Back to Search Start Over

On the Cross-Layer Network Planning for Flexible Ethernet Over Elastic Optical Networks

Authors :
Hui Liang
Nelson L. S. da Fonseca
Zuqing Zhu
Source :
IEEE Transactions on Network and Service Management. 18:3691-3705
Publication Year :
2021
Publisher :
Institute of Electrical and Electronics Engineers (IEEE), 2021.

Abstract

This article studies the cross-layer network planning that tries to combine flexible Ethernet (FlexE) and elastic optical networks (EONs), for FlexE-over-EONs. We focus our investigation on the most challenging setting, i.e. , the FlexE-over-EONs based on the FlexE-aware architecture, and consider both single-hop and multi-hop scenarios for the cross-layer planning. For the single-hop scenario, we assume that all the client flows are routed over end-to-end lightpaths in the EON. We formulate a mixed integer linear programming (MILP) model for this problem, transform it into the class constrained bin packing problem (CCBP), and leverage the primal-dual interior-point (PDIP) method to propose a polynomial-time approximation algorithm for it. Then, for the multi-hop scenario, we use a more realistic assumption that each client flow can be routed over multiple lightpaths in the EON. We show that after solving the virtual topology design, the cross-layer planning in this scenario can be transformed into that in the single-hop scenario. Therefore, an integer linear programming (ILP) model is formulated to tackle the virtual topology design, and we design a polynomial-time approximation algorithm for it by modifying the well-known branch-and-bond method. To evaluate the performance of our two-step method for the multi-hop scenario, we also propose a heuristic algorithm. Extensive simulations verify that regarding large-scale cross-layer planning for FlexE-over-EONs, our approximation algorithms are significantly more time-efficient than the ILP/MILP models, and their solutions have bounded gaps to the optimal ones and are much better than those of the heuristic.

Details

ISSN :
23737379
Volume :
18
Database :
OpenAIRE
Journal :
IEEE Transactions on Network and Service Management
Accession number :
edsair.doi...........124ff8d20cf54f50fb723ff7fca6e48f