SD-WAN for Cloud Edge Computing Continuum interconnection

Mokhtari, Ayoub; Ksentini, Adlen
GLOBECOM 2024, IEEE Global Communications Conference: Communications Software and Multimedia, 8-12 December 2024, Cape Town, South Africa

Driven by the emergence of IoT services and the need to process data as close as possible to its source, the shift from centralized Cloud computing to Cloud-Edge computing has become essential. This transition has given rise to a new comput-ing paradigm known as the Cloud Edge Computing Continuum (CECC), where microservices that constitute Cloud-native applica-tions are geographically dispersed across different federated CECC nodes. This transformation has brought with it new challenges in terms of interconnecting geographically dispersed CECC nodes through Wide Area Networks (WANs) to support microservices interconnection needs (bandwidth, latency, etc.). To address this challenge, in this paper, we leverage Software Defined Wide Area Networks (SD-WAN) for interconnecting the CECC nodes. We first introduce a comprehensive CECC SD-WAN architecture by demonstrating how dynamic application-aware routing facilitated by SD-WAN can overcome these limitations, ensuring efficient and flexible network performance across geographically dispersed Cloud sites. Then, for the instantiation of the architecture, we adapt an open-source SD-WAN implementation to the context of CECC. We also show performance improvement, with our approach being 10% more efficient in terms of memory usage than the VRF lite approach in the open-source implementation.


Type:
Conférence
City:
Cape Town
Date:
2024-12-08
Department:
Systèmes de Communication
Eurecom Ref:
7823
Copyright:
© 2024 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

PERMALINK : https://www.eurecom.fr/publication/7823