Exploring the logical layer to support differentiated resilience classes in multilayer networks

The concept of providing differentiated classes of resilient services over communications networks has received attention in the literature. A number of proposals tried to address the problem by provisioning different resilience classes of service with various protection schemes. However, most of th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Annales des télécommunications 2018-02, Vol.73 (1-2), p.63-79
Hauptverfasser: Alashaikh, Abdulaziz, Tipper, David, Gomes, Teresa
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The concept of providing differentiated classes of resilient services over communications networks has received attention in the literature. A number of proposals tried to address the problem by provisioning different resilience classes of service with various protection schemes. However, most of these works are applied to a single layer and lack cross-layer coordination in multilayer scenarios. In addition, there is an increasing need for supporting services with very high resilience requirements over future communications networks. In this paper, we utilize the spine concept idea of embedding a subnetwork at the physical layer with comparatively high availability link and node values (Gomes et al. 2014 ; Alashaikh et al. Comput Netw 82:4–19 2015 ), to provide a foundation for resilience differentiation between multiple classes of flows. Cross-layer mapping and spine-aware routing are performed in a way that transfers the spine differentiation capability to the upper layer network and flows. We provide joint routing-mapping optimization formulations with different protection configurations and evaluate their performance in a multilayer scenario. Furthermore, we compare providing protection at the lower layer versus protection at the upper layer in terms of QoR class availability differentiation and resource requirements.
ISSN:0003-4347
1958-9395
DOI:10.1007/s12243-017-0616-1