CompRess: Composing overlay service resources for end‐to‐end network slices using semantic user intents. Issue 1 (6th September 2019)

• Record Type: Journal Article
• Authors: Chen, Xi; Yu, Hongfang; Xu, Shizhong; Du, Xiaojiang
• Other Names: Aguiar Rui L. guestEditor.; Jeon Seil guestEditor.; Pasquini Rafael guestEditor.; Rawat Danda B. guestEditor.; Wang Qi guestEditor.
• Is Part Of: Transactions on emerging telecommunications technologies. Volume 31:Issue 1(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-09-06
• Subjects: Telecommunication -- Periodicals; 384.05
• DOI: 10.1002/ett.3728
• Languages: English
• ISSNs: 2161-5748

Abstract

Abstract: Network slicing is considered as a promising approach to bring flexible overlay services to end users, especially in the emerging 5G networks. Vertical network slices have been introducing new players into the arena of networked service provisioning such as automobile, healthcare, etc. Software‐Defined Networking (SDN) and Network Function Virtualization (NFV) are the key enabling technologies for network slicing. The combination of SDN and NFV brings benefits such as network programmability, flexible resources composition, etc; nevertheless, it also introduces complexity in expressing resources composition logics. Following the Intent‐Based Networking (IBN) ideology, this paper presents CompRess, an IBN practice and implementation of overlay service resources composition for end‐to‐end network slices based on Semantic Web techniques in SDN/NFV networks. Web Ontology Language and Resource Description Framework are used for resources modeling and description whereas SPARQL, a structural human‐readable query language, is adopted for user intent expression to bring high‐level abstraction and coarser‐grained encapsulation of low‐level operations. Meanwhile, back‐end K Shortest Paths–based algorithms are derived to compose a multichain Virtual Network Function Forwarding Graph (VNF‐FG) in an IBN flavor where several service function chains can be orchestrated for traffic steering under the hood of front‐end SPARQL‐expressed user intents. The simulation results demonstrate CompRess' applicability in multichain VNF‐FG composition for end‐to‐end network slices. Abstract : Semantic Web based user intent expression. The extended SPARQL language to support complex overlay service resource composition logics and requirements. The proposed multi‐chain VNF‐FG composition algorithm works in a topology‐agnostic fashion, decoupled from the underlay network. High‐performance deviation method for multi‐chain VNF‐FG composition.