Collaborative Research: Framework Implementations: CSSI: CANDY: Cyberinfrastructure for Accelerating Innovation in Network Dynamics

合作研究：框架实施：CSSI：CANDY：加速网络动态创新的网络基础设施

• 批准号: 2104076
• 负责人: Sanjukta Bhowmick
• 金额: $ 44.94万
• 依托单位: University of North Texas
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2025-08-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2104076&HistoricalAwards=false
• 关键词: Collaborative; Research; Framework; Implementations; CSSI

Efficient analysis of dynamic networks is highly important in diverse multidisciplinary real-life applications, such as data mining and analytics, social and biological networks, epidemiology, cyber-physical infrastructures, transportation networks, surface mining, and cybersecurity. Although numerous software exists for analyzing static networks, a comprehensive cyberinfrastructure that supports innovative research challenges in large-scale, complex, dynamic networks is lacking. This multi-university proposal addresses this gap by developing a novel platform, called CANDY (Cyberinfrastructure for Accelerating Innovation in Network Dynamics), based on efficient, scalable parallel algorithm design for dynamic networks and high-performance software development with performance optimization. For broader impact and outreach activities, the investigators will (1) collaborate with multidisciplinary research groups to evaluate the effectiveness of the developed platform, algorithms and software tools; (2) host workshops, webinars, and tutorials to educate research community about the cyberinfrastructure; (3) disseminate project outcomes via a dedicated website, keynote and invited talks, demos, and high-quality publications in peer-reviewed journals and conferences; and (4) train next generation data scientists in the development of CANDY platform, by engaging women and underrepresented minority students, including high school students and rural communities in Missouri, Hispanic and African-American communities in Texas, and First Nation (Native American) community in Oregon.This project will develop the first parallel, scalable, extendable, and user-friendly software platform for updating important properties of dynamic networks. It will also provide the requisite functionalities and tools to modify existing algorithms or create new ones, catering to basic, intermediate and advanced users with different levels of expertise. The CANDY cyberinfrastructure platform will be implemented on different architectures, such as distributed memory, shared memory, and graphics processor units providing user-friendly interfaces. Significant research and development innovations include: (1) a novel hierarchical taxonomy of network analysis algorithms that allows for layered specification of parallel algorithms based on multiple parameters; (2) templates for creating new scalable algorithms for dynamic network analysis; (3) algorithms to partition the streaming set of nodes and edges into network snapshots at changing points; and (4) invariant-based quantifiable performance metrics for analyzing large-scale dynamic networks. As a case study, the developed software will be evaluated on two disparate domains -- fast processing of genomic data on dynamic trees, and cost-effective operation of complex mining engineering applications.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.

动态网络的有效分析在不同的多学科现实生活应用中非常重要，例如数据挖掘和分析，社会和生物网络，流行病学，网络物理基础设施，运输网络，表面挖掘和网络安全。尽管存在许多用于分析静态网络的软件，但是缺乏支持大规模，复杂，动态网络中创新研究挑战的全面网络基础结构。该多元大学提案通过开发一个名为Candy（用于加速网络动力学加速创新的网络基础结构）的新型平台，基于有效的，可扩展的并行算法设计，用于动态网络和高性能软件开发具有性能优化。对于更广泛的影响和外展活动，研究人员将（1）与多学科研究小组合作，以评估开发平台，算法和软件工具的有效性； （2）主持研讨会，网络研讨会和教程，以对研究社区进行有关网络基础设施的教育； （3）通过专门的网站，主题演讲和受邀演讲，演示以及高质量评论的期刊和会议的高质量出版物来传播项目成果； （4）通过吸引妇女和代表性不足的少数族裔学生，包括高中生和德克萨斯州的西班牙裔和非裔美国人社区，以及俄勒冈州原住民（American）社区，这些项目将开发出一个可扩展的，可扩展和用户友好的典型范围，培训了下一代数据科学家。它还将提供必要的功能和工具来修改现有算法或创建新算法，以适应具有不同专业知识水平的基本，中级和高级用户。 Candy Cyber​​infrastructure平台将在不同的体系结构（例如分布式内存，共享内存和图形处理器单元）上实现，可提供用户友好的接口。重大的研发创新包括：（1）网络分析算法的新型层次分类法，该算法允许基于多个参数对平行算法进行分层规范； （2）用于创建用于动态网络分析的新的可扩展算法的模板； （3）算法将节点和边缘的流集划分为变化点的网络快照； （4）基于不变的可量化性能指标，用于分析大规模动态网络。作为一个案例研究，将对开发的软件进行评估，对两个不同的领域进行评估 - 快速处理动态树的基因组数据以及复杂的采矿工程应用程序的成本效益操作。该奖项反映了NSF的法定任务，并被认为是值得通过基金会的知识分子和更广泛影响的审查标准来通过评估来通过评估来获得支持的。

项目成果

Parallel Vertex Color Update on Large Dynamic Networks

• DOI: 10.1109/hipc56025.2022.00027
• 发表时间: 2022-12
• 期刊: 2022 IEEE 29th International Conference on High Performance Computing, Data, and Analytics (HiPC)
• 作者: Arindam Khanda;S. Bhowmick;Xin Liang;Sajal K. Das