Capacity-efficient strategy for 100 dual-failure restorability in optical mesh networks utilising reconfigurable p-cycles and a forcer filling concept

Chuan-Ching Sue, J. Y. Du

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The problem of achieving 100 dual-failure restorability utilising reconfigurable p-cycle mechanisms has been investigated via three different p-cycle mechanisms derived from the integer linear programming model; complete-repair (CRP), incremental-repair (IRP) and dynamic-repair (DRP). An enhanced 100 dual-failure restoration mechanism is proposed, designated as modified DRP with forcer filling concept (MDRP/wFF), in which network protection is achieved using only reconfigurable span-protecting p-cycles without the addition of path-segment-protecting p-cycles as in DRP. As a result, the overhead incurred in differentiating between the two p-cycles becomes non-existent upon network restoration. Furthermore, a FF concept is employed to reduce the amount of spare capacity required to restore the network. The performance of the MDRP/wFF scheme was benchmarked against that of the CRP, IRP and DRP mechanisms. The simulation results show that MDRP/wFF achieves a performance tradeoff among CRP, IRP and DRP in terms of its spare capacity requirements and average number of reconfigured spans during the reconfiguration process [designated as average reconfiguration overhead (ARO)]. The MDRP/wFF scheme was found to have a lower spare capacity requirement but incurs some additional ARO compared with DRP.

Original languageEnglish
Pages (from-to)198-208
Number of pages11
JournalIET Communications
Volume3
Issue number2
DOIs
Publication statusPublished - 2009 Feb 16

Fingerprint

Repair
Restoration
Linear programming

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

@article{b03f785218b341069c6d3fcc021d5687,
title = "Capacity-efficient strategy for 100 dual-failure restorability in optical mesh networks utilising reconfigurable p-cycles and a forcer filling concept",
abstract = "The problem of achieving 100 dual-failure restorability utilising reconfigurable p-cycle mechanisms has been investigated via three different p-cycle mechanisms derived from the integer linear programming model; complete-repair (CRP), incremental-repair (IRP) and dynamic-repair (DRP). An enhanced 100 dual-failure restoration mechanism is proposed, designated as modified DRP with forcer filling concept (MDRP/wFF), in which network protection is achieved using only reconfigurable span-protecting p-cycles without the addition of path-segment-protecting p-cycles as in DRP. As a result, the overhead incurred in differentiating between the two p-cycles becomes non-existent upon network restoration. Furthermore, a FF concept is employed to reduce the amount of spare capacity required to restore the network. The performance of the MDRP/wFF scheme was benchmarked against that of the CRP, IRP and DRP mechanisms. The simulation results show that MDRP/wFF achieves a performance tradeoff among CRP, IRP and DRP in terms of its spare capacity requirements and average number of reconfigured spans during the reconfiguration process [designated as average reconfiguration overhead (ARO)]. The MDRP/wFF scheme was found to have a lower spare capacity requirement but incurs some additional ARO compared with DRP.",
author = "Chuan-Ching Sue and Du, {J. Y.}",
year = "2009",
month = "2",
day = "16",
doi = "10.1049/iet-com:20080029",
language = "English",
volume = "3",
pages = "198--208",
journal = "IET Communications",
issn = "1751-8628",
publisher = "Institution of Engineering and Technology",
number = "2",

}

TY - JOUR

T1 - Capacity-efficient strategy for 100 dual-failure restorability in optical mesh networks utilising reconfigurable p-cycles and a forcer filling concept

AU - Sue, Chuan-Ching

AU - Du, J. Y.

PY - 2009/2/16

Y1 - 2009/2/16

N2 - The problem of achieving 100 dual-failure restorability utilising reconfigurable p-cycle mechanisms has been investigated via three different p-cycle mechanisms derived from the integer linear programming model; complete-repair (CRP), incremental-repair (IRP) and dynamic-repair (DRP). An enhanced 100 dual-failure restoration mechanism is proposed, designated as modified DRP with forcer filling concept (MDRP/wFF), in which network protection is achieved using only reconfigurable span-protecting p-cycles without the addition of path-segment-protecting p-cycles as in DRP. As a result, the overhead incurred in differentiating between the two p-cycles becomes non-existent upon network restoration. Furthermore, a FF concept is employed to reduce the amount of spare capacity required to restore the network. The performance of the MDRP/wFF scheme was benchmarked against that of the CRP, IRP and DRP mechanisms. The simulation results show that MDRP/wFF achieves a performance tradeoff among CRP, IRP and DRP in terms of its spare capacity requirements and average number of reconfigured spans during the reconfiguration process [designated as average reconfiguration overhead (ARO)]. The MDRP/wFF scheme was found to have a lower spare capacity requirement but incurs some additional ARO compared with DRP.

AB - The problem of achieving 100 dual-failure restorability utilising reconfigurable p-cycle mechanisms has been investigated via three different p-cycle mechanisms derived from the integer linear programming model; complete-repair (CRP), incremental-repair (IRP) and dynamic-repair (DRP). An enhanced 100 dual-failure restoration mechanism is proposed, designated as modified DRP with forcer filling concept (MDRP/wFF), in which network protection is achieved using only reconfigurable span-protecting p-cycles without the addition of path-segment-protecting p-cycles as in DRP. As a result, the overhead incurred in differentiating between the two p-cycles becomes non-existent upon network restoration. Furthermore, a FF concept is employed to reduce the amount of spare capacity required to restore the network. The performance of the MDRP/wFF scheme was benchmarked against that of the CRP, IRP and DRP mechanisms. The simulation results show that MDRP/wFF achieves a performance tradeoff among CRP, IRP and DRP in terms of its spare capacity requirements and average number of reconfigured spans during the reconfiguration process [designated as average reconfiguration overhead (ARO)]. The MDRP/wFF scheme was found to have a lower spare capacity requirement but incurs some additional ARO compared with DRP.

UR - http://www.scopus.com/inward/record.url?scp=59849109660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59849109660&partnerID=8YFLogxK

U2 - 10.1049/iet-com:20080029

DO - 10.1049/iet-com:20080029

M3 - Article

AN - SCOPUS:59849109660

VL - 3

SP - 198

EP - 208

JO - IET Communications

JF - IET Communications

SN - 1751-8628

IS - 2

ER -