Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions

Derek Doran; Lester Lipsky; Steve Thompson

doi:10.1109/NCA.2010.11

Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions

Derek Doran, Lester Lipsky, Steve Thompson

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

An analytic cost model is presented for M/G/1/N queueing systems. It considers the cost of customer loss versus customer delays, by varying buffer size and processor speed. We find optimal (and near optimal) configurations for a wide variety of service-time distributions. The model can provide insight into when it is better to invest in increased processor speed than to supply more buffer space. It is seen that different distributions may need very different hardware for optimal performance, and that it may actually be better to reject customers.

Original language	American English
Title of host publication	2010 Ninth IEEE International Symposium on Network Computing and Applications
Publisher	IEEE
Pages	28-35
Number of pages	8
ISBN (Electronic)	978-1-4244-7627-5 , 978-0-7695-4118-1
ISBN (Print)	978-1-4244-7628-2
DOIs	https://doi.org/10.1109/NCA.2010.11
State	Published - 2010
Event	2010 9th IEEE International Symposium on Network Computing and Applications, NCA 2010 - Cambridge, MA, United States Duration: Jul 15 2010 → Jul 17 2010

Conference

Conference	2010 9th IEEE International Symposium on Network Computing and Applications, NCA 2010
Country/Territory	United States
City	Cambridge, MA
Period	7/15/10 → 7/17/10

ASJC Scopus Subject Areas

Computer Networks and Communications
Computer Science Applications

Keywords

Servers
Shape
Organizations
Equations
Queueing analysis
Optimized production technology
Cost function

Disciplines

Computer Sciences
Engineering

Access to Document

10.1109/NCA.2010.11License: Unspecified

Cite this

Doran, D, Lipsky, L & Thompson, S 2010, Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions. in 2010 Ninth IEEE International Symposium on Network Computing and Applications., 5598233, IEEE, pp. 28-35, 2010 9th IEEE International Symposium on Network Computing and Applications, NCA 2010, Cambridge, MA, United States, 7/15/10. https://doi.org/10.1109/NCA.2010.11

@inproceedings{8d57ad85ff0141b19f171d996eb29f5d,

title = "Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions",

abstract = " An analytic cost model is presented for M/G/1/N queueing systems. It considers the cost of customer loss versus customer delays, by varying buffer size and processor speed. We find optimal (and near optimal) configurations for a wide variety of service-time distributions. The model can provide insight into when it is better to invest in increased processor speed than to supply more buffer space. It is seen that different distributions may need very different hardware for optimal performance, and that it may actually be better to reject customers.",

keywords = "Servers, Shape, Organizations, Equations, Queueing analysis, Optimized production technology, Cost function",

author = "Derek Doran and Lester Lipsky and Steve Thompson",

year = "2010",

doi = "10.1109/NCA.2010.11",

language = "American English",

isbn = "978-1-4244-7628-2",

pages = "28--35",

booktitle = "2010 Ninth IEEE International Symposium on Network Computing and Applications",

publisher = "IEEE",

note = "2010 9th IEEE International Symposium on Network Computing and Applications, NCA 2010 ; Conference date: 15-07-2010 Through 17-07-2010",

}

TY - GEN

T1 - Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions

AU - Doran, Derek

AU - Lipsky, Lester

AU - Thompson, Steve

PY - 2010

Y1 - 2010

N2 - An analytic cost model is presented for M/G/1/N queueing systems. It considers the cost of customer loss versus customer delays, by varying buffer size and processor speed. We find optimal (and near optimal) configurations for a wide variety of service-time distributions. The model can provide insight into when it is better to invest in increased processor speed than to supply more buffer space. It is seen that different distributions may need very different hardware for optimal performance, and that it may actually be better to reject customers.

AB - An analytic cost model is presented for M/G/1/N queueing systems. It considers the cost of customer loss versus customer delays, by varying buffer size and processor speed. We find optimal (and near optimal) configurations for a wide variety of service-time distributions. The model can provide insight into when it is better to invest in increased processor speed than to supply more buffer space. It is seen that different distributions may need very different hardware for optimal performance, and that it may actually be better to reject customers.

KW - Servers

KW - Shape

KW - Organizations

KW - Equations

KW - Queueing analysis

KW - Optimized production technology

KW - Cost function

UR - http://www.scopus.com/inward/record.url?scp=78449299936&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78449299936&partnerID=8YFLogxK

UR - https://corescholar.libraries.wright.edu/cse/259

U2 - 10.1109/NCA.2010.11

DO - 10.1109/NCA.2010.11

M3 - Conference contribution

SN - 978-1-4244-7628-2

SP - 28

EP - 35

BT - 2010 Ninth IEEE International Symposium on Network Computing and Applications

PB - IEEE

T2 - 2010 9th IEEE International Symposium on Network Computing and Applications, NCA 2010

Y2 - 15 July 2010 through 17 July 2010

ER -

Cost-Based Optimization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions

Abstract

Conference

ASJC Scopus Subject Areas

Keywords

Disciplines

Access to Document

Other files and links

Cite this