Modeling and predicting end-to-end response times in multi-tier internet applications

Sandjai Bhulai; Swaminathan Sivasubramanian; Rob van der Mei; Maarten van Steen

Modeling and predicting end-to-end response times in multi-tier internet applications

Sandjai Bhulai^*, Swaminathan Sivasubramanian, Rob van der Mei, Maarten van Steen

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

11 Citations (Scopus)

Abstract

Many Internet applications employ multi-tier software architectures. The performance of such multi-tier Internet applications is typically measured by the end-to-end response times. Most of the earlier works in modeling the response times of such systems have limited their study to modeling the mean. However, since the user-perceived performance is highly influenced by the variability in response times, the variance of the response times is important as well. We first develop a simple model for the end-to-end response times for multi-tiered Internet applications. We validate the model by real data from two large-scale applications that are widely deployed on the Internet. Second, we derive exact and approximate expressions for the mean and the variance, respectively, of the end-to-end response times. Extensive numerical validation shows that the approximations match very well with simulations. These observations make the results presented highly useful for capacity planning and performance prediction of large-scale multi-tiered Internet applications.

Original language	English
Title of host publication	Managing Traffic Performance in Converged Networks
Subtitle of host publication	20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings
Editors	Lorne Mason, Tadeusz Drwiega, James Yan
Place of Publication	Berlin, Heidelberg
Publisher	Springer
Pages	519-532
Number of pages	14
ISBN (Electronic)	978-3-540-72990-7
ISBN (Print)	978-3-540-72989-1
Publication status	Published - 1 Dec 2007
Externally published	Yes
Event	20th International Teletraffic Congress 2007 - Ottawa, Canada Duration: 17 Jun 2007 → 21 Jun 2007 Conference number: 20

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	4516
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	20th International Teletraffic Congress 2007
Abbreviated title	ITC 2007
Country/Territory	Canada
City	Ottawa
Period	17/06/07 → 21/06/07

Keywords

End-to-end response times
Internet services
Multi-tier architectures
Variance approximation

Cite this

Bhulai, S., Sivasubramanian, S., van der Mei, R., & van Steen, M. (2007). Modeling and predicting end-to-end response times in multi-tier internet applications. In L. Mason, T. Drwiega, & J. Yan (Eds.), Managing Traffic Performance in Converged Networks: 20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings (pp. 519-532). (Lecture Notes in Computer Science; Vol. 4516). Springer.

Bhulai, Sandjai ; Sivasubramanian, Swaminathan ; van der Mei, Rob et al. / Modeling and predicting end-to-end response times in multi-tier internet applications. Managing Traffic Performance in Converged Networks: 20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings. editor / Lorne Mason ; Tadeusz Drwiega ; James Yan. Berlin, Heidelberg : Springer, 2007. pp. 519-532 (Lecture Notes in Computer Science).

@inproceedings{62bccc44bd164ea49f910f8750472bba,

title = "Modeling and predicting end-to-end response times in multi-tier internet applications",

abstract = "Many Internet applications employ multi-tier software architectures. The performance of such multi-tier Internet applications is typically measured by the end-to-end response times. Most of the earlier works in modeling the response times of such systems have limited their study to modeling the mean. However, since the user-perceived performance is highly influenced by the variability in response times, the variance of the response times is important as well. We first develop a simple model for the end-to-end response times for multi-tiered Internet applications. We validate the model by real data from two large-scale applications that are widely deployed on the Internet. Second, we derive exact and approximate expressions for the mean and the variance, respectively, of the end-to-end response times. Extensive numerical validation shows that the approximations match very well with simulations. These observations make the results presented highly useful for capacity planning and performance prediction of large-scale multi-tiered Internet applications.",

keywords = "End-to-end response times, Internet services, Multi-tier architectures, Variance approximation",

author = "Sandjai Bhulai and Swaminathan Sivasubramanian and {van der Mei}, Rob and {van Steen}, Maarten",

year = "2007",

month = dec,

day = "1",

language = "English",

isbn = "978-3-540-72989-1",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "519--532",

editor = "Lorne Mason and Tadeusz Drwiega and James Yan",

booktitle = "Managing Traffic Performance in Converged Networks",

address = "Germany",

note = "20th International Teletraffic Congress 2007, ITC 2007 ; Conference date: 17-06-2007 Through 21-06-2007",

