课题基金 / 基金详情

MRI: Acquisition of High Speed Network Infrastructure for High Performance Distributed Computation in the Colleges of Science and Engineering

MRI: Acquisition of High Speed Network Infrastructure for High Performance Distributed Computation in the Colleges of Science and Engineering
MRI:科学与工程学院高性能分布式计算的高速网络基础设施的购置
批准号:
0922702
负责人:
Jose Castillo
金额:
$25.59万
依托单位国家:
美国
项目类别:
Standard Grant
财政年份:
2009
资助国家:
美国
项目状态:
已结题
起止时间:
2009-09-01 至 2013-02-28

项目摘要

项目成果

Jose Castillo的其他基金

相似基金

相关文献

中文摘要
翻译
该奖项将资助购买思科10千兆/秒交换机和波分复用设备,以在圣地亚哥州立大学(SDSU)计算科学研究中心(CSRC)附属的科学和工程研究小组之间构建三路高速数据链路,圣地亚哥超级计算机中心(SDSC)的TeraGrid中心,以及位于加州大学圣地亚哥分校(UCSD)的加州电信和信息技术研究所(CalIT 2)。 该网络基础设施将为SDSU的物理化学、天体物理学、天文学、生物信息学、热科学、燃烧工程、流体动力学、非线性动力学、海洋建模和计算机科学研究小组提供核心数据传输资源,这些研究小组经常使用SDSC和CalIT 2计算资源。 波分复用设备将使两个校园之间的快速网络交易,从而使SDSU的研究人员能够快速传输TeraGrid集群产生的数据到SDSU进行本地可视化和后处理。核心10千兆/秒交换机将首先连接SDSU两个独立学院的服务器机房设施:理学院和工程学院。 该交换机还将为各个研究实验室提供多个额外的10千兆/秒接口,以连接并从而在SDSU校园内形成高速研究骨干网络。 将立即利用该网络基础设施的研究项目包括:(1)开发用于模拟合成气燃烧中气体涡轮机NOx排放的化学平衡和动力学服务的面向服务的体系结构/网络基础设施,(2)开发通用Currency海洋模型(GCOM)的网络基础设施,(3)松软活性中间体的计算量子力学研究,(4)用于脉冲爆震发动机模拟的高阶粒子源单元(PSIC)方法的数值研究,以及(5)PetaShake的开发,PetaShake是TeraShake并行滞弹性波模型(AWM)代码的千万亿次实现,其使用数十亿个网格点模拟地震场景。 这些项目目前对现有的SDSU校园网络提出了很高的要求。 SDSU的研究人员现在需要更大的吞吐量和更短的延迟交换数据集时,这笔赠款将提供一个急需的改进SDSU-UCSD网络连接通过资助必要的设备,以促进两所大学之间的快速数据传输。 除了教师和研究生使用,该网络基础设施还将用于从事以下项目的本科生:(1)计算机科学和计算机工程的高级论文研究项目,(2)计算机网络,高速网络设计,客户端-服务器编程和分布式计算等课程中的编程项目,以及(3)在现有的,外部资助的奖学金计划。SDSU占据了一个强大的地位,使这些机会提供给少数民族在科学和工程研究中的代表性不足,在全国十大校园授予学士学位的少数民族之一。
英文摘要
This award will fund the purchase of a Cisco 10 Gigabit/s switch and wavelength division multiplexing equipment to construct a three-way high-speed data link between science and engineering research groups affiliated with the Computational Science Research Center (CSRC) at San Diego State University (SDSU), the TeraGrid hub at the San Diego Supercomputer Center (SDSC), and the California Institute for Telecommunications and Information Technology(CalIT2) at the University of California, San Diego (UCSD). This network infrastructure will provide core data transmission resources for research groups at SDSU in physical chemistry, geophysics, astronomy, bioinformatics, thermal science, combustion engineering, fluid dynamics, nonlinear dynamics, ocean modeling, and computer science who routinely use SDSC and CalIT2 computing resources. The wavelength division multiplexing equipment will enable rapid network transactions between the two campuses, thereby allowing researchers at SDSU to rapidly transfer terabytes of data generated on TeraGrid clusters back to SDSU for local visualization and post-processing.The core 10 Gigabit/s switch will initially interconnect server room facilities at SDSU in two separate colleges: the College of Science and the College of Engineering. The switch will also provide several additional 10 Gigabit/s interfaces for individual research laboratories to connect and thereby form a high-speed research backbone network on the SDSU campus. Research projects that will immediately make use of this network infrastructure include (1) the development of a service oriented architecture/cyberinfrastructure of chemical equilibrium and kinetic services used to model gas turbine NOx emissions in syngas combustion, (2) the development of a cyberinfrastructure for the General Curvilinear Ocean Model (GCOM),(3) computational quantum mechanical studies of floppy reactive intermediates, (4) the numerical investigation of high-order particle-source-in-cell (PSIC) methods for the simulation of pulse detonation engines, and (5) the development of PetaShake, a petascale implementation of the TeraShake parallel Anelastic Wave Model (AWM) code that simulates earthquake scenarios using several billion grid points. These projects currently place heavy demands on the existing SDSU campus network. SDSU researchers now require greater throughput and shorter delay when exchanging datasets and this grant will provide a much needed improvement to SDSU-UCSD network connectivity by funding the necessary equipment to facilitate rapid data transfer between the two universities. In addition to faculty and graduate student use, this network infrastructure will be used by undergraduate students working on projects that include (1) senior thesis research projects in computer science and computer engineering, (2) programming projects in courses such as computer networking, high-speed network design, client-server programming, and distributed computing, and (3) undergraduate research carried out under existing, externally funded fellowship programs. SDSU occupies a strong position for making these opportunities available to minorities underrepresented in science and engineering research, as one of the top ten campuses in the nation granting bachelor's degrees to ethnic minorities.
期刊论文(1)
专著(0)
科研奖励(0)
会议论文
Parallel Implementation of a PETSc-Based Framework for the General Curvilinear Coastal Ocean Model
基于 PETSc 的通用曲线沿海海洋模型框架的并行实现
DOI: 10.3390/jmse7060185
发表时间: 2019
期刊: Journal of Marine Science and Engineering
影响因子: 2.9
作者: [Valera, Manuel, Thomas, Mary P., Garcia, Mariangel, Castillo, Jose E.]
通讯作者: Castillo, Jose E.
CC* Compute: Central Computing with Advanced Implementation at San Diego State University
Broadening Opportunities for Interdisciplinary Computational Science and Engineering Research Training (ICSERT)
CC-NIE Network Infrastructure: Implementation of a Science DMZ at San Diego State University to Facilitate High-Performance Data Transfer for Scientific Applications
Acquisition of a Computational Mathematics Cluster
海外基金