國立高雄大學圖資館 |

Language: English

Back

Reliable and Efficient Network Virtualization in Edge-Cloud Computing.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Reliable and Efficient Network Virtualization in Edge-Cloud Computing.
Author:	Shang, Xiaojun.
Published:	Ann Arbor : ProQuest Dissertations & Theses, 2023
Description:	144 p.
Notes:	Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
Notes:	Advisor: Yang, Yuanyuan.
Contained By:	Dissertations Abstracts International85-03B.
Subject:	Computer engineering.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30567509
ISBN:	9798380366113

Reliable and Efficient Network Virtualization in Edge-Cloud Computing.
Shang, Xiaojun.

Reliable and Efficient Network Virtualization in Edge-Cloud Computing. - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 144 p.

Source: Dissertations Abstracts International, Volume: 85-03, Section: B.

Thesis (Ph.D.)--State University of New York at Stony Brook, 2023.

This item must not be sold to any third party vendors.

Network virtualization involves transforming traditional hardware-based network components into software-based alternatives for benefits such as scalability, flexibility, and cost savings. Along with the proliferation of IoT devices and edge-native data, network virtualization is expanding to edge clouds and the proximity of IoT devices for pervasive connectivity, on-demand protection, and utmost privacy, which are necessities of the upcoming 6G communication. Nevertheless, heterogeneous and unpredictable edge environments present new challenges to implementing network virtualization effectively, such as reliability flaws, network congestion, and management complexity. To address such challenges, in this dissertation, we investigate how to provision, deploy, manage, and maintain edge-based virtual network services in a highly reliable and efficient manner with the following main contributions.First, we introduce a novel framework that enhances the reliability of virtual network functions (VNFs) in or close to the proximity of IoT and end users. This framework ensures the end-to-end reliability of interconnected VNFs under limited budgets and non-stationary failures, which contains multiple novel online approximation algorithms handling resource provision, backup allocation, and flow rerouting in a consistent way. Additionally, the framework includes a self-adaptive VNF backup architecture in case of even more constrained scenarios, such as VNFs located near IoT devices with few resources to support basic redundancy. It dynamically adjusts the number and distribution of VNF backups over the edge and the cloud to find the sweet spot between service reliability and cost efficiency.Regarding the risks of network congestion due to aggregated VNFs and edge resource limitations, we further propose a joint VNF placement and flow routing strategy to achieve low operating costs and network congestion control at the same time. It is worth noting that frequent traffic fluctuations at the network edge make it necessary to timely migrate VNFs and reroute corresponding traffic flows for real-time congestion control. Our solution makes it possible by combining both previous and future information to trade off the overhead of service migration and the gain of congestion relief.Besides algorithmic contributions, we also implement a new edge computing system that practically provides on-demand and high-quality virtual services to IoT and mobile users using a Function-as-a-Service (FaaS) pattern. The system utilizes a centralized control paradigm to migrate virtual services among edge nodes according to users’ movement for delay minimization. It also contains a distributed control paradigm on each edge server to trade off energy consumption and Quality of Experience (QoE) enhancement. Our design supports a broader range of virtual services other than network functions, paving the way to the integration of networking and computing for future data-intensive edge computing scenarios.In the end, we discuss the future research directions such as combining machine learning power with network virtualization in edge-cloud environments towards the upcoming pervasive network intelligence.

ISBN: 9798380366113Subjects--Topical Terms:

212944
Computer engineering.
Subjects--Index Terms:

Network virtualization

Reliable and Efficient Network Virtualization in Edge-Cloud Computing.
LDR:04494nmm a2200409 4500 001 655810
005 20240414211936.5
006 m o d
007 cr#unu||||||||
008 240620s2023 ||||||||||||||||| ||eng d
020 $a 9798380366113
035 $a (MiAaPQ)AAI30567509
035 $a AAI30567509
040 $a MiAaPQ $c MiAaPQ
100 1 $a Shang, Xiaojun. $3 966922
245 1 0 $a Reliable and Efficient Network Virtualization in Edge-Cloud Computing.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 144 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
500 $a Advisor: Yang, Yuanyuan.
502 $a Thesis (Ph.D.)--State University of New York at Stony Brook, 2023.
506 $a This item must not be sold to any third party vendors.
520 $a Network virtualization involves transforming traditional hardware-based network components into software-based alternatives for benefits such as scalability, flexibility, and cost savings. Along with the proliferation of IoT devices and edge-native data, network virtualization is expanding to edge clouds and the proximity of IoT devices for pervasive connectivity, on-demand protection, and utmost privacy, which are necessities of the upcoming 6G communication. Nevertheless, heterogeneous and unpredictable edge environments present new challenges to implementing network virtualization effectively, such as reliability flaws, network congestion, and management complexity. To address such challenges, in this dissertation, we investigate how to provision, deploy, manage, and maintain edge-based virtual network services in a highly reliable and efficient manner with the following main contributions.First, we introduce a novel framework that enhances the reliability of virtual network functions (VNFs) in or close to the proximity of IoT and end users. This framework ensures the end-to-end reliability of interconnected VNFs under limited budgets and non-stationary failures, which contains multiple novel online approximation algorithms handling resource provision, backup allocation, and flow rerouting in a consistent way. Additionally, the framework includes a self-adaptive VNF backup architecture in case of even more constrained scenarios, such as VNFs located near IoT devices with few resources to support basic redundancy. It dynamically adjusts the number and distribution of VNF backups over the edge and the cloud to find the sweet spot between service reliability and cost efficiency.Regarding the risks of network congestion due to aggregated VNFs and edge resource limitations, we further propose a joint VNF placement and flow routing strategy to achieve low operating costs and network congestion control at the same time. It is worth noting that frequent traffic fluctuations at the network edge make it necessary to timely migrate VNFs and reroute corresponding traffic flows for real-time congestion control. Our solution makes it possible by combining both previous and future information to trade off the overhead of service migration and the gain of congestion relief.Besides algorithmic contributions, we also implement a new edge computing system that practically provides on-demand and high-quality virtual services to IoT and mobile users using a Function-as-a-Service (FaaS) pattern. The system utilizes a centralized control paradigm to migrate virtual services among edge nodes according to users’ movement for delay minimization. It also contains a distributed control paradigm on each edge server to trade off energy consumption and Quality of Experience (QoE) enhancement. Our design supports a broader range of virtual services other than network functions, paving the way to the integration of networking and computing for future data-intensive edge computing scenarios.In the end, we discuss the future research directions such as combining machine learning power with network virtualization in edge-cloud environments towards the upcoming pervasive network intelligence.
590 $a School code: 0771.
650 4 $a Computer engineering. $3 212944
650 4 $a Computer science. $3 199325
650 4 $a Information technology. $3 184390
653 $a Network virtualization
653 $a Network components
653 $a IoT devices
653 $a Virtual network functions
653 $a Edge-native data
653 $a Edge clouds
690 $a 0464
690 $a 0489
690 $a 0984
710 2 $a State University of New York at Stony Brook. $b Computer Engineering. $3 766090
773 0 $t Dissertations Abstracts International $g 85-03B.
790 $a 0771
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30567509