An SLA-Aware Network Function Selection Algorithm for SFCs

Garg, Gaurav and Reddy, Venkatarami and Sathya, Vanlin and et al, . (2019) An SLA-Aware Network Function Selection Algorithm for SFCs. In: IEEE 2nd 5G World Forum (5GWF), 30 September - 2 October 2019, Dresden, Germany.

Full text not available from this repository. (Request a copy)

Abstract

5G is expected to support diverse services with different service requirements. Network Slicing (NS) is a new paradigm which aims to assign different services to a logically dedicated network. A network slice can host multiple services of similar requirements. The traffic flows belonging to a service are processed by a sequence of Virtual Network Functions (VNFs) which form a Service Function Chain (SFC). To utilize the resources efficiently, an operator can share the same VNF instance for multiple SFCs. The CPU utilization of a VNF instance can increase when new flows are accepted. This may increase the VNF delay which can result in the violation of SLA requirements of the existing flows. Therefore, an efficient VNF selection to form SFCs while meeting the SLA requirements is an important aspect that needs to be addressed. In this paper, we address the VNF selection problem for an SFC with the goal of supporting more number of SFCs by considering the dynamic variation of VNF delay. We develop a heuristic algorithm namely PENDANT which considers the effect of CPU utilization on VNF delay and migrates flows to other VNF instances to avoid SLA violations.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Franklin, AntonyUNSPECIFIED
Tamma, Bheemarjuna ReddyUNSPECIFIED
Item Type: Conference or Workshop Item (Paper)
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 04 Dec 2019 04:15
Last Modified: 04 Dec 2019 04:15
URI: http://raiithold.iith.ac.in/id/eprint/7103
Publisher URL: http://doi.org/10.1109/5GWF.2019.8911716
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 7103 Statistics for this ePrint Item