• Convergence Security Journal
• /
• v.23 no.1
• /
• pp.3-8
• /
• 2023

Multi-domain optical transport networks are not fundamentally interoperable and require an integrated orchestration mechanism and path provision mechanism at the entire network level. In addition, ensuring network survivability is one of the important issues. MPLS-TP (Multi-Protocol Label Switching-Transport Profile) defines various protection/recovery methods as standards, but does not mention how to calculate and select protection/recovery paths. Therefore, an algorithm that minimizes protection/recovery collisions at the optical circuit packet integrated network level and calculates and sets a path that can be rapidly protected/recovered over the entire integrated network area is required. In this paper, we proposed an algorithm that calculates and sets up a path that can be rapidly protected and restored in a T-SDN network composed of multiple ring-mesh topology.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.12B
• /
• pp.1051-1059
• /
• 2003

The key feature of GMPLS is the provision of network resources and the automatic set-up of a path. And the mechanism of protection and restoration of a path is presented when network component fails. This paper suggests CR-LDP Static Detour Mechanism and Dynamic Detour Mechanism. CR-LDP Detours are a mechanism to set up detour paths automatically at every node to protect working path. Hence suggested mechanism performance is compared with existing mechanisms by computer simulation.

• The KIPS Transactions:PartC
• /
• v.12C no.3 s.99
• /
• pp.401-408
• /
• 2005

We propose a new MPLS restoration scheme that uses network resources more efficiently and minimizes the backup path cost effectively. Contrary to other restoration strategies, the proposed restoration scheme starts the recovery action at the selected node of a working LSP. At LSP setup, the working and backup path cost is evaluated and the starting node of restoration is designated. By doing so, the restoration speed could be further increased and resource utilization could be maximized. We simulated the proposed scheme and compared with other restoration and protection schemes. The result shows that our scheme can provide fast restoration with acceptable delay and loss characteristics.

• The KIPS Transactions:PartC
• /
• v.10C no.2
• /
• pp.155-162
• /
• 2003

Fast restoration time and service guarantee are the important goals to achieve the network reliability. In the protection switching scheme, one way to guarantee service fro an application session if a network happens to fail is to establish the restoration path that amounts to the same bandwidth of the working path of the session at the same time. When we setup the restoration path, we can reduce the bandwidth consumption by the restoration path if the path can share the bandwidth required by the other paths. This paper explains the methods how to determine the shared bandwidth of the restoration path in the protection switching scheme, given the maximum bandwidth assigned to a link along the working path. We point out that such sharing algorithm can not reduce the bandwidth consumption by the restoration paths in some cases, which contradict the general conception. We explain why this can happen, and show the simulation results in real network topologies to prove our arguments. We explain the reason of the failure of the sharing effect by the simple sharing algorithm. Finally we propose the way of how we can overcome the failure of the sharing effect, using the complete sharing algorithm based on the link database and showing the results.

• The KIPS Transactions:PartC
• /
• v.12C no.1 s.97
• /
• pp.77-82
• /
• 2005

Protection of user service becomes increasingly important since even very short interruption of service due to link or node failure will cause huge data loss and incur tremendous restoration cost in high speed network environment. Thus fast and efficient protection and restoration is one of the most important issues to be addressed. Protection methods have been proposed to provide efficiency and stability in optical networks. In this paper, an original network is separated into several domains using Hamiltonian cycle. and link protection is performed on the cycles of the domains. We have shown that protection path length can be decreased up to $57{\%}$ with marginal increase of backup capacity. Our proposed method can provide high-speed protection with marginal increase of protection capacity.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.631-633
• /
• 2005

본 논문에서는 MEF(Metro Ethernet Forum) 기술 문서를 기반으로 메트로 이더넷 망을 위한 보호 기술을 소개하고 메트로 이더넷 망을 위한 링크 보호 알고리즘을 설계한다. 제안한 링크 보호 알고리즘은 각 서비스가 요구하는 링크 복구 시간(restoration time)을 고려하여 백업 경로를 미리 설정하는 방법과 장애가 발생한 이후에 우회 경로를 설정하는 방법을 사용한다. 백업 경로를 미리 설정함에 따라 자원의 이용률이 감소하므로 이를 해결하기 위하여 백업 경로 간 자원의 공유 방법을 도입하였으며, 서비스 특성에 따라 복구 경로에서 제한된 QoS를 제공하도록 하여 망의 가용성을 증가시일 수 있다.

• Information and Communications Magazine
• /
• v.32 no.7
• /
• pp.9-16
• /
• 2015

본고에서는 Transport SDN의 고확장성과 네트워크 장애를 신속하게 복구하기 위한 고가용성을 제공하는 SDN 컨트롤러 구조와 데이터 플레인 장애 복구 기술 동향을 소개한다. SDN 컨트롤러 기술에 대해서는 최근 많은 관심을 받고 있는 오픈소스 프레임워크 기반의 SDN 컨트롤러를 중심으로 확장성 및 고가용성 기술 동향을 살펴본다. 그리고 데이터 플레인에서 노드 및 경로 장애 발생 시 신속하게 경로를 복구하기 위한 보호기술과 복구기술 동향을 소개한다.

• Convergence Security Journal
• /
• v.12 no.3
• /
• pp.35-43
• /
• 2012

The explosive growth of Internet traffic cause by smart equipment such as smart phone has led to a dramatic increase in demand for data transmission capacity and network control architecture, which requires high transmission rates beyond the conventional transmission capability. Next generation networks are expected to be controlled by Generalized Multi-Protocol Label Switching(GMPLS) protocol suite and operating at multiple switching layers. In order to ensure the most efficient utilization of multilayer network resources, effective global provisioning that providing the network with the possibility of reacting in advance to traffic changes should be provided. In this paper, we proposes a new path selection scheme in multilayer optical networks based on the vertical PCE architecture and a different approach to efficiently exploit multiple PCE cooperation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39B no.1
• /
• pp.41-43
• /
• 2014

In this paper, we proposed a new protection scheme of the WDM mesh networks that can utilize the network resources efficiently while satisfying with the recovery time defined in the SLA (service level agreement). The protection scheme and the backup path are selected based on the comparison of the expected protection time with the SLA recovery time. The superior performance of the proposed protection scheme was demonstrated by comparing with ones of the existing protection schemes.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.5
• /
• pp.53-61
• /
• 2004

This paper proposes two types of sub-network based on dynamic restoration schemes to improve survivability of GMPLS networks and analyzes characteristics of these two schemes. The first proposed scheme divides with a whole GMPLS network into several sub-networks, applies a mixture of both restoration and protection methods according to fault location. The other scheme divides a whole GMPLS network into primary and secondary sub-networks, applies a restoration method in each sub-network according to fault location. In our simulation, we evaluate the performances of network resource utilization, restoration success rate, and mean restoration time and conduct its comparative analysis with conventional schemes. The simulation results show that the efficiency of network resource utilization in the proposed schemes is increased as compared with conventional restoration schemes (l＋l, 1:1, 1:N) in case of single-failed link. By contrast, we found that the performances of restoration success rate and mean restoration time in case of multi-failed link is lower than conventional restoration schemes. However, the probability that multi-failed link is occurred is very low so that the problem in practical GMPLS network is negligible.