2007Journal of Optical NetworkingRequires access

Performance study of IP and SONET grooming in optical WDM mesh networks

Ramakrishna Shenai, Krishna M. Sivalingam

Open publisher page 1 citations

Abstract

We study the problem of IP traffic grooming in optical wavelength-division-multiplexed (WDM) mesh networks. Two different grooming architectures are considered: (i) IP/MPLS grooming, and (ii) SONET grooming. The IP/MPLS grooming approach provides a fine level of grooming granularity realized via software-based implementations but has router-processing-related bottlenecks. The SONET grooming approach is realized in dedicated hardware, but it is more expensive and typically offers limited grooming granularity choices. We also consider multipath routing as used in SONET virtual concatenation techniques, and we study the impact of the differential delay constraint. A detailed comparison of the different grooming options in terms of blocking probability, lightpath utilization, and survivability issues is presented. The simulation analysis helps quantify the improved performance of finer granularity grooming approaches over the coarser SONET-based approaches.

About this research paper

What this paper is about

We study the problem of IP traffic grooming in optical wavelength-division-multiplexed (WDM) mesh networks. Two different grooming architectures are considered: (i) IP/MPLS grooming, and (ii) SONET grooming. The IP/MPLS grooming approach provides a fine level of grooming granularity realized via software-based implementations but has router-processing-related bottlenecks. The SONET grooming approach is realized in dedicated hardware, but it is more expensive and typically offers limited grooming granularity choices. We also consider multipath routing as used in SONET virtual concatenation techniques, and we study the impact of the differential delay constraint. A detailed comparison of the different grooming options in terms of blocking probability, lightpath utilization, and survivability issues is presented. The simulation analysis helps quantify the improved performance of finer granularity grooming approaches over the coarser SONET-based approaches.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We study the problem of IP traffic grooming in optical wavelength-division-multiplexed (WDM) mesh networks. Two different grooming architectures are considered: (i) IP/MPLS grooming, and (ii) SONET grooming. The IP/MPLS grooming approach provides a fine level of grooming granularity realized via software-based implementations but has router-processing-related bottlenecks. The SONET grooming approach is realized in dedicated hardware, but it is more expensive and typically offers limited grooming granularity choices. We also consider multipath routing as used in SONET virtual concatenation techniques, and we study the impact of the differential delay constraint. A detailed comparison of the different grooming options in terms of blocking probability, lightpath utilization, and survivability issues is presented. The simulation analysis helps quantify the improved performance of finer granularity grooming approaches over the coarser SONET-based approaches.

Key concepts: Synchronous optical networking, Traffic grooming, Computer network, Optical mesh network, Multiprotocol Label Switching, Computer science, Ethernet over SDH, Wavelength-division multiplexing

Related papers

Back to paper searchBrowse research topicsOriginal source
Performance study of IP and SONET grooming in optical WDM mesh networks — Research Paper | ScholarLens