CC*DNI Networking Infrastructure: Data Driven Network Infrastructure Upgrade for Lamar University Research

拉马尔大学正在推出一个科学DMZ，旨在高效地为关键的STEM科学研究项目实现更高级别的性能、可靠性和可预测性，并满足下一代大数据研究需求。拟议的网络基础设施提供与州和国家高性能计算设施的连接，包括通过NSF/XSEDE计划支持的设施，从校园内的多个研究地点。拉马尔大学的教师研究包括但不限于生物、化学、环境科学、电子系统、生物医学诊断学、自然灾害、工程学以及高性能计算和数据分析。专门的科学DMZ通过10G光纤主干将校园内的研究密集区相互连接，同时消除研究实验室与外部协作计算和分析设施之间高效数据流的障碍，从而为研究和教育项目提供变革性的能力。新设计提供多个10Gbps路由端口、用于链路整合和聚合的DMZ交换机、高性能数据传输节点以及用于性能监控和测试的Perfsonar节点。此外，光纤网络升级为单模光纤连接，用于研究密集型领域，以及分布和接入层交换，以提供10G容量。新网络还通过将项目设计和操作融入课堂教学、学生参与密集的计算和数据驱动的研究以及独立的学生研究工作，为研究生和本科生提供更广泛的影响。

Lamar University is introducing a Science DMZ designed to efficiently achieve higher levels of performance, reliability and predictability for critical STEM science research projects and address next generation big data research needs. The proposed cyber infrastructure provides connectivity to state and national high performance computing facilities, including those supported through the NSF/XSEDE program, from multiple research locations across campus. Lamar University faculty research includes, but is not limited to, biology, chemistry, environmental science, electronic systems, biomedical diagnostics, natural disasters, engineering, and high performance computing and data analysis. The dedicated Science DMZ provides transformative capability for both research and educational programs by interconnecting research intensive areas on campus to one another through a 10G fiber backbone while removing obstacles to efficient data flows between research laboratories and external collaborative computational and analytical facilities. The new design provides multiple 10Gbps-routed ports, DMZ switches for link consolidation and aggregation, a high performance data transfer node, and a Perfsonar node for performance monitoring and testing. In addition, the optical fiber network is upgraded to single-mode fiber connections for research-intensive areas along with distribution and access layer switching to provide 10G capacity.The new network also provides broader impact benefits to graduate and undergraduate students by incorporating project design and operation into classroom lecture, student engagement in intensive computational and data driven research and independent student research efforts.