MRI: Acquisition of a High Performance Big Data Analysis Platform

MRI：收购高性能大数据分析平台

• 批准号: 1828521
• 负责人: Nihat Altiparmak
• 金额: $ 47.87万
• 依托单位: University of Louisville Research Foundation Inc
• 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: 2018-10-01 至 2020-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1828521&HistoricalAwards=false
• 关键词: MRI; Acquisition; Performance; Big; Data

This project from the University of Louisville Research Foundation will support acquisition of a Big Data Analysis Platform (BDAP) with high performance data storage, networking, and processing capabilities. The platform will support research in multimedia, biomedicine, metagenomics, and health data analysis. In addition, the platform will allow general research on efficient management and analysis of big data. Moreover, strengthening scientific research across the university, the researchers will engage with K-12 students and teachers.The project will advance the state-of-the-art in big data management and analysis by enabling five research thrusts that will use the BDAP equipment purchased by this award. First, BDAP will provide improvement over existing big data management techniques by introducing self-optimizing and energy efficient big data platforms through dynamic data placement, retrieval, and reorganization algorithms, as well as enabling efficient big data analysis through novel heterogeneous and multi-source data clustering algorithms. Second, BDAP will allow experimentation with novel algorithms guided by deep neural networks to analyze big multimedia data. Third, BDAP will enable experimentation with new paradigms for integrating big biomedical data from multiple sources including image, genomic, quantitative, biological, and observational data. Fourth, BDAP will enable testing of techniques to generate, store, analyze and integrate large microbiome, health and socioeconomic data sets to determine the causal relationship between specific microbial profiles and human health via bioinformatics and data mining approaches. Finally, the project will contribute to the development of new statistical methods for analyzing high dimensional data for epigenetic, pharmacogenomics, and genome association studies. This in-house instrument enables energy consumption measurements and permanent storage for sensitive data sets of petabyte-range. While advancing engineering knowledge, the instrumentation supports computational, statistical, and bioengineering research, that applies engineering principles to several problems in biology and medicine.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.

路易斯维尔大学研究基金会的该项目将支持获得具有高性能数据存储，网络和处理能力的大数据分析平台（BDAP）。 该平台将支持多媒体，生物医学，宏基因组学和健康数据分析的研究。 此外，该平台将允许对大数据有效管理和分析的一般研究。 此外，加强大学的科学研究，研究人员将与K-12学生和老师互动。该项目将通过启用五个将使用该奖项购买的BDAP设备的研究推力来推进大数据管理和分析的最新技术。首先，BDAP将通过动态数据放置，检索和重组算法以及通过新颖的异质和多组分数据集群算法来实现有效的大数据分析，从而通过动态数据放置，检索和重组算法引入自我优化和能源有效的大数据平台，从而提供对现有大数据管理技术的改进。其次，BDAP将允许通过深层神经网络引导的新算法进行实验，以分析大型多媒体数据。第三，BDAP将通过新的范式进行实验，以整合来自多个来源的大型生物医学数据，包括图像，基因组，定量，生物学和观察数据。第四，BDAP将能够测试技术，以生成，存储，分析和整合大型微生物组，健康和社会经济数据集，以通过生物信息学和数据挖掘方法来确定特定微生物概况与人类健康之间的因果关系。 最后，该项目将有助于开发新的统计方法，以分析用于表观遗传学，药物基因组学和基因组关联研究的高维数据。 该内部仪器可实现敏感数据集的能源消耗测量和永久存储。 在促进工程知识的同时，该仪器支持计算，统计和生物工程研究，该研究将工程原理应用于生物学和医学的几个问题。该奖项反映了NSF的法定任务，并被认为是通过基金会的知识分子优点和更广泛的影响审查标准来通过评估来进行评估的。

项目成果

Monte Carlo Based Server Consolidation for Energy Efficient Cloud Data Centers

• DOI: 10.1109/cloudcom.2019.00046
• 发表时间: 2019-12
• 期刊: 2019 IEEE International Conference on Cloud Computing Technology and Science (CloudCom)
• 作者: B. Harris;Nihat Altiparmak

Ultra-Low Latency SSDs' Impact on Overall Energy Efficiency

• 发表时间: 2020
• 作者: B. Harris;Nihat Altiparmak