Development of resilient and resource-efficient service function chaining in 5G networks.
Abstract
This is a research based project on resilient and resource e cient service function
chaining in 5G networks.
This project addresses the challenge on how to guarantee survivability of critical
services while e ciently utilizing the exhaustible resources of the shared substrate
network.
First, this project proposes a generic multi-layer graph based algorithm as an
alternative algorithm for SFC deployment across a single substrate network.
Based on the proposed algorithm, a migration-aware algorithm that enables non-
critical service applications to share the unused backup resources of critical ser-
vices in a way that minimizes future preemption of these users is proposed. The
migration-aware algorithm results into minimal resource consumption in most sce-
narios compared to a dedicated backup scheme and an improved performance in
terms of the number of service preemptions compared to a cost based algorithm.
In addition,the low priority users are pre-empted from the shared resources incase
of failure of the critial user, therefore the project proposes a QoS-aware global-
rerouting algorithm for remapping such users while minimizing the level of service
interruption.
The migration-aware algorithm resulted in more than 8% resource saving in most
scenarios compared to a dedicated backup scheme and more than 70% performance
improvement in terms of the number of service preemptions compared to a cost
based algorithm. Additionally, whenever low priority users are preempted from the
shared resources, the project proposed a QoS-aware global-rerouting algorithm for
remapping such users while minimizing the level of service interruption resulting
from migration of surviving VNFs and virtual links to other substrate nodes and
paths. The proposed algorithm shown to outperform a service restoration scheme
based on local rerouting in terms of successful service restoration and resource
consumption.