Abstract
We propose a variable step fluid model for communication network in this paper. Our main goal in this research is simulation speedup of a packet-level simulator while maintaining the accuracy. The variable step fluid model not only reduces complexity but also accurately estimates simulation details such as round trip time, queue sizes, TCP windows, and packet drops. In addition, the variable step fluid model reduces event explosions, ripple effects, which have been observed in the traditional fluid models. We validate our model against ns-2 simulation with a mixture of TCP and UDP flows under various background traffic scenarios. Our model achieves significant speedup compared to packet-level simulators. For example, the speedup of our fluid model for 20 Mb bottleneck is 40 to 70 against ns-2.
| Original language | English |
|---|---|
| Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Pages | 87-97 |
| Number of pages | 11 |
| Volume | 3976 LNCS |
| Publication status | Published - 2006 Jul 19 |
| Event | 5th International IFIP-TC6 Networking Conference, Networking 2006 - Coimbra, Portugal Duration: 2006 May 15 → 2006 May 19 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 3976 LNCS |
| ISSN (Print) | 03029743 |
| ISSN (Electronic) | 16113349 |
Other
| Other | 5th International IFIP-TC6 Networking Conference, Networking 2006 |
|---|---|
| Country | Portugal |
| City | Coimbra |
| Period | 06/5/15 → 06/5/19 |
Fingerprint
ASJC Scopus subject areas
- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science
Cite this
Variable step fluid simulation for communication network. / Kim, Hongjoong; Lee, Junsoo.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 3976 LNCS 2006. p. 87-97 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3976 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Variable step fluid simulation for communication network
AU - Kim, Hongjoong
AU - Lee, Junsoo
PY - 2006/7/19
Y1 - 2006/7/19
N2 - We propose a variable step fluid model for communication network in this paper. Our main goal in this research is simulation speedup of a packet-level simulator while maintaining the accuracy. The variable step fluid model not only reduces complexity but also accurately estimates simulation details such as round trip time, queue sizes, TCP windows, and packet drops. In addition, the variable step fluid model reduces event explosions, ripple effects, which have been observed in the traditional fluid models. We validate our model against ns-2 simulation with a mixture of TCP and UDP flows under various background traffic scenarios. Our model achieves significant speedup compared to packet-level simulators. For example, the speedup of our fluid model for 20 Mb bottleneck is 40 to 70 against ns-2.
AB - We propose a variable step fluid model for communication network in this paper. Our main goal in this research is simulation speedup of a packet-level simulator while maintaining the accuracy. The variable step fluid model not only reduces complexity but also accurately estimates simulation details such as round trip time, queue sizes, TCP windows, and packet drops. In addition, the variable step fluid model reduces event explosions, ripple effects, which have been observed in the traditional fluid models. We validate our model against ns-2 simulation with a mixture of TCP and UDP flows under various background traffic scenarios. Our model achieves significant speedup compared to packet-level simulators. For example, the speedup of our fluid model for 20 Mb bottleneck is 40 to 70 against ns-2.
UR - http://www.scopus.com/inward/record.url?scp=33745878534&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745878534&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:33745878534
SN - 3540341927
SN - 9783540341925
VL - 3976 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 87
EP - 97
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ER -