modeling of important processes in science and technology and hence are of benefit to scientists and engineers in a wide variety of disciplines. For example, the research on molecular dynamics methods is used to obtain insights in the interactions between drug molecules and their protein targets, numerical simulations of interacting neurons in the visual cortex can lead to an improved understanding of high-level visual processing events, such as "edge-detection," and studies of surface evolution in crystalline materials aid in the design of novel microelectronic devices. In addition, the described research promotes interdisciplinary collaborations between applied mathematicians and scientists in diverse areas. Graduate students and postdocs involved in the research receive training in state-of-the-art numerical techniques.

在核磁共振计划的支持下，新泽西理工学院数学科学系(DMS)和应用数学与统计中心(CAMS)获得了一个用于数学科学研究的节点Beowulf级计算机集群。集群的每个计算节点由两个2.0 GHz的位处理器(AMD Opteron CPU)组成；集群总共包括256 GB的内存、海量存储设备、科学软件和用于高速Myrnet网络的硬件。这台机器致力于支持CAMS和DMS的教职员工和研究生的研究，并用于涉及数学建模和计算技术开发的项目，以研究物理科学和生物学的基本过程。这些项目的例子包括：开发高效的分子动力学方法，应用于纳米设备中的流体流动和药物分子/蛋白质靶标相互作用；模拟视觉皮质中的大型神经元相互作用系统；颗粒系统的调查；生物学表面过程的介观模型研究；晶体材料表面演化的模拟；以及研究电磁波传播方面的改进的数值方法。研究活动主要涉及科学和技术中重要过程的数学建模，因此对各种学科的科学家和工程师都有好处。例如，分子动力学方法的研究被用来了解药物分子与其蛋白质靶标之间的相互作用，对视觉皮质中相互作用的神经元的数值模拟可以提高对高级视觉处理事件的理解，如“边缘检测”，以及对晶体材料表面演化的研究有助于设计新型微电子设备。此外，所描述的研究促进了应用数学家和科学家在不同领域的跨学科合作。参与这项研究的研究生和博士后都接受了最先进的数值技术培训。