A System for Managing Transport-network Recovery Using Hybrid-backup Operation Planes according to Degree of Network Failure

Suzuki, Toshiaki; Kubo, Hiroyuki; Hoshihara, Hayato; Sakamoto, Kenichi; Inouchi, Hidenori; Ogawa, Taro

Home // International Journal On Advances in Networks and Services, volume 9, numbers 3 and 4, 2016 // View article

A System for Managing Transport-network Recovery Using Hybrid-backup Operation Planes according to Degree of Network Failure

Authors:
Toshiaki Suzuki
Hiroyuki Kubo
Hayato Hoshihara
Kenichi Sakamoto
Hidenori Inouchi
Taro Ogawa

Keywords: network management; protection; disaster recovery; packet transport

Abstract:
A system for managing transport-network recovery using hybrid-backup operation planes according to the degree of a network failure is proposed. Under this management system, an entire network is separated into multiple areas. A network-management server prepares a three-step recovery procedure to cover the degree of network failure. In the first step of the recovery, an inside-area protection scheme is used to recover current data-transmission paths in each area. In the second step, an end-to-end protection scheme is applied to the current data-transmission paths. In the third step, the operation plane is changed. Each assumed operation plane is composed of recovery configurations for restoring failure paths for assumed area-based network failures. If a small network failure occurs, it is recovered by the inside-area protection and end-to-end protection schemes. If a catastrophic network failure (caused by a disaster) that cannot be recovered by those protection schemes occurs, it is recovered by changing the operation plane in accordance with the damaged areas. A prototype system composed of a network-management server and 96 emulated packet-transport nodes was developed and evaluated by configuring 1000 data-transmission paths. In case of a small network failure, 500 data-transmission paths were damaged, and they were reconfigured by the inside-area protection scheme and end-to-end protection scheme in about 5 seconds. If the network failure was not recovered by those protection schemes, 1000 data-transmission paths were reconfigured in about 1.2 seconds after the network-management server decided to change the operation plane. As a result, the proposed system could localize a network failure and recover a transport network according to the degree of network failures.

Pages: 38 to 48

Publication date: December 31, 2016

Published in: journal

ISSN: 1942-2644