TY - GEN
T1 - Alternative decompositions for distributed maximization of network utility
T2 - INFOCOM 2006: 25th IEEE International Conference on Computer Communications
AU - Palomar, Daniel P.
AU - Chiang, Mung
N1 - Copyright:
Copyright 2008 Elsevier B.V., All rights reserved.
PY - 2006
Y1 - 2006
N2 - Network utility maximization (NUM) problems provide an important approach to conduct network resource management and to view layering as optimization decomposition. In the existing literature, distributed implementations are typically achieved by the means of the so-called dual decomposition technique. However, the span of decomposition possibilities includes many other elements which thus far have not been fully exploited, such as the use of the primal decomposition technique, the versatile introduction of auxiliary variables, and the potential of multilevel decompositions. This paper presents a systematic framework to exploit the potential of the alternative decomposition structures as a way to obtain different distributed algorithms, each with a different tradeoff among convergence speed, message passing amount and asymmetry, and distributed computation architecture. Many specific applications are considered to illustrate the proposed framework, including resourceconstrained and direct-control rate allocation, and rate allocation among QoS classes and with multipath routing. For each of these applications, the associated generalized NUM formulation is first presented, followed by the development of novel alternative decompositions and numerical experiments on the resulting new distributed algorithms.
AB - Network utility maximization (NUM) problems provide an important approach to conduct network resource management and to view layering as optimization decomposition. In the existing literature, distributed implementations are typically achieved by the means of the so-called dual decomposition technique. However, the span of decomposition possibilities includes many other elements which thus far have not been fully exploited, such as the use of the primal decomposition technique, the versatile introduction of auxiliary variables, and the potential of multilevel decompositions. This paper presents a systematic framework to exploit the potential of the alternative decomposition structures as a way to obtain different distributed algorithms, each with a different tradeoff among convergence speed, message passing amount and asymmetry, and distributed computation architecture. Many specific applications are considered to illustrate the proposed framework, including resourceconstrained and direct-control rate allocation, and rate allocation among QoS classes and with multipath routing. For each of these applications, the associated generalized NUM formulation is first presented, followed by the development of novel alternative decompositions and numerical experiments on the resulting new distributed algorithms.
KW - Congestion control
KW - Distributed algorithm
KW - Mathematical programming/optimization
KW - Network control by pricing
KW - Network utility maximization
KW - Rate control
KW - Resource allocation
UR - http://www.scopus.com/inward/record.url?scp=33747206316&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33747206316&partnerID=8YFLogxK
U2 - 10.1109/INFOCOM.2006.257
DO - 10.1109/INFOCOM.2006.257
M3 - Conference contribution
AN - SCOPUS:33747206316
SN - 1424402212
SN - 9781424402212
T3 - Proceedings - IEEE INFOCOM
BT - Proceedings - INFOCOM 2006
Y2 - 23 April 2006 through 29 April 2006
ER -