Skip to Main content Skip to Navigation
Conference papers

Using deep learning for recommending and completing deployment descriptors in NFV

Abstract : Future software networks promises to fully automate the network management, configuration, deployment and operations of devices. Currently the deployment automation is enabled by orchestrators using descriptor files associated with Virtual Network Functions (VNFs) and Network Services (NSs), called VNF Descriptors (VNFDs) and NS Descriptors (NSDs). We focus our attention in this demo paper on VNFDs and we propose a framework for VNFD-Mining with Word Embeddings and Deep Neural Networks. The aim of the framework is to augment orchestrators with an ability to select, recommend and complete NFV descriptors given an initial description. The framework is trained initially with a catalogue of predefined VNFDs.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02482849
Contributor : Walid Gaaloul Connect in order to contact the contributor
Submitted on : Tuesday, February 18, 2020 - 12:07:57 PM
Last modification on : Tuesday, February 2, 2021 - 2:26:02 PM

Identifiers

Citation

Wassim Sellil Atoui, Imen Grida Ben Yahia, Walid Gaaloul. Using deep learning for recommending and completing deployment descriptors in NFV. NETSOFT 2019: IEEE Conference on Network Softwarization, Jun 2019, Paris, France. pp.233-235, ⟨10.1109/NETSOFT.2019.8806704⟩. ⟨hal-02482849⟩

Share

Metrics

Record views

35