elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

A DTMC Model for Performance Evaluation of Irregular Interconnection Networks with Asymmetric Spatial Traffic Distributions

Lüdtke, Daniel and Tutsch, Dietmar (2016) A DTMC Model for Performance Evaluation of Irregular Interconnection Networks with Asymmetric Spatial Traffic Distributions. In: Lecture Notes in Computer Science, 9845, pp. 193-209. Springer International Publishing Switzerland. Analytical and Stochastic Modelling Techniques and Applications ASMTA 2016, 23.-26. Aug. 2016, Cardiff, UK. DOI: 10.1007/978-3-319-43904-4_14 ISBN 978-3-319-43903-7

[img] PDF (Author's Version)
745kB

Official URL: http://dx.doi.org/10.1007/978-3-319-43904-4_14

Abstract

Several mathematical models have been proposed to evaluate the performance of interconnection networks used for high-speed connections for supercomputers, switches and routers for local and wide area networks, as well as networks on a chip. Often these models are based on state space reduction by exploiting symmetries of the network and requiring uniform traffic patterns. If an interconnection network is built for a specific application with non-uniform spatial traffic distribution, models that are more general are needed. This paper proposes a mathematical model for performance evaluation of application-specific interconnection networks based on inhomogeneous discrete time Markov chains (DTMC). It supports store and forward routing, irregular network topologies, and asymmetric spatial traffic distributions. The model is described in a generalized way so that it can support arbitrary switching element sizes within the network and its input buffers.

Item URL in elib:https://elib.dlr.de/106669/
Document Type:Conference or Workshop Item (Speech)
Title:A DTMC Model for Performance Evaluation of Irregular Interconnection Networks with Asymmetric Spatial Traffic Distributions
Authors:
AuthorsInstitution or Email of AuthorsAuthors ORCID iD
Lüdtke, DanielDaniel.Luedtke (at) dlr.dehttps://orcid.org/0000-0002-6758-1562
Tutsch, Dietmartutsch (at) uni-wuppertal.deUNSPECIFIED
Date:4 August 2016
Journal or Publication Title:Lecture Notes in Computer Science
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:Yes
Volume:9845
DOI :10.1007/978-3-319-43904-4_14
Page Range:pp. 193-209
Editors:
EditorsEmail
Wittevrongel, SabineUNSPECIFIED
Phung-Duc, TuanUNSPECIFIED
Publisher:Springer International Publishing Switzerland
Series Name:Lecture Notes in Computer Science
ISBN:978-3-319-43903-7
Status:Published
Keywords:network topology, packet-switching networks, store and forward networks, modeling techniques, performance attributes, distribution functions, Markov processes, Performance, fixed-point iteration, traffic model, mathematical model
Event Title:Analytical and Stochastic Modelling Techniques and Applications ASMTA 2016
Event Location:Cardiff, UK
Event Type:international Conference
Event Dates:23.-26. Aug. 2016
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Space Technology
DLR - Research area:Raumfahrt
DLR - Program:R SY - Technik für Raumfahrtsysteme
DLR - Research theme (Project):R - Scosa Onboard Computing
Location: Braunschweig
Institutes and Institutions:Institut of Simulation and Software Technology > Software for Space Systems and Interactive Visualisation
Deposited By: Lüdtke, Daniel
Deposited On:25 Oct 2016 10:23
Last Modified:31 Jul 2019 20:04

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.