{"title":"ODRAD: An optical wireless DCN dynamic-bandwidth reconfiguration with AWGR and deep reinforcement learning","authors":"Kassahun Geresu, Huaxi Gu, Meaad Fadhel, Wenting Wei, Xiaoshan Yu","doi":"10.1016/j.osn.2024.100771","DOIUrl":null,"url":null,"abstract":"<div><p>The rapid growth of Data Center Network (DCN) traffic has brought new challenges, such as limited bandwidth, high latency, and packet loss to existing DCNs based on electrical switches. Because of its theoretically unlimited bandwidth and faster data transmission speeds, optical switching can overcome the problems of electrically switched DCNs. Additionally, numerous research works have been devoted to optical wired DCNs. However, static and fixed-topology DCNs based on optical interconnects significantly limit their flexibility, scalability, and reconfigurability to provide adaptive bandwidth for traffic with heterogeneous characteristics. In this study, we propose and conduct performance evaluations on a reconfigurable optical wireless DCN architecture based on distributed Software-Defined Networking (SDN), Deep Reinforcement Learning (DRL), Semiconductor Optical Amplifier (SOA), and Arrayed Waveguide Grating Router (AWGR). Our architecture is called ODRAD (which stands for Optical Wireless DCN Dynamic-bandwidth Reconfiguration with AWGR and Deep Reinforcement Learning). A Mininet simulation model is established to further verify the reconfigurability of the ODRAD network for various server scales. Based on experimental verification, ODRAD achieves an average end-to-end server latency of <span><math><mrow><mn>5</mn><mo>.</mo><mn>2</mn><mspace></mspace><mi>μ</mi><mi>s</mi></mrow></math></span> under a load of 99%. Compression results demonstrate a 17.36% improvement in packet rate latency performance compared to RotorNet and a 15.21% improvement compared to OPSquare at a load of 99% as the ODRAD network scales from 2,560 to 40,960 servers. Furthermore, ODRAD exhibits effective throughput across different routing protocols, DCN scales and loads.</p></div>","PeriodicalId":54674,"journal":{"name":"Optical Switching and Networking","volume":null,"pages":null},"PeriodicalIF":1.9000,"publicationDate":"2024-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Optical Switching and Networking","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1573427724000018","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0
Abstract
The rapid growth of Data Center Network (DCN) traffic has brought new challenges, such as limited bandwidth, high latency, and packet loss to existing DCNs based on electrical switches. Because of its theoretically unlimited bandwidth and faster data transmission speeds, optical switching can overcome the problems of electrically switched DCNs. Additionally, numerous research works have been devoted to optical wired DCNs. However, static and fixed-topology DCNs based on optical interconnects significantly limit their flexibility, scalability, and reconfigurability to provide adaptive bandwidth for traffic with heterogeneous characteristics. In this study, we propose and conduct performance evaluations on a reconfigurable optical wireless DCN architecture based on distributed Software-Defined Networking (SDN), Deep Reinforcement Learning (DRL), Semiconductor Optical Amplifier (SOA), and Arrayed Waveguide Grating Router (AWGR). Our architecture is called ODRAD (which stands for Optical Wireless DCN Dynamic-bandwidth Reconfiguration with AWGR and Deep Reinforcement Learning). A Mininet simulation model is established to further verify the reconfigurability of the ODRAD network for various server scales. Based on experimental verification, ODRAD achieves an average end-to-end server latency of under a load of 99%. Compression results demonstrate a 17.36% improvement in packet rate latency performance compared to RotorNet and a 15.21% improvement compared to OPSquare at a load of 99% as the ODRAD network scales from 2,560 to 40,960 servers. Furthermore, ODRAD exhibits effective throughput across different routing protocols, DCN scales and loads.
期刊介绍:
Optical Switching and Networking (OSN) is an archival journal aiming to provide complete coverage of all topics of interest to those involved in the optical and high-speed opto-electronic networking areas. The editorial board is committed to providing detailed, constructive feedback to submitted papers, as well as a fast turn-around time.
Optical Switching and Networking considers high-quality, original, and unpublished contributions addressing all aspects of optical and opto-electronic networks. Specific areas of interest include, but are not limited to:
• Optical and Opto-Electronic Backbone, Metropolitan and Local Area Networks
• Optical Data Center Networks
• Elastic optical networks
• Green Optical Networks
• Software Defined Optical Networks
• Novel Multi-layer Architectures and Protocols (Ethernet, Internet, Physical Layer)
• Optical Networks for Interet of Things (IOT)
• Home Networks, In-Vehicle Networks, and Other Short-Reach Networks
• Optical Access Networks
• Optical Data Center Interconnection Systems
• Optical OFDM and coherent optical network systems
• Free Space Optics (FSO) networks
• Hybrid Fiber - Wireless Networks
• Optical Satellite Networks
• Visible Light Communication Networks
• Optical Storage Networks
• Optical Network Security
• Optical Network Resiliance and Reliability
• Control Plane Issues and Signaling Protocols
• Optical Quality of Service (OQoS) and Impairment Monitoring
• Optical Layer Anycast, Broadcast and Multicast
• Optical Network Applications, Testbeds and Experimental Networks
• Optical Network for Science and High Performance Computing Networks