Survivable Design of Multi-Domain Multi-Layer Backbone Networks

多域多层骨干网络的生存设计

• 批准号: 341465-2012
• 负责人: Wu, Jing
• 金额: $ 1.53万
• 依托单位: University of Ottawa
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=572265
• 关键词: Survivable; Design; Multi; Domain; Layer

Reliable communication networks and services are critical to modern societies. Our focus is to assess and improve reliability of backbone networks carrying aggregated user communication traffic within a metropolitan area, or across a region or country, or among continents. Currently, backbone networks are technologically organized in a hierarchy, which consists of multiple layers. At an upper layer, Internet Protocol (IP)-based user traffic is dominant, carried by data packets. Many IP services require certain quality of service, including service reliability, and thus are carried as IP flows, as opposed to individual packets. IP links in the upper layer are implemented in lower layers as reconfigurable circuits, such as lightpaths in Wavelength Division Multiplexing (WDM) networks. Protections to IP flows can be accomplished by providing backup IP paths at the upper layer, or providing backup lightpaths at lower layers, or a combination of both. Although single layer network and service reliability has been extensively studies for two decades, in a multi-layer network environment, a lot of important problems remain unsolved, for example, how to quantify the availability of IP flows when IP-link failures are not independent as a result of complicated mappings of IP links to lower-layer circuits, how to find the locations and then isolate network faults or attacks, how to protect against multiple network failures or attacks, etc. Independent of technological hierarchy, backbone networks also are administered as multiple administration domains or by different network operators. For end users, it is end-to-end service reliability that matters, not the reliability of network links or segments. Therefore, design and coordination of protections across multiple domains are important, however, also very challenging due to limited information sharing across domains and sometime competitive relations among network operators. This proposed project will investigate the design issues of service protection in a multi-domain multi-layer network environment.

可靠的通信网络和服务对于现代社会至关重要。我们的重点是评估和提高在大都市区、跨地区、国家或各大洲之间承载聚合用户通信流量的骨干网络的可靠性。目前，骨干网络在技术上是按层次结构组织的，该层次结构由多个层组成。在上层，基于互联网协议（IP）的用户流量占主导地位，由数据包承载。许多IP服务需要一定的服务质量，包括服务可靠性，因此作为IP流而不是单独的数据包来传送。上层中的 IP 链路在下层中实现为可重新配置电路，例如波分复用 (WDM) 网络中的光路。对IP流的保护可以通过在上层提供备份IP路径、或在下层提供备份光路、或两者的组合来实现。尽管单层网络和服务可靠性已经被广泛研究了二十年，但在多层网络环境中，许多重要问题仍未解决，例如，当IP链路故障不独立时如何量化IP流的可用性由于IP链路到下层电路的复杂映射，如何找到位置并隔离网络故障或攻击，如何防范多重网络故障或攻击等。骨干网也独立于技术层次进行管理作为多个管理域或由不同的网络运营商。对于最终用户而言，重要的是端到端服务的可靠性，而不是网络链路或网段的可靠性。因此，跨多个域的保护设计和协调很重要，但由于跨域的信息共享有限以及有时网络运营商之间的竞争关系，也非常具有挑战性。该项目将研究多域多层网络环境中的服务保护设计问题。