Efficient nonlinear optimizations of queuing systems

Author

Chiang, Mung ; Sutivong, Arak ; Boyd, Stephen

Author_Institution

Dept. of Electr. Eng., Stanford Univ., CA, USA

Volume

3

fYear

2002

fDate

17-21 Nov. 2002

Firstpage

2425

Abstract

We present a systematic treatment of efficient nonlinear optimizations of queuing systems. The suite of formulations uses the computational tool of convex optimization, with fast polynomial time algorithms to obtain the global optimum for these nonlinear problems under various constraints. We first show convexity structures of several queuing systems, including some surprising transition patterns, followed by formulating and showing numerical examples of several convex performance optimizations for both single queues and queuing networks. Blocking probability minimization and service rate allocation through the effective bandwidth approach is also presented.

Keywords

optimisation; polynomials; probability; queueing theory; blocking probability minimization; communications data networks; computer multiprocessor networks; convex performance optimization; convexity structures; efficient nonlinear optimization; fast polynomial time algorithms; global optimum; nonlinear problems; queuing networks; queuing systems; service rate allocation; transition patterns; Bandwidth; Communication switching; Computer networks; Constraint optimization; Functional programming; Measurement; Packet switching; Polynomials; Quality of service; Switches;

fLanguage

English

Publisher

ieee

Conference_Titel

Global Telecommunications Conference, 2002. GLOBECOM '02. IEEE

Print_ISBN

0-7803-7632-3

Type

conf

DOI

10.1109/GLOCOM.2002.1189066

Filename

1189066