Resource Allocation Over EON-Based Infrastructures in a Network Virtualization Environment
| Autor: | Pablo Andrés Maya, Juan Felipe Botero, Paola Soto |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Flexibility (engineering)
Computer Networks and Communications Computer science Distributed computing Network virtualization 020206 networking & telecommunications 02 engineering and technology Virtualization computer.software_genre Scalability 0202 electrical engineering electronic engineering information engineering Resource allocation Resource management Electrical and Electronic Engineering Heuristics computer Integer programming |
| Zdroj: | IEEE Transactions on Network and Service Management. 16:13-26 |
| ISSN: | 2373-7379 |
| DOI: | 10.1109/tnsm.2018.2883488 |
| Popis: | Through network virtualization, new networking paradigms can be easily implemented by sharing a common physical infrastructure between different users and applications. A fundamental aspect in this environment is how to efficiently map virtual resources onto physical resources, frequently referred as virtual network embedding. Integer linear programming, heuristics and meta-heuristics have been commonly used to solve this problem. However, efficient allocation algorithms cannot ignore the main characteristics of the physical infrastructure. Simultaneously, elastic optical networks (EONs), are gaining attention of researchers and industry due to their high spectrum efficiency, flexibility and adaptability that allow higher transmission rates able to cope with the tremendous IP traffic demand of new applications and services. In this paper, we formulate a mathematical model for the offline virtual optical network embedding assuming a transparent EON-based infrastructure. We then propose a meta-heuristic (PRVONE) based on an optical-path ranking system to provide scalable embeddings within reasonable running time. We evaluate our work through extensive simulations and compare it with state-of-the-art approaches. Results show that PRVONE can achieve improvements in terms of acceptance ratio in all test case scenarios while providing load balanced embeddings. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|