Carrier-grade ethernet:etherpath protection vs. ethertunnel protection

Ethernet is a success story in local area networks. Efforts to extend its boundaries beyond LAN to the carriers´ backbone networks are in progress. We study the problem of designing reliable and cost-efficient high-rate (100 Gb/s) carrier-grade Ethernet. Reliability is considered for application at one of two levels: Etherpath protection (where an Etherpath is a lightpath established by Ethernet interfaces) or Ethertunnel protection (where an Ethertunnel is an Ethernet connection). We develop the protection algorithms for the two cases, and study the trade-offs in their impact on network cost and implementation complexity. We consider practical settings for the envisioned next-generation carrier-grade Ethernet networks with the following characteristics: high-rate (up to 100 Gb/s) Etherpaths constrained by the signal´s maximum all-optical transmission range before which 3R signal regeneration is required, and an Ethernet network with mixed line rates. Our algorithms, tested on a 17-node German study network, lead to the following findings: ETP is more resource-efficient than EPP, but EPP incurs less implementation complexity than ETP.