CSR: Small: Cascading Failures Modeling and Mitigation in the Internet of Things

当单一事件引发连锁反应中系统组件的连续故障时，就会发生连锁故障。如果没有及时有效的缓解机制，连锁故障往往会对整个系统甚至环境和社会造成灾难性的影响。物联网（IoT）系统中现有的级联故障理论与现实之间存在很大差距。该项目旨在物联网级联故障和可靠性研究方面取得突破，弥合差距，减轻故障造成的严重后果。此外，该项目还包括教育部分，其目标是帮助增加代表性不足的少数民族，特别是妇女参与计算机和工程，并通过新课程的编制和对研究生和本科生的研究培训，将研究与教育结合起来。 该项目的总体目标是通过开发有效的级联故障建模和分析方法以及有效的缓解策略来增强物联网系统对级联故障的恢复能力，从而提高基于物联网的系统的可靠性和鲁棒性。具体而言，该项目将（1）开发新的综合数学模型，以表征物联网中级联故障的复杂物理机制;（2）创建高效和可扩展的可靠性评估方法，以提供准确和快速的物联网系统可靠性预测，考虑级联故障和各种依赖关系的影响;（3）开发创新的过程中和多阶段缓解机制，以有效应对连锁故障，并预防或至少减轻其后果;以及（4）使用真实世界物联网系统的案例研究以及独立同行和专家评估的出版物来评估和验证拟议的研究。提出的级联故障模型和缓解策略以及可靠性分析方法是对现代和不断发展的物联网系统的可靠性和鲁棒性设计和操作知识体系的基本贡献。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Cascading failures take place when a single incident triggers successive malfunction of system components in a chain reaction. Without timely and effective mitigation mechanisms, cascading failures often cause catastrophic impacts on the entire system and even the environment and society. There is a big gap between the existing cascading failure theories and realities in the Internet of Things (IoT) systems. This project aims to make breakthroughs in the IoT cascading failure and reliability research to bridge the gap, mitigating serious consequences caused by the failures. In addition, the project has educational components with the goals to help increase participation of underrepresented minorities, particularly women in computing and engineering, and to integrate research and education through new curriculum development and research training of both graduate and undergraduate students. The overarching objective of this project is to enhance the reliability and robustness of IoT-based systems by developing efficient cascading failure modeling and analysis methods as well as effective mitigation strategies to strengthen IoT systems’ resilience to cascading failures. Specifically, the project will (1) develop new comprehensive mathematical models to characterize the complex physical mechanism of cascading failures in IoT; (2) create efficient and scalable reliability evaluation methodologies to provide an accurate and fast prediction of the IoT system reliability considering impacts of cascading failures and various dependencies; (3) develop innovative in-process and multi-phase mitigation mechanisms to effectively respond to cascading failures and prevent or at least alleviate their consequence; and (4) evaluate and validate the proposed research using case studies of real-world IoT-based systems as well as publications evaluated by independent peers and experts. The proposed cascading failure models and mitigation strategies, and reliability analysis methods are fundamental contributions to the body of knowledge on the reliable and robust design and operation of modern and evolving IoT-based systems.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.