A Network Slice Resource Allocation and Optimization Model for End-to-End Mobile Networks

Abstract

With the fifth generation of mobile networks (5G) diverse new use cases arise, including enhanced Mobile Broadband (eMBB), industrial and sensor networks as well as highly safety and security critical communication services. These new use cases introduce very different requirements on the mobile communication networks, e.g., network reliability, latency and throughput. Massive network virtualization and end-to-end mobile network slicing are seen as key enablers to handle those differing requirements and providing mobile network services for the various 5G use cases by a shared physical network infrastructure. However, resource allocation and mobile network slice embedding are still unsolved problems. Therefore, in this paper a standardized and easy to understand Integer Linear Program for offline mobile network slice embedding, especially focusing on resource allocation and virtual node as well as link mapping, will be presented. Moreover, the question of how to efficiently determine a nearly optimal end-to-end mobile network slice embedding on a shared network infrastructure will be answered.

An Efficient Model for Mobile Network Slice Embedding under Resource Uncertainty

“…This model is based on the nearly optimal NSLE model previously published in [13] and advanced in [14]. This previous model assumes full and exact knowledge about resource availability.…”

“…The needed graph-theoretical notations are based on [16] and have been used similarly in [13] and [14]. Undirected graphs G, defined as an ordered pair G = (V, E), are used to model the physical network as well as the NSLs.…”

“…), as proposed in this paper, is compared to a simple objective function SIMPLE (SIM. ), as presented in [13]. γ is set to 1.5 for the UNC.…”

A Formal Optimization Model for 5G Mobile Network Slice Resource Allocation

“…However, solving the dynamic NSE is subject to ongoing research. This paper enhances the NSL resource allocation and optimization model and algorithm for end-to-end mobile networks proposed in [4]. The following research questions will be answered in this paper: regarding their specifications and resource as well as application chaining constraints?…”

“…3) What is the optimal embedding of the NSLs considering communication latency? Application function chaining is introduced into the Integer Linear Program (ILP) proposed in [4]. In addition to that, path splitting is added to the model, i.e., the possibility to use several paths in the physical network to serve a single virtual link.…”