Complex networks as control paradigm for complex systems: towards 'evolve-able' architectures for resilient enetworked applications

复杂网络作为复杂系统的控制范例：面向弹性网络应用的“可进化”架构

• 批准号: 217201-2008
• 负责人: Ulieru, Mihaela
• 金额: $ 1.21万
• 依托单位: University of New Brunswick
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2008
• 资助国家: 加拿大
• 起止时间: 2008-01-01 至 2009-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=393161
• 关键词: Complex; networks; control; paradigm; complex

The Internet of the future will be a nervous system for the entire economy, integrating 'opportunistic ecosystems' of single devices / departments / enterprises into a larger and more complex infrastructure called the Cyber-Physical Ecosystem (CPE). In the CPE, the individual properties or attributes of single entities will be dynamically combined to achieve an emergent desired behavior of the ecosystem. It is extremely hard - if not impossible - to control large scale CPE by building a global logic 'top-down' system able to rapidly adapt to changes by instructing each element what to do at each step. Using the latest knowledge of complexity science, we aim through this proposal to develop a methodological framework for designing for 'evolve-ability' large scale CPE capable of generating resilient and scalable structure from the 'bottom-up' by evolving self-organized basic architectural component 'cells.' These cells will be adaptively crafted through dynamic protocols enabling service composition into novel architectural components. The statistical properties displayed by the underlying network structure of the complex distributed system will reveal the appropriate parameters on which efficient reliable operation depends. The parameters will be tuned using an eNetwork middleware embedded into the complex system's fabric similar to how DNA molds the fundamental cells in natural systems such that they can evolve to accommodate gradual or abrupt change in the environment or internal operating conditions. Validation on the state-of-the-art testbed recently deployed in the Adaptive Risk Management Lab led by the applicant at UNB will enable proof of concept as well as applications that will revolutionize several areas of crucial importance, including: blackout-free electricity generation and distribution, optimization of energy consumption, disaster response through deployment of holistic security ecosystems, pandemic mitigation, networked transportation and manufacturing, and environmental monitoring and sustainability assessment.

未来的互联网将是整个经济的神经系统，将单个设备/部门/企业的“机会主义生态系统”整合到一个更大、更复杂的基础设施中，称为网络物理生态系统（CPE）。在CPE中，单个实体的各个属性或属性将被动态地组合以实现生态系统的紧急期望行为。通过构建一个全局逻辑“自上而下”的系统来控制大规模CPE是非常困难的-如果不是不可能的话-该系统能够通过指示每个元素在每个步骤中做什么来快速适应变化。利用复杂性科学的最新知识，我们的目标是通过这个建议，开发一个方法框架，设计的“进化能力”的大规模CPE能够产生弹性和可扩展的结构，从“自下而上”的发展自组织的基本架构组件的细胞。这些单元将通过动态协议自适应地制作，使服务组合成为新的架构组件。复杂分布式系统的底层网络结构所显示的统计特性将揭示有效可靠运行所依赖的适当参数。这些参数将使用嵌入到复杂系统结构中的eNetwork中间件进行调整，类似于DNA如何塑造自然系统中的基本细胞，以便它们可以进化以适应环境或内部操作条件的逐渐或突然变化。最近在UNB申请人领导的自适应风险管理实验室中部署的最先进的测试平台上进行的验证将实现概念验证以及应用程序，这些应用程序将彻底改变几个至关重要的领域，包括：无停电发电和配电，优化能源消耗，通过部署整体安全生态系统应对灾害，缓解流行病，联网运输和制造，以及环境监测和可持续性评估。