<abstract xmlns="http://www.w3.org/1999/xhtml"><p>In recent years elastic optical networks have been perceived as a prospective choice for future optical networks due to better adjustment and utilization of optical resources than is the case with traditional wavelength division multiplexing networks. In the paper we investigate the elastic architecture as the communication network for distributed data centers. We address the problems of optimization of routing and spectrum assignment for large-scale computing systems based on an elastic optical architecture; particularly, we concentrate on anycast user to data center traffic optimization. We assume that computational resources of data centers are limited. For this offline problems we formulate the integer linear programming model and propose a few heuristics, including a meta-heuristic algorithm based on a tabu search method. We report computational results, presenting the quality of approximate solutions and efficiency of the proposed heuristics, and we also analyze and compare some data center allocation scenarios.</p></abstract>

In recent years elastic optical networks have been perceived as a prospective choice for future optical networks due to better adjustment and utilization of optical resources than is the case with traditional wavelength division multiplexing networks. In the paper we investigate the elastic architecture as the communication network for distributed data centers. We address the problems of optimization of routing and spectrum assignment for large-scale computing systems based on an elastic optical architecture; particularly, we concentrate on anycast user to data center traffic optimization. We assume that computational resources of data centers are limited. For this offline problems we formulate the integer linear programming model and propose a few heuristics, including a meta-heuristic algorithm based on a tabu search method. We report computational results, presenting the quality of approximate solutions and efficiency of the proposed heuristics, and we also analyze and compare some data center allocation scenarios.

Heuristic algorithms for joint optimization of unicast and anycast traffic in elastic optical network–based large–scale computing systems

Department of Systems and Computer Networks, Faculty of Electronics

Wrocław University of Science and Technology

International Journal of Applied Mathematics and Computer Science

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

In recent years elastic optical networks have been perceived as a prospective choice for future optical networks due to better adjustment and utilization of...

Heuristic algorithms for joint optimization of unicast and anycast traffic in elastic optical network–based large–scale computing systems

Publié en ligne: 23 sept. 2017

Pages: 605 - 622

Reçu: 01 sept. 2016

Accepté: 03 juin 2017

DOI: https://doi.org/10.1515/amcs-2017-0043

Mots clés
elastic optical networks, optimization, distributed data center, algorithm, tabu search

© 2017 Marcin Markowski, published by De Gruyter Open

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Heuristic algorithms for joint optimization of unicast and anycast traffic in elastic optical network–based large–scale computing systems

Publié en ligne: 23 sept. 2017

Pages: 605 - 622

Reçu: 01 sept. 2016

Accepté: 03 juin 2017

DOI: https://doi.org/10.1515/amcs-2017-0043

Mots cléselastic optical networks, optimization, distributed data center, algorithm, tabu search

© 2017 Marcin Markowski, published by De Gruyter Open

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Mots clés
elastic optical networks, optimization, distributed data center, algorithm, tabu search