All-to-all broadcasting in torus Network on Chip

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper proposes and evaluates the performance of an all-to-all broadcasting algorithm for a 2D torus Network on Chip (NoC). The proposed algorithm uses special spanning trees called NEWS spanning trees. These trees are link conflict free which implies that the communication steps of the all-to-all algorithm are contention free. The proposed all-to-all broadcasting algorithm is optimal in terms of transmission time and does not need any additional buffer memory like in the all-to-all algorithm for the 2D torus (IEEE Trans Comput 50:1029–1032, 2001). Reducing the amount of buffer space is a very important issue in NoC architectures. Our algorithm is therefore a more efficient solution for all-to-all broadcasting in 2D torus NoC multi-core systems compared to previously proposed algorithms.

Original languageEnglish
JournalJournal of Supercomputing
DOIs
Publication statusAccepted/In press - Mar 21 2015

Fingerprint

Broadcasting
Torus
Spanning tree
Buffer
Contention
Network on chip
Network-on-chip
Efficient Solution
Imply
Data storage equipment
Evaluate
Communication

Keywords

  • 2D torus
  • All-to-all broadcasting
  • Network on chip (NoC)
  • Spanning trees

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Information Systems
  • Theoretical Computer Science

Cite this

All-to-all broadcasting in torus Network on Chip. / Touzene, Abderezak; Day, Khaled.

In: Journal of Supercomputing, 21.03.2015.

Research output: Contribution to journalArticle

@article{535dc540b1b74313b415fa202b04cc41,
title = "All-to-all broadcasting in torus Network on Chip",
abstract = "This paper proposes and evaluates the performance of an all-to-all broadcasting algorithm for a 2D torus Network on Chip (NoC). The proposed algorithm uses special spanning trees called NEWS spanning trees. These trees are link conflict free which implies that the communication steps of the all-to-all algorithm are contention free. The proposed all-to-all broadcasting algorithm is optimal in terms of transmission time and does not need any additional buffer memory like in the all-to-all algorithm for the 2D torus (IEEE Trans Comput 50:1029–1032, 2001). Reducing the amount of buffer space is a very important issue in NoC architectures. Our algorithm is therefore a more efficient solution for all-to-all broadcasting in 2D torus NoC multi-core systems compared to previously proposed algorithms.",
keywords = "2D torus, All-to-all broadcasting, Network on chip (NoC), Spanning trees",
author = "Abderezak Touzene and Khaled Day",
year = "2015",
month = "3",
day = "21",
doi = "10.1007/s11227-015-1406-z",
language = "English",
journal = "Journal of Supercomputing",
issn = "0920-8542",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - All-to-all broadcasting in torus Network on Chip

AU - Touzene, Abderezak

AU - Day, Khaled

PY - 2015/3/21

Y1 - 2015/3/21

N2 - This paper proposes and evaluates the performance of an all-to-all broadcasting algorithm for a 2D torus Network on Chip (NoC). The proposed algorithm uses special spanning trees called NEWS spanning trees. These trees are link conflict free which implies that the communication steps of the all-to-all algorithm are contention free. The proposed all-to-all broadcasting algorithm is optimal in terms of transmission time and does not need any additional buffer memory like in the all-to-all algorithm for the 2D torus (IEEE Trans Comput 50:1029–1032, 2001). Reducing the amount of buffer space is a very important issue in NoC architectures. Our algorithm is therefore a more efficient solution for all-to-all broadcasting in 2D torus NoC multi-core systems compared to previously proposed algorithms.

AB - This paper proposes and evaluates the performance of an all-to-all broadcasting algorithm for a 2D torus Network on Chip (NoC). The proposed algorithm uses special spanning trees called NEWS spanning trees. These trees are link conflict free which implies that the communication steps of the all-to-all algorithm are contention free. The proposed all-to-all broadcasting algorithm is optimal in terms of transmission time and does not need any additional buffer memory like in the all-to-all algorithm for the 2D torus (IEEE Trans Comput 50:1029–1032, 2001). Reducing the amount of buffer space is a very important issue in NoC architectures. Our algorithm is therefore a more efficient solution for all-to-all broadcasting in 2D torus NoC multi-core systems compared to previously proposed algorithms.

KW - 2D torus

KW - All-to-all broadcasting

KW - Network on chip (NoC)

KW - Spanning trees

UR - http://www.scopus.com/inward/record.url?scp=84925436809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925436809&partnerID=8YFLogxK

U2 - 10.1007/s11227-015-1406-z

DO - 10.1007/s11227-015-1406-z

M3 - Article

AN - SCOPUS:84925436809

JO - Journal of Supercomputing

JF - Journal of Supercomputing

SN - 0920-8542

ER -