Impact of Access Bandwidth on Packet Loss: A Flow-level Analysis

被引:0
|
作者
Mehmood, Muhammad Amir [1 ]
Sarrar, Nadi [2 ]
Uhlig, Steve [3 ]
Feldmann, Anja [2 ]
机构
[1] UET, Al Khawarizmi Inst Comp Sci, Lahore, Pakistan
[2] Tech Univ Berlin, T Labs, Berlin, Germany
[3] Queen Many Univ London, London, England
关键词
Access networks; Flows; Burst losses;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
High-speed Internet connectivity is becoming more ubiquitous due to advances in access technologies. On the one hand, the increased access speed allows flows to consume the larger available bandwidth. On the other hand, large aggressive flows lead to more bursty traffic. In this paper, we study the impact of higher broadband access speeds on the resulting packet losses, and dissect those losses into their constituent flows. Based on tightly controlled experiments using open source tools, we show that as ISPs are increasing the access bandwidth of their residential customers, the effects on flow-level properties are non-trivial. Surprisingly, the average number of flows that participate in a packet loss burst remains small, as compared to the burst size. As expected, higher access bandwidth translates into larger packet loss bursts. Furthermore, large flows dominate packet losses, and even more so as access bandwidth increases. In contrast, irrespective of the access bandwidth, a few small flows systematically participate in loss episodes, and suffer significantly due to their limited size.
引用
收藏
页码:259 / 264
页数:6
相关论文
共 50 条
  • [1] Flow-Level Loss Detection with Δ-Sketches
    Feibish, Shir Landau
    Liu, Zaoxing
    Ivkin, Nikita
    Chen, Xiaoqi
    Braverman, Vladimir
    Rexford, Jennifer
    [J]. PROCEEDINGS OF THE 2022 SYMPOSIUM ON SDN RESEARCH, SOSR 2022, 2022, : 25 - 32
  • [2] On the Flow-level Dynamics of a Packet-switched Network
    Moallemi, Ciamac
    Shah, Devavrat
    [J]. SIGMETRICS 2010: PROCEEDINGS OF THE 2010 ACM SIGMETRICS INTERNATIONAL CONFERENCE ON MEASUREMENT AND MODELING OF COMPUTER SYSTEMS, 2010, 38 (01): : 83 - 94
  • [3] Estimating Bandwidth Requirements Using Flow-Level Measurements
    Schmidt, Ricardo de O.
    Pras, Aiko
    [J]. MANAGING THE DYNAMICS OF NETWORKS AND SERVICES, 2011, 6734 : 169 - 172
  • [4] Towards Bandwidth Estimation Using Flow-Level Measurements
    Schmidt, Ricardo de O.
    Sperotto, Anna
    Sadre, Ramin
    Pras, Aiko
    [J]. DEPENDABLE NETWORKS AND SERVICES, 2012, 7279 : 127 - 138
  • [5] Utility-Optimized Flow-Level Bandwidth Allocationin Hybrid SDNs
    Huang, Xiaohong
    Yuan, Tingting
    Ma, Maode
    [J]. IEEE ACCESS, 2018, 6 : 20279 - 20290
  • [6] Supporting features for flow-level packet analysis towards cyber threat detection: A pilot study
    Ogu, Emmanuel C.
    Ojesanmi, Olusegun A.
    Awodele, Oludele
    Kuyoro, Shade O.
    [J]. International Journal of Information and Computer Security, 2021, 16 (1-2) : 137 - 149
  • [7] OpenFlow-Based Flow-Level Bandwidth Provisioning for CICQ Switches
    Jin, Hao
    Pan, Deng
    Liu, Jason
    Pissinou, Niki
    [J]. IEEE TRANSACTIONS ON COMPUTERS, 2013, 62 (09) : 1799 - 1812
  • [8] μ service level agreement:: A new bandwidth guarantee of flow-level granularity in Internet VPN
    Su, CC
    Lee, WN
    Horng, MF
    Hsu, JP
    Kuo, YH
    [J]. 7TH INTERNATIONAL CONFERENCE ON ADVANCED COMMUNICATION TECHNOLOGY, VOLS 1 AND 2, PROCEEDINGS, 2005, : 324 - 329
  • [9] Packet-Level and Flow-Level Network Intrusion Detection Based on Reinforcement Learning and Adversarial Training
    Yang, Bin
    Arshad, Muhammad Haseeb
    Zhao, Qing
    [J]. ALGORITHMS, 2022, 15 (12)
  • [10] An online flow-level Packet Classification Method on Multi-core Network Processor
    Li Wei
    Yu Xiufen
    [J]. 2015 11TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY (CIS), 2015, : 407 - 411