CLB-LP: Controller Load Balancing Based on Load Prediction Using Deep Learning for Software-Defined IoT Networks

IF 6.7 2区计算机科学 Q1 ENGINEERING, MULTIDISCIPLINARY

IEEE Transactions on Network Science and Engineering Pub Date : 2024-10-28 DOI:10.1109/TNSE.2024.3487355

Quanze Liu;Yong Liu;Qian Meng;Tianyi Yu

{"title":"CLB-LP: Controller Load Balancing Based on Load Prediction Using Deep Learning for Software-Defined IoT Networks","authors":"Quanze Liu;Yong Liu;Qian Meng;Tianyi Yu","doi":"10.1109/TNSE.2024.3487355","DOIUrl":null,"url":null,"abstract":"By integrating Software-Defined Networking (SDN), Software-Defined Internet of Things (SD-IoT) simplifies network configuration while enhancing controllability. The expansion of the IoT scale has led to the emergence of the multiple controller architecture. However, it introduces the challenge of controller load imbalances. Existing schemes primarily focus on dynamic switch migration. Nonetheless, conventional strategies use real-time network information for load measurement and selection of candidate switches, which reduces load balancing performance due to inaccurate load measurement. Moreover, existing approaches struggle to balance load balancing rate and migration cost when selecting the target controllers. Therefore, we propose the controller load balancing based on load prediction (CLB-LP) scheme, which uses historical load data to predict future load, thereby avoiding unnecessary switch migrations. Additionally, we introduce a switch selection algorithm that combines load prediction and migration probability to select candidate switches, effectively improving load balancing performance. Furthermore, we present a target controller selection algorithm based on the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), which improves the load balancing rate while reducing migration cost. Finally, we evaluate the effectiveness of CLB-LP, and compared to existing schemes, its load balancing rate and response time are 29.4% higher and 28.5% lower, respectively.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"12 1","pages":"173-185"},"PeriodicalIF":6.7000,"publicationDate":"2024-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Network Science and Engineering","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10737143/","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

By integrating Software-Defined Networking (SDN), Software-Defined Internet of Things (SD-IoT) simplifies network configuration while enhancing controllability. The expansion of the IoT scale has led to the emergence of the multiple controller architecture. However, it introduces the challenge of controller load imbalances. Existing schemes primarily focus on dynamic switch migration. Nonetheless, conventional strategies use real-time network information for load measurement and selection of candidate switches, which reduces load balancing performance due to inaccurate load measurement. Moreover, existing approaches struggle to balance load balancing rate and migration cost when selecting the target controllers. Therefore, we propose the controller load balancing based on load prediction (CLB-LP) scheme, which uses historical load data to predict future load, thereby avoiding unnecessary switch migrations. Additionally, we introduce a switch selection algorithm that combines load prediction and migration probability to select candidate switches, effectively improving load balancing performance. Furthermore, we present a target controller selection algorithm based on the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), which improves the load balancing rate while reducing migration cost. Finally, we evaluate the effectiveness of CLB-LP, and compared to existing schemes, its load balancing rate and response time are 29.4% higher and 28.5% lower, respectively.

查看原文本刊更多论文

CLB-LP：基于负载预测的基于深度学习的软件定义物联网控制器负载均衡

软件定义物联网（SD-IoT）通过集成SDN （software defined Networking）技术，简化了网络配置，增强了可控性。物联网规模的扩大导致了多控制器架构的出现。然而，它引入了控制器负载不平衡的挑战。现有方案主要关注动态交换机迁移。然而，传统的策略使用实时网络信息进行负载测量和候选交换机的选择，由于不准确的负载测量，降低了负载均衡的性能。此外，现有的方法在选择目标控制器时难以平衡负载均衡率和迁移成本。因此，我们提出基于负载预测的控制器负载均衡（CLB-LP）方案，该方案使用历史负载数据预测未来负载，从而避免不必要的交换机迁移。此外，我们还引入了一种结合负载预测和迁移概率的交换机选择算法来选择候选交换机，有效地提高了负载均衡性能。在此基础上，提出了一种基于TOPSIS （Order of Preference by Similarity to Ideal Solution）的目标控制器选择算法，在降低迁移成本的同时提高了负载均衡率。最后，我们评估了CLB-LP的有效性，与现有方案相比，它的负载均衡率和响应时间分别提高了29.4%和28.5%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Transactions on Network Science and Engineering Engineering-Control and Systems Engineering

CiteScore

12.60

自引率

9.10%

发文量

393

期刊介绍： The proposed journal, called the IEEE Transactions on Network Science and Engineering (TNSE), is committed to timely publishing of peer-reviewed technical articles that deal with the theory and applications of network science and the interconnections among the elements in a system that form a network. In particular, the IEEE Transactions on Network Science and Engineering publishes articles on understanding, prediction, and control of structures and behaviors of networks at the fundamental level. The types of networks covered include physical or engineered networks, information networks, biological networks, semantic networks, economic networks, social networks, and ecological networks. Aimed at discovering common principles that govern network structures, network functionalities and behaviors of networks, the journal seeks articles on understanding, prediction, and control of structures and behaviors of networks. Another trans-disciplinary focus of the IEEE Transactions on Network Science and Engineering is the interactions between and co-evolution of different genres of networks.