}

Bhulai, S, Sivasubramanian, S, van der Mei, R & van Steen, M 2007, Modeling and predicting end-to-end response times in multi-tier internet applications. in L Mason, T Drwiega & J Yan (eds), Managing Traffic Performance in Converged Networks: 20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings. Lecture Notes in Computer Science, vol. 4516, Springer, Berlin, Heidelberg, pp. 519-532, 20th International Teletraffic Congress 2007, Ottawa, Canada, 17/06/07.

Modeling and predicting end-to-end response times in multi-tier internet applications. / Bhulai, Sandjai; Sivasubramanian, Swaminathan; van der Mei, Rob et al.
Managing Traffic Performance in Converged Networks: 20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings. ed. / Lorne Mason; Tadeusz Drwiega; James Yan. Berlin, Heidelberg: Springer, 2007. p. 519-532 (Lecture Notes in Computer Science; Vol. 4516).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Modeling and predicting end-to-end response times in multi-tier internet applications

AU - Bhulai, Sandjai

AU - Sivasubramanian, Swaminathan

AU - van der Mei, Rob

AU - van Steen, Maarten

N1 - Conference code: 20

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Many Internet applications employ multi-tier software architectures. The performance of such multi-tier Internet applications is typically measured by the end-to-end response times. Most of the earlier works in modeling the response times of such systems have limited their study to modeling the mean. However, since the user-perceived performance is highly influenced by the variability in response times, the variance of the response times is important as well. We first develop a simple model for the end-to-end response times for multi-tiered Internet applications. We validate the model by real data from two large-scale applications that are widely deployed on the Internet. Second, we derive exact and approximate expressions for the mean and the variance, respectively, of the end-to-end response times. Extensive numerical validation shows that the approximations match very well with simulations. These observations make the results presented highly useful for capacity planning and performance prediction of large-scale multi-tiered Internet applications.

AB - Many Internet applications employ multi-tier software architectures. The performance of such multi-tier Internet applications is typically measured by the end-to-end response times. Most of the earlier works in modeling the response times of such systems have limited their study to modeling the mean. However, since the user-perceived performance is highly influenced by the variability in response times, the variance of the response times is important as well. We first develop a simple model for the end-to-end response times for multi-tiered Internet applications. We validate the model by real data from two large-scale applications that are widely deployed on the Internet. Second, we derive exact and approximate expressions for the mean and the variance, respectively, of the end-to-end response times. Extensive numerical validation shows that the approximations match very well with simulations. These observations make the results presented highly useful for capacity planning and performance prediction of large-scale multi-tiered Internet applications.

KW - End-to-end response times

KW - Internet services

KW - Multi-tier architectures

KW - Variance approximation

UR - http://www.scopus.com/inward/record.url?scp=38149008147&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:38149008147

SN - 978-3-540-72989-1

T3 - Lecture Notes in Computer Science

SP - 519

EP - 532

BT - Managing Traffic Performance in Converged Networks

A2 - Mason, Lorne

A2 - Drwiega, Tadeusz

A2 - Yan, James

PB - Springer

CY - Berlin, Heidelberg

T2 - 20th International Teletraffic Congress 2007

Y2 - 17 June 2007 through 21 June 2007

ER -

Bhulai S, Sivasubramanian S, van der Mei R, van Steen M. Modeling and predicting end-to-end response times in multi-tier internet applications. In Mason L, Drwiega T, Yan J, editors, Managing Traffic Performance in Converged Networks: 20th International Teletraffic Congress, ITC20 2007, Ottawa, Canada, June 17-21, 2007. Proceedings. Berlin, Heidelberg: Springer. 2007. p. 519-532. (Lecture Notes in Computer Science).

Modeling and predicting end-to-end response times in multi-tier internet applications

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this