Abstract
Knowledge of link capacity is important for network design, management, and utilization. With the increasing popularity of asymmetric link technologies (such as DSL, 1xRTT, and satellite links), it is desirable to have a capacity estimation technique, which can simultaneously measure forward and backward direction link capacities on an Internet path. Moreover, this estimation must often be "sender only", because of receiver limitations or lack of standards. In this study, we propose a simple, fast and accurate technique, called AsymProbe, to estimate asymmetric link capacities. AsymProbe is a "sender only", round trip procedure. It achieves asymmetric link capacity estimation by strategically altering the ratio of probe and acknowledgement packet sizes. Using simulation and testbed experiments, we validate AsymProbe with a variety of network configurations. The results show that AsymProbe can correctly estimate the asymmetric link capacities as long as an appropriate packet size ratio can be employed.
| Original language | English |
|---|---|
| Pages (from-to) | 780-791 |
| Number of pages | 12 |
| Journal | Lecture Notes in Computer Science |
| Volume | 3462 |
| Publication status | Published - 2005 Sep 26 |
| Event | 4th International IFIP-TC6 Networking Conference: Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile and Wireless Communications Systems, NETWORKING 2005 - Waterloo, Ont., Canada Duration: 2005 May 2 → 2005 May 6 |
Fingerprint
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)
Cite this
End-to-end asymmetric link capacity estimation. / Chen, Ling Jyh; Sun, Tony; Yang, Guang; Sanadidi, M. Y.; Gerla, Mario.
In: Lecture Notes in Computer Science, Vol. 3462, 26.09.2005, p. 780-791.Research output: Contribution to journal › Conference article
}
TY - JOUR
T1 - End-to-end asymmetric link capacity estimation
AU - Chen, Ling Jyh
AU - Sun, Tony
AU - Yang, Guang
AU - Sanadidi, M. Y.
AU - Gerla, Mario
PY - 2005/9/26
Y1 - 2005/9/26
N2 - Knowledge of link capacity is important for network design, management, and utilization. With the increasing popularity of asymmetric link technologies (such as DSL, 1xRTT, and satellite links), it is desirable to have a capacity estimation technique, which can simultaneously measure forward and backward direction link capacities on an Internet path. Moreover, this estimation must often be "sender only", because of receiver limitations or lack of standards. In this study, we propose a simple, fast and accurate technique, called AsymProbe, to estimate asymmetric link capacities. AsymProbe is a "sender only", round trip procedure. It achieves asymmetric link capacity estimation by strategically altering the ratio of probe and acknowledgement packet sizes. Using simulation and testbed experiments, we validate AsymProbe with a variety of network configurations. The results show that AsymProbe can correctly estimate the asymmetric link capacities as long as an appropriate packet size ratio can be employed.
AB - Knowledge of link capacity is important for network design, management, and utilization. With the increasing popularity of asymmetric link technologies (such as DSL, 1xRTT, and satellite links), it is desirable to have a capacity estimation technique, which can simultaneously measure forward and backward direction link capacities on an Internet path. Moreover, this estimation must often be "sender only", because of receiver limitations or lack of standards. In this study, we propose a simple, fast and accurate technique, called AsymProbe, to estimate asymmetric link capacities. AsymProbe is a "sender only", round trip procedure. It achieves asymmetric link capacity estimation by strategically altering the ratio of probe and acknowledgement packet sizes. Using simulation and testbed experiments, we validate AsymProbe with a variety of network configurations. The results show that AsymProbe can correctly estimate the asymmetric link capacities as long as an appropriate packet size ratio can be employed.
UR - http://www.scopus.com/inward/record.url?scp=24944563523&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=24944563523&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:24944563523
VL - 3462
SP - 780
EP - 791
JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SN - 0302-9743
ER -