Control transmission pace at IP layer to avoid packet drop

Guojun Jin, Haina Tang
{"title":"Control transmission pace at IP layer to avoid packet drop","authors":"Guojun Jin, Haina Tang","doi":"10.1109/IPOM.2004.1547593","DOIUrl":null,"url":null,"abstract":"Avoiding packet loss is critical for time sensitive network applications, such as multimedia streams for video/voice. Delaying and dropping low priority packets to ensure high priority and time sensitive data stream deliver during network congestion is a basic QoS (quality of service) mechanism over current network infrastructure. This mechanism works if time sensitive data stream is the minority of the network traffic and if the network is not very congested. The methodology of dropping low priority data does not scale when time sensitive data stream uses high percentage of network bandwidth. This is because bandwidth required by video/audio applications can vary in very wide range when real-time data becomes majority network traffic, that is, television (TV), telephone, visual telephone, videoconferencing, gaming, and other video/audio based applications are all deployed on Internet. Then, what is the proper percentage of bandwidth to reserve? and which packets should be dropped if available bandwidth is less than demanding? A major issue is that letting bottleneck routers drop packets is not a proper methodology to guarantee quality of service. If packets cannot be delivered due to exhausted network bandwidth, these packets should be tossed as earlier as possible to reduce bandwidth waste or should be delayed at transmission hosts for later transmission. Also, applications should have right to selectively toss data for enhancing service quality, rather than let routers randomly drop packets. Therefore, mechanisms to avoid packet drop need to be deployed in Internet infrastructure. This paper studies how well priority (class) based traffic shaping can help time sensitive data delivery, addresses technology of packet drop avoidance (PDA), and shows how packet drop avoidance mechanism improves real-time applications' performance by reducing bandwidth waste, packet delay and loss. This paper then addresses why PDA should be deployed in Internet protocol (IP) layer.","PeriodicalId":197627,"journal":{"name":"2004 IEEE International Workshop on IP Operations and Management","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2004 IEEE International Workshop on IP Operations and Management","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IPOM.2004.1547593","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

Avoiding packet loss is critical for time sensitive network applications, such as multimedia streams for video/voice. Delaying and dropping low priority packets to ensure high priority and time sensitive data stream deliver during network congestion is a basic QoS (quality of service) mechanism over current network infrastructure. This mechanism works if time sensitive data stream is the minority of the network traffic and if the network is not very congested. The methodology of dropping low priority data does not scale when time sensitive data stream uses high percentage of network bandwidth. This is because bandwidth required by video/audio applications can vary in very wide range when real-time data becomes majority network traffic, that is, television (TV), telephone, visual telephone, videoconferencing, gaming, and other video/audio based applications are all deployed on Internet. Then, what is the proper percentage of bandwidth to reserve? and which packets should be dropped if available bandwidth is less than demanding? A major issue is that letting bottleneck routers drop packets is not a proper methodology to guarantee quality of service. If packets cannot be delivered due to exhausted network bandwidth, these packets should be tossed as earlier as possible to reduce bandwidth waste or should be delayed at transmission hosts for later transmission. Also, applications should have right to selectively toss data for enhancing service quality, rather than let routers randomly drop packets. Therefore, mechanisms to avoid packet drop need to be deployed in Internet infrastructure. This paper studies how well priority (class) based traffic shaping can help time sensitive data delivery, addresses technology of packet drop avoidance (PDA), and shows how packet drop avoidance mechanism improves real-time applications' performance by reducing bandwidth waste, packet delay and loss. This paper then addresses why PDA should be deployed in Internet protocol (IP) layer.
控制IP层的传输速度,避免丢包
避免丢包对于时间敏感的网络应用是至关重要的,比如视频/语音的多媒体流。在网络拥塞时,延迟和丢弃低优先级的数据包以保证高优先级和时间敏感的数据流的传输是当前网络基础设施的基本QoS (service quality)机制。如果对时间敏感的数据流只占网络流量的一小部分,并且网络不是很拥挤,这种机制就可以工作。当时间敏感型数据流占用的网络带宽比例较大时,这种丢弃低优先级数据的方法不适用。这是因为当实时数据成为主要的网络流量时,视频/音频应用程序所需的带宽可能会有很大的变化,也就是说,电视(TV)、电话、可视电话、视频会议、游戏和其他基于视频/音频的应用程序都部署在Internet上。那么,预留带宽的适当百分比是多少呢?如果可用带宽低于要求,应该丢弃哪些数据包?一个主要的问题是,让瓶颈路由器丢弃数据包并不是保证服务质量的合适方法。如果由于网络带宽耗尽而无法发送数据包,则应尽早丢弃这些数据包以减少带宽浪费,或者将这些数据包延迟到传输主机以供后续传输。此外,应用程序应该有权有选择地丢弃数据以提高服务质量,而不是让路由器随机丢弃数据包。因此,需要在Internet基础设施中部署避免丢包的机制。本文研究了基于优先级(类)的流量整形如何帮助时间敏感数据传输,讨论了丢包避免技术(PDA),并展示了丢包避免机制如何通过减少带宽浪费、数据包延迟和丢失来提高实时应用程序的性能。然后,本文阐述了为什么PDA应该部署在互联网协议(IP)层。
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
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信