Prototyping and Load Balancing the Service Based Architecture of 5G Core Using NFV

Kiran Buyakar, Tulja Vamshi and Tamma, Bheemarjuna Reddy and Franklin, Antony and et al, . (2019) Prototyping and Load Balancing the Service Based Architecture of 5G Core Using NFV. In: IEEE Conference on Network Softwarization (NetSoft), 24-28 June 2019, Paris, France.

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

Abstract

3GPP has chosen Service Based Architecture (SBA) for 5G Core (5GC). In this work, we build a prototype of SBA of 5GC from scratch using open source tools in Network Functions Virtualization (NFV) environment. In SBA, Network Functions (NFs) are designed to expose capabilities to consumers with interfaces, called Service Based Interface (SBI). To reduce the latency and the load on NFs, we use gRPC, a modern open-source Remote Procedure Call (RPC) framework, instead of REST for implementing SBI. We present a distributed service registration and discovery framework for 5GC using a software solution named Consul. We then propose usage of a Look Aside Load Balancer (LALB) for load balancing multiple NFs of 5GC. The control plane latency is reduced by routing traffic across multiple instances of Access and Mobility Management Function (AMF) via an LALB. Experimental results suggest that carefully chosen load balancing algorithms can significantly lessen the control plane latency when compared to simple random or round-robin schemes.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Tamma, Bheemarjuna ReddyUNSPECIFIED
Franklin, AntonyUNSPECIFIED
Item Type: Conference or Workshop Item (Paper)
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 03 Sep 2019 04:35
Last Modified: 03 Sep 2019 04:35
URI: http://raiithold.iith.ac.in/id/eprint/6097
Publisher URL: http://doi.org/10.1109/NETSOFT.2019.8806648
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 6097 Statistics for this ePrint Item