Queen: Estimating Packet Loss Rate between Arbitrary Internet Hosts

  • Y. Angela Wang
  • Cheng Huang
  • Jin Li
  • Keith W. Ross
Conference paper

DOI: 10.1007/978-3-642-00975-4_6

Part of the Lecture Notes in Computer Science book series (LNCS, volume 5448)
Cite this paper as:
Wang Y.A., Huang C., Li J., Ross K.W. (2009) Queen: Estimating Packet Loss Rate between Arbitrary Internet Hosts. In: Moon S.B., Teixeira R., Uhlig S. (eds) Passive and Active Network Measurement. PAM 2009. Lecture Notes in Computer Science, vol 5448. Springer, Berlin, Heidelberg

Abstract

Estimate of packet-loss rates between arbitrary Internet hosts is critical for many large-scale distributed applications, including overlay routing, P2P media streaming, VoIP, and edge-server location in CDNs. iPlane has been recently proposed to estimate delay, packet-loss rates, and bandwidth between arbitrary hosts [1]. To our knowledge, iPlane is the only published technique for estimating loss rates between arbitrary Internet hosts. In this paper, we present Queen, a new methodology for estimating packet-loss rates between arbitrary hosts. Queen, extending the King [2] methodology for estimating delay, takes advantage of the open recursive DNS name servers. Queen requires neither additional infrastructure deployment nor control of the DNS recursive servers. After describing the methodology, we present an extensive measurement validation of Queen’s accuracy. Our validation shows that Queen’s accuracy is reasonably high and, in particular, significantly better than that of iPlane for packet-loss rate estimation.

Keywords

Recursive DNS Retransmission Pattern Loss Rate 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Y. Angela Wang
    • 1
  • Cheng Huang
    • 2
  • Jin Li
    • 2
  • Keith W. Ross
    • 1
  1. 1.Polytechnic Institute of NYUBrooklynUSA
  2. 2.Microsoft ResearchRedmondUSA

Personalised recommendations