VLSI models of network-on-chip interconnect

Dimitrios N. Serpanos, Wayne Wolf

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

We use VLSI circuit models to analyze the relative delay of interconnect subsystems for networks-on-chips (NoCs). Most work in NoCs has selected a network topology based on higher-level performance models, such as packet delay. Our model parameterizes the interconnect subsystem size by N, the number of IP cores (processors, memories, etc.) to be connected. This paper analyzes busses, crossbars, and some multi-stage networks. We compare the delay required transfer a specific amount of information (bits) between two cores. Considering the data transfer parallelism in crossbars, we make 2 different comparisons: (i) transfer between 2 devices, and (ii) parallel transfers between all devices.

Original languageEnglish
Title of host publication2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC
Pages72-77
Number of pages6
DOIs
Publication statusPublished - 1 Dec 2007
Externally publishedYes
Event2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC - Atlanta, GA, United States
Duration: 15 Oct 200717 Oct 2007

Other

Other2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC
CountryUnited States
CityAtlanta, GA
Period15/10/0717/10/07

Fingerprint

VLSI circuits
Data transfer
Topology
Data storage equipment
Network-on-chip
Intellectual property core

ASJC Scopus subject areas

  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Serpanos, D. N., & Wolf, W. (2007). VLSI models of network-on-chip interconnect. In 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC (pp. 72-77). [4402475] https://doi.org/10.1109/VLSISOC.2007.4402475

VLSI models of network-on-chip interconnect. / Serpanos, Dimitrios N.; Wolf, Wayne.

2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC. 2007. p. 72-77 4402475.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Serpanos, DN & Wolf, W 2007, VLSI models of network-on-chip interconnect. in 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC., 4402475, pp. 72-77, 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC, Atlanta, GA, United States, 15/10/07. https://doi.org/10.1109/VLSISOC.2007.4402475
Serpanos DN, Wolf W. VLSI models of network-on-chip interconnect. In 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC. 2007. p. 72-77. 4402475 https://doi.org/10.1109/VLSISOC.2007.4402475
Serpanos, Dimitrios N. ; Wolf, Wayne. / VLSI models of network-on-chip interconnect. 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC. 2007. pp. 72-77
@inproceedings{9918956563ef4884ba7bfc6d4c73bbca,
title = "VLSI models of network-on-chip interconnect",
abstract = "We use VLSI circuit models to analyze the relative delay of interconnect subsystems for networks-on-chips (NoCs). Most work in NoCs has selected a network topology based on higher-level performance models, such as packet delay. Our model parameterizes the interconnect subsystem size by N, the number of IP cores (processors, memories, etc.) to be connected. This paper analyzes busses, crossbars, and some multi-stage networks. We compare the delay required transfer a specific amount of information (bits) between two cores. Considering the data transfer parallelism in crossbars, we make 2 different comparisons: (i) transfer between 2 devices, and (ii) parallel transfers between all devices.",
author = "Serpanos, {Dimitrios N.} and Wayne Wolf",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/VLSISOC.2007.4402475",
language = "English",
isbn = "9781424417100",
pages = "72--77",
booktitle = "2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC",

}

TY - GEN

T1 - VLSI models of network-on-chip interconnect

AU - Serpanos, Dimitrios N.

AU - Wolf, Wayne

PY - 2007/12/1

Y1 - 2007/12/1

N2 - We use VLSI circuit models to analyze the relative delay of interconnect subsystems for networks-on-chips (NoCs). Most work in NoCs has selected a network topology based on higher-level performance models, such as packet delay. Our model parameterizes the interconnect subsystem size by N, the number of IP cores (processors, memories, etc.) to be connected. This paper analyzes busses, crossbars, and some multi-stage networks. We compare the delay required transfer a specific amount of information (bits) between two cores. Considering the data transfer parallelism in crossbars, we make 2 different comparisons: (i) transfer between 2 devices, and (ii) parallel transfers between all devices.

AB - We use VLSI circuit models to analyze the relative delay of interconnect subsystems for networks-on-chips (NoCs). Most work in NoCs has selected a network topology based on higher-level performance models, such as packet delay. Our model parameterizes the interconnect subsystem size by N, the number of IP cores (processors, memories, etc.) to be connected. This paper analyzes busses, crossbars, and some multi-stage networks. We compare the delay required transfer a specific amount of information (bits) between two cores. Considering the data transfer parallelism in crossbars, we make 2 different comparisons: (i) transfer between 2 devices, and (ii) parallel transfers between all devices.

UR - http://www.scopus.com/inward/record.url?scp=50149114527&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50149114527&partnerID=8YFLogxK

U2 - 10.1109/VLSISOC.2007.4402475

DO - 10.1109/VLSISOC.2007.4402475

M3 - Conference contribution

SN - 9781424417100

SP - 72

EP - 77

BT - 2007 IFIP International Conference on Very Large Scale Integration, VLSI-SoC

ER -