Noisy neighbor detection and avoidance for network slicing in 5G
Résumé
This paper addresses the problem of noisy neighbor in the context of network slicing in 5G networks. Noisy neighbor is considered as an important issue since VNFs or VMs running on the same physical machine compete for resources, such as CPU, memory or bandwidth, which can degrade their performance. In this paper, we present a new approach based on supervised learning and optimization that detects and avoids bad neighboring VNFs through intelligent VNFs placement and migration.
Domaines
Informatique [cs]
Fichier principal
Noisy neighbor detection and avoidance for network.pdf (1.29 Mo)
Télécharger le fichier
Origine | Fichiers produits par l'(les) auteur(s) |
---|