MRI: Acquisition of Hardware for the Enhancement of the ELSA High Performance Computing Cluster to Enable Computational Research at The College of New Jersey

MRI：采购硬件以增强 ELSA 高性能计算集群，以支持新泽西学院的计算研究

• 批准号: 1828163
• 负责人: Joseph Baker
• 金额: $ 65.1万
• 依托单位: The College of New Jersey
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-01 至 2021-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1828163&HistoricalAwards=false
• 关键词: MRI; Acquisition; Hardware; Enhancement; ELSA

The College of New Jersey (TCNJ) will acquire equipment to significantly upgrade and enhance the Electronic Laboratory for Science and Analysis (ELSA) High Performance Computing cluster. TCNJ is a primarily undergraduate institution promoting a deep engagement of undergraduate students in research. Many of TCNJ's School of Science faculty members are working at the cutting edge of computational research in their fields, which include a broad range of areas including biochemistry/biophysics, genetics, bioinformatics, astrophysics, machine learning, and mathematical biology. In order to maintain a diverse and state of the art resource that meets the current and future computational needs of TCNJ's faculty and undergraduate students the current ELSA cluster requires targeted hardware enhancements. The new instrument will (1) enhance the research capacity and resulting scientific discovery of TCNJ's School of Science faculty members and their undergraduate research teams; (2) expose a greater number of undergraduate students and researchers to this powerful computational infrastructure through a series of newly developed High Performance Computing and data visualization short courses and workshops; and (3) improve access to the ELSA cluster for students traditionally underrepresented in STEM, as well as to researchers beyond TCNJ through a new collaboration with Open Science Grid.This project will expand the research programs of more than 13 faculty members (many of whom are early career faculty) spanning all five of TCNJ's School of Science departments. The computationally intensive work that will be supported through this project includes a diverse array of scientific efforts, including studies of pilus biomechanics, estimations of cell signaling processes, methods for investigating the strength of passwords and security of password systems, and improving our understanding of the most energetic objects in the universe. Currently, the research programs of TCNJ faculty members in these and other areas are restricted by inadequate graphic processing unit (GPU) resources and by the slow speed aging central processing unit (CPU) servers part of the current instrumentation. The new ELSA cluster will allow faculty and student researchers at TCNJ to run workflows ranging from embarrassingly parallel computations, to those that necessitate high levels of parallelization over hundreds of cores, intensive GPU computations, and remote visualization of simulation results. The instrument will thus ensure that these research programs are able to reach their full potential. This project will also benefit nearly 100 undergraduate student researchers each year who are part of these labs and work directly on the cluster. As a result, in addition to improving the capacity for scientific discovery, the proposed acquisition will help TCNJ meet the demands of developing an undergraduate workforce that is ready to leverage increasingly powerful High-Performance Computing resources, now and in their future careers.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

新泽西学院 (TCNJ) 将购买设备来显着升级和增强科学与分析电子实验室 (ELSA) 高性能计算集群。 TCNJ 是一所以本科生为主的机构，致力于促进本科生深入参与研究。 TCNJ 理学院的许多教员都在各自领域的计算研究前沿工作，这些领域涉及生物化学/生物物理学、遗传学、生物信息学、天体物理学、机器学习和数学生物学等广泛领域。为了保持多样化和最先进的资源，满足 TCNJ 教师和本科生当前和未来的计算需求，当前的 ELSA 集群需要有针对性的硬件增强。新仪器将 (1) 提高 TCNJ 理学院教职员工及其本科生研究团队的研究能力和由此产生的科学发现； （2）通过一系列新开发的高性能计算和数据可视化短期课程和研讨会，让更多的本科生和研究人员接触到这一强大的计算基础设施； (3) 通过与 Open Science Grid 的新合作，为传统上在 STEM 领域代表性不足的学生以及 TCNJ 以外的研究人员提供更多进入 ELSA 集群的机会。该项目将扩大超过 13 名教职人员的研究项目（其中许多人是是早期职业教师），涵盖 TCNJ 理学院的所有五个系。该项目将支持的计算密集型工作包括各种各样的科学工作，包括菌毛生物力学的研究、细胞信号传导过程的估计、研究密码强度和密码系统安全性的方法，以及提高我们对密码系统的理解。宇宙中最具能量的物体。目前，TCNJ 教员在这些和其他领域的研究项目受到图形处理单元 (GPU) 资源不足以及当前仪器中中央处理单元 (CPU) 服务器老化缓慢的限制。新的 ELSA 集群将允许 TCNJ 的教师和学生研究人员运行各种工作流程，从令人尴尬的并行计算到需要数百个核心的高水平并行化、密集 GPU 计算和模拟结果远程可视化的工作流程。因此，该工具将确保这些研究项目能够充分发挥其潜力。该项目每年还将惠及近 100 名本科生研究人员，他们是这些实验室的一部分，直接在集群上工作。因此，除了提高科学发现的能力外，拟议的收购还将帮助 TCNJ 满足培养本科生队伍的需求，使他们能够在现在和未来的职业生涯中利用日益强大的高性能计算资源。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

A Neural Network Approach for Birds of a Feather Solvability Prediction

物以类聚的神经网络方法可解性预测

• DOI: 10.1609/aaai.v33i01.33019706
• 发表时间: 2019
• 期刊: Proceedings of the ... AAAI Conference on Artificial Intelligence
• 作者: Sang, Benjamin;Yoon, Sejong
• 通讯作者: Yoon, Sejong

Theory of Change to Practice: How Experimentalist Teaching Enabled Faculty to Navigate the COVID-19 Disruption

• DOI: 10.1021/acs.jchemed.0c00731
• 发表时间: 2020-09-08
• 期刊: JOURNAL OF CHEMICAL EDUCATION
• 影响因子: 3
• 作者: Chan, Benny C.;Baker, Joseph L.;Triano, Rebecca M.
• 通讯作者: Triano, Rebecca M.

Unveiling molecular interactions that stabilize bacterial adhesion pili.

• DOI: 10.1016/j.bpj.2022.04.036
• 发表时间: 2022-06-07
• 期刊: BIOPHYSICAL JOURNAL
• 影响因子: 3.4
• 作者: Dahlberg, Tobias;Baker, Joseph L.;Bullitt, Esther;Andersson, Magnus
• 通讯作者: Andersson, Magnus

Stopping Active Learning Based on Predicted Change of F Measure for Text Classification

• DOI: 10.1109/icosc.2019.8665646
• 发表时间: 2019-01
• 期刊: 2019 IEEE 13th International Conference on Semantic Computing (ICSC)
• 作者: Michael Altschuler;Michael Bloodgood

Predicting Crowd Egress and Environment Relationships to Support Building Design Optimization

• DOI: 10.1016/j.cag.2020.03.005
• 发表时间: 2020-05
• 期刊: Comput. Graph.
• 作者: Kaidong Hu;Sejong Yoon;V. Pavlovic;P. Faloutsos;M. Kapadia