On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods

William J. Stewart; A. Touzene

On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods

William J. Stewart, A. Touzene

Computer Science

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

2 Citations (Scopus)

Abstract

Iterative aggregation/disaggregation (IAD) methods are powerful tools for solving Markov chain models whose transition probability matrices are nearly completely decomposable (NCD). Such models arise frequently during the performance and reliability analysis of computer and telecommunication systems. IAD methods require the solution of a stochastic coupling matrix whose elements denote transition probabilities among blocks. The coupling matrices are often large and in NCD models necessarily have diagonal elements close to one and small off-diagonal elements. This makes their solution by either iterative or direct methods rather difficult. In this paper we propose a modification of the coupling matrix that allows us to accurate and efficiently compute its stationary probability vector.

Original language	English
Title of host publication	Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems
Editors	Vijay Madisetti, Erol Gelenbe, Jean Walrand
Publisher	Publ by IEEE
Pages	255-262
Number of pages	8
ISBN (Print)	0818652926
Publication status	Published - 1994
Event	Proceedings of the 2nd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - Durham, NC, USA Duration: Jan 31 1994 → Feb 2 1994

Other

Other	Proceedings of the 2nd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems
City	Durham, NC, USA
Period	1/31/94 → 2/2/94

ASJC Scopus subject areas

Engineering(all)

Access to Document

Fingerprint Dive into the research topics of 'On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods'. Together they form a unique fingerprint.

Cite this

Stewart, W. J., & Touzene, A. (1994). On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods. In V. Madisetti, E. Gelenbe, & J. Walrand (Eds.), Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (pp. 255-262). Publ by IEEE.

On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods. / Stewart, William J.; Touzene, A.

Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems. ed. / Vijay Madisetti; Erol Gelenbe; Jean Walrand. Publ by IEEE, 1994. p. 255-262.

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

Stewart, WJ & Touzene, A 1994, On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods. in V Madisetti, E Gelenbe & J Walrand (eds), Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems. Publ by IEEE, pp. 255-262, Proceedings of the 2nd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, Durham, NC, USA, 1/31/94.

@inproceedings{9830ba076c10440d87c7c3f8b566eedc,

title = "On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods",

abstract = "Iterative aggregation/disaggregation (IAD) methods are powerful tools for solving Markov chain models whose transition probability matrices are nearly completely decomposable (NCD). Such models arise frequently during the performance and reliability analysis of computer and telecommunication systems. IAD methods require the solution of a stochastic coupling matrix whose elements denote transition probabilities among blocks. The coupling matrices are often large and in NCD models necessarily have diagonal elements close to one and small off-diagonal elements. This makes their solution by either iterative or direct methods rather difficult. In this paper we propose a modification of the coupling matrix that allows us to accurate and efficiently compute its stationary probability vector.",

author = "Stewart, {William J.} and A. Touzene",

year = "1994",

language = "English",

isbn = "0818652926",

pages = "255--262",

editor = "Vijay Madisetti and Erol Gelenbe and Jean Walrand",

booktitle = "Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems",

publisher = "Publ by IEEE",

note = "Proceedings of the 2nd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems ; Conference date: 31-01-1994 Through 02-02-1994",

}

TY - GEN

T1 - On solving stochastic coupling matrices arising in iterative aggregation/disaggregation methods

AU - Stewart, William J.

AU - Touzene, A.

PY - 1994

Y1 - 1994

N2 - Iterative aggregation/disaggregation (IAD) methods are powerful tools for solving Markov chain models whose transition probability matrices are nearly completely decomposable (NCD). Such models arise frequently during the performance and reliability analysis of computer and telecommunication systems. IAD methods require the solution of a stochastic coupling matrix whose elements denote transition probabilities among blocks. The coupling matrices are often large and in NCD models necessarily have diagonal elements close to one and small off-diagonal elements. This makes their solution by either iterative or direct methods rather difficult. In this paper we propose a modification of the coupling matrix that allows us to accurate and efficiently compute its stationary probability vector.

AB - Iterative aggregation/disaggregation (IAD) methods are powerful tools for solving Markov chain models whose transition probability matrices are nearly completely decomposable (NCD). Such models arise frequently during the performance and reliability analysis of computer and telecommunication systems. IAD methods require the solution of a stochastic coupling matrix whose elements denote transition probabilities among blocks. The coupling matrices are often large and in NCD models necessarily have diagonal elements close to one and small off-diagonal elements. This makes their solution by either iterative or direct methods rather difficult. In this paper we propose a modification of the coupling matrix that allows us to accurate and efficiently compute its stationary probability vector.

UR - http://www.scopus.com/inward/record.url?scp=0028026503&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028026503&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0028026503

SN - 0818652926

SP - 255

EP - 262

BT - Proceedings of the IEEE International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems

A2 - Madisetti, Vijay

A2 - Gelenbe, Erol

A2 - Walrand, Jean

PB - Publ by IEEE

T2 - Proceedings of the 2nd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems

Y2 - 31 January 1994 through 2 February 1994

ER -