Backup paths are widely used in IP networks to protect IP links from failures. However, existing solutions such asthe commonly used independent model and Shared Risk Link Group (SRLG) model do not accurately reflect the correlation betweenIP link failures, and thus may not choose reliable backup paths. We propose a cross-layer approach for minimizing routing disruptioncaused by IP link failures. We develop a probabilistically correlated failure (PCF) model to quantify the impact of IP link failure onthe reliability of backup paths. With the PCF model, we propose an algorithm to choose multiple reliable backup paths to protecteach IP link. When an IP link fails, its traffic is split onto multiple backup paths to ensure that the rerouted traffic load on each IP linkdoes not exceed the usable bandwidth. We evaluate our approach using real ISP networks with both optical and IP layer topologies.Experimental results show that two backup paths are adequate for protecting a logical link. Compared with existing works, the backuppaths selected by our approach are at least 18 percent more reliable and the routing disruption is reduced by at least 22 percent.Unlike prior works, the proposed approach prevents the rerouted traffic from interfering with normal traffic. Cross-Layer Approach for Minimizing Routing Disruption in IP Networks
IPlink failures are fairly common in the Internet for variousreasons. In high speed IP networks like the Internetbackbone, disconnection of a link for several seconds canlead to millions of packets being dropped . Therefore,quickly recovering from IP link failures is important forenhancing Internet reliability and availability, and hasreceived much attention in recent years. Currently, backuppath-based protection , and is widely used byInternet Service Providers (ISPs) to protect their domains. Inthis approach, backup paths are precomputed, configured,and stored in routers. When a link failure is detected, trafficoriginally traversing the link is immediately switched to thebackup path of this link. Through this, the routing disruptionduration is reduced to the failure detection time which istypically less than 50 ms.
Selecting backup paths is a critical problem in backuppath-based protection. Existing approaches mainly focuson choosing reliable backup paths to reduce the routingdisruption caused by IP link failures. However, they sufferfrom two limitations. First, the widely used failure modelsdo not accurately reflect the correlation between IP linkfailures. As a result, the selected backup paths may beunreliable. Second, most prior works consider backup path selection as a connectivity problem, but ignore the trafficload and bandwidth constraint of IP links