Survey of Recovery Schemes in MPLS Networks

Abstract

To provide dependable services MPLS networks make use of a set of procedures (detection, notiJcation and fault recovery) which seek to ensure appropriate protection for the trafic carried in the LSPs. When a fault happens i n the active LSP, the recovery scheme must re-direct the trafic to a recovery path (the protection LSP or recovery LSP) which bypasses the fault. The two basic recovery models used to redirect trafic are rerouting and protection switching. Protection switching is faster than rerouting but cannot handle simultaneous faults in the active and the recovery path. On the other hand, rerouting is generally slow, and cannot offer QoS guaranties upon failure, but can use resources in a more eficient way. In this work we present a survey of several fault recovery schemes, firstly some approaches for protection switching will be reviewed and secondly the characteristics of some rerouting approaches will be discussed. For these schemes, we also allude to the extensions required by the signalling and trafic engineering routing protocols. Finally, the reviewed recovery mechanisms are classified according to a set of characteristics considered relevant.

A Survey of Fast-Recovery Mechanisms in Packet-Switched Networks

“…With our focus on packet-switched networks we hence explicitly do not consider recovery in, e.g., optical networks for which there exists much equipment for performing protection. Furthermore, there already are many good surveys on reliable routing in specific communication technologies, such as Ethernet [34], IP [23], [25], [35], [36] (and more recently [23], [25]), MPLS [37], [38], BGP [39], or SDN [40]. However, to the best of our knowledge, this is the first complete and up-to-date survey on the timely topic of fastrecovery mechanisms in packet-switched networks.…”

A Survey of Fast Recovery Mechanisms in the Data Plane

“…There already exist many good surveys on reliable routing in specific communication technologies, such as Ethernet [34], IP [23], [25], [35], [36] (and more recently [23], [25]), MPLS [37], [38], BGP [39], or SDN [40]. However, to the best of our knowledge, this is the first complete and up-todate tutorial-like survey on the timely topic of fast-recovery mechanisms in the data plane.…”

Fast reroute with pre-established bypass tunnel in MPLS

“…In Efficient Pre-Qualify [15], a backup path for each LSP is computed when the network states change, but without actual establishing the backup path. When an LSR detects a failure, the LSR establishes the previously calculated backup path, and reroutes the traffic to the backup path.…”

“…The recently presented methods for restoration of label switched paths (LSP) are fast reroute extensions to RSVP-TE for LSP [8], Efficient Pre-Qualify [14,15], distributed label switched path (D-LSP) [16], simple pre-provisioning scheme [17] and the scheme of Hundessa [18,19] and Das [20].…”

“…Paths must also be rerouted quickly so that only a few packets are lost. Based on the above two criteria, this study presents the pre-established bypass tunnels (PBT) method, which exhibits good performance in terms of correctly received ratio (number of packets received without link failure divided by number of packets received with link failure) and successful rerouting ratio (number of successfully rerouted LSPs divided by number of affected LSP when a link fails or a LSR fails) when compared with Extension RSVP [8] and Efficient Pre-Qualify [14,15]. Section 2 introduces the recently proposed mechanisms.…”