Bandwidth allocation for multiple qualities of service using generalized processor sharing

G. De Veciana; George Kesidis

doi:10.1109/18.481801

Bandwidth allocation for multiple qualities of service using generalized processor sharing

G. De Veciana, George Kesidis

Research output: Contribution to journal › Article › peer-review

28 Scopus citations

Abstract

We consider the asymptotic behavior of the queue length distribution in segregated buffers sharing a deterministic server via a class of generalized processor sharing (GPS) policies. Such policies have been proposed as a means to guarantee individual quality of service constraints to heterogeneous streams in integrated services digital networks. These results exhibit the manner in which spare capacity is shared by statistically multiplexed traffic streams. The framework corresponds to a natural relaxation of a single GPS node subject to (σ. ρ)-constrained flows where, instead of studying the worst case behavior, we consider statistical bounds on the performance of individual traffic streams.

Original language	English (US)
Pages (from-to)	268-272
Number of pages	5
Journal	IEEE Transactions on Information Theory
Volume	42
Issue number	1
DOIs	https://doi.org/10.1109/18.481801
State	Published - 1996

All Science Journal Classification (ASJC) codes

Information Systems
Computer Science Applications
Library and Information Sciences

Access to Document

10.1109/18.481801

Cite this

@article{80d42e65f3af4f48bd06bb7671622693,

title = "Bandwidth allocation for multiple qualities of service using generalized processor sharing",

abstract = "We consider the asymptotic behavior of the queue length distribution in segregated buffers sharing a deterministic server via a class of generalized processor sharing (GPS) policies. Such policies have been proposed as a means to guarantee individual quality of service constraints to heterogeneous streams in integrated services digital networks. These results exhibit the manner in which spare capacity is shared by statistically multiplexed traffic streams. The framework corresponds to a natural relaxation of a single GPS node subject to (σ. ρ)-constrained flows where, instead of studying the worst case behavior, we consider statistical bounds on the performance of individual traffic streams.",

author = "{De Veciana}, G. and George Kesidis",

note = "Funding Information: When traffic streams are bursty and share a buffered link, (1) no longer has a clear-cut counterpart.{\textquoteright} Indeed, the interactions among traffic streams and resources in such networks is typically not linear G. de Veciana was supported in part by the National Science Foundation under Grant NCR-9409722. The research of G. Kesidis was supported by NSERC of Canada.",

year = "1996",

doi = "10.1109/18.481801",

language = "English (US)",

volume = "42",

pages = "268--272",

journal = "IEEE Transactions on Information Theory",

issn = "0018-9448",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Bandwidth allocation for multiple qualities of service using generalized processor sharing

AU - De Veciana, G.

AU - Kesidis, George

N1 - Funding Information: When traffic streams are bursty and share a buffered link, (1) no longer has a clear-cut counterpart.’ Indeed, the interactions among traffic streams and resources in such networks is typically not linear G. de Veciana was supported in part by the National Science Foundation under Grant NCR-9409722. The research of G. Kesidis was supported by NSERC of Canada.

PY - 1996

Y1 - 1996

N2 - We consider the asymptotic behavior of the queue length distribution in segregated buffers sharing a deterministic server via a class of generalized processor sharing (GPS) policies. Such policies have been proposed as a means to guarantee individual quality of service constraints to heterogeneous streams in integrated services digital networks. These results exhibit the manner in which spare capacity is shared by statistically multiplexed traffic streams. The framework corresponds to a natural relaxation of a single GPS node subject to (σ. ρ)-constrained flows where, instead of studying the worst case behavior, we consider statistical bounds on the performance of individual traffic streams.

AB - We consider the asymptotic behavior of the queue length distribution in segregated buffers sharing a deterministic server via a class of generalized processor sharing (GPS) policies. Such policies have been proposed as a means to guarantee individual quality of service constraints to heterogeneous streams in integrated services digital networks. These results exhibit the manner in which spare capacity is shared by statistically multiplexed traffic streams. The framework corresponds to a natural relaxation of a single GPS node subject to (σ. ρ)-constrained flows where, instead of studying the worst case behavior, we consider statistical bounds on the performance of individual traffic streams.

UR - http://www.scopus.com/inward/record.url?scp=0029754216&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029754216&partnerID=8YFLogxK

U2 - 10.1109/18.481801

DO - 10.1109/18.481801

M3 - Article

AN - SCOPUS:0029754216

SN - 0018-9448

VL - 42

SP - 268

EP - 272

JO - IEEE Transactions on Information Theory

JF - IEEE Transactions on Information Theory

IS - 1

ER -

Bandwidth allocation for multiple qualities of service using generalized processor sharing

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this