Buffer depth and traffic influence on 3D NoCs performance

Yan Ghidini; Thais Webber; Edson Moreno; Fernando Grando; Rubem Fagundes; César Marcon

doi:10.1109/RSP.2012.6380684

Buffer depth and traffic influence on 3D NoCs performance

Yan Ghidini^*, Thais Webber, Edson Moreno, Fernando Grando, Rubem Fagundes, César Marcon

^*Corresponding author for this work

School of Computer Science

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

Abstract

3D NoC-based architectures have emerged to reduce the network latency, the energy consumption and total area in comparison to 2D NoC topologies. However, they are characterized by various trade-offs with regard to the three dimensional structure and its performance specifications. In this paper, we present a 3D NoC mesh architecture called Lasio, whose latency and the throughput achieved, for both network and application, are evaluated considering two types of traffic patterns, varied buffer depth and a range of packet sizes. Cycle-accurate simulations demonstrated that there is a high impact of buffer depth and packet size on the NoC latency and on the application latency. Applying an appropriate buffer depth, for several sizes of packets, the application latency is reduced and throughput is increased.

Original language	English
Title of host publication	Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping
Subtitle of host publication	Shortening the Path from Specification to Prototype, RSP 2012
Pages	9-15
Number of pages	7
DOIs	https://doi.org/10.1109/RSP.2012.6380684
Publication status	Published - 2012
Event	23rd IEEE International Symposium on Rapid System Prototyping, RSP 2012 - Tampere, Finland Duration: 11 Oct 2012 → 12 Oct 2012

Publication series

Name	Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP
ISSN (Print)	2150-5500
ISSN (Electronic)	2150-5519

Conference

Conference	23rd IEEE International Symposium on Rapid System Prototyping, RSP 2012
Country/Territory	Finland
City	Tampere
Period	11/10/12 → 12/10/12

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

3D NoC
Buffer depth
Latency
Throughput

Access to Document

10.1109/RSP.2012.6380684

Cite this

Ghidini, Y., Webber, T., Moreno, E., Grando, F., Fagundes, R., & Marcon, C. (2012). Buffer depth and traffic influence on 3D NoCs performance. In Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 2012 (pp. 9-15). Article 6380684 (Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP). https://doi.org/10.1109/RSP.2012.6380684

@inproceedings{4564eb63950f46af88e4bfd33530726d,

title = "Buffer depth and traffic influence on 3D NoCs performance",

abstract = "3D NoC-based architectures have emerged to reduce the network latency, the energy consumption and total area in comparison to 2D NoC topologies. However, they are characterized by various trade-offs with regard to the three dimensional structure and its performance specifications. In this paper, we present a 3D NoC mesh architecture called Lasio, whose latency and the throughput achieved, for both network and application, are evaluated considering two types of traffic patterns, varied buffer depth and a range of packet sizes. Cycle-accurate simulations demonstrated that there is a high impact of buffer depth and packet size on the NoC latency and on the application latency. Applying an appropriate buffer depth, for several sizes of packets, the application latency is reduced and throughput is increased.",

keywords = "3D NoC, Buffer depth, Latency, Throughput",

author = "Yan Ghidini and Thais Webber and Edson Moreno and Fernando Grando and Rubem Fagundes and C{\'e}sar Marcon",

year = "2012",

doi = "10.1109/RSP.2012.6380684",

language = "English",

isbn = "9781467327862",

series = "Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP",

pages = "9--15",

booktitle = "Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping",

}

Ghidini, Y, Webber, T, Moreno, E, Grando, F, Fagundes, R & Marcon, C 2012, Buffer depth and traffic influence on 3D NoCs performance. in Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 2012., 6380684, Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP, pp. 9-15, 23rd IEEE International Symposium on Rapid System Prototyping, RSP 2012, Tampere, Finland, 11/10/12. https://doi.org/10.1109/RSP.2012.6380684

Buffer depth and traffic influence on 3D NoCs performance. / Ghidini, Yan; Webber, Thais; Moreno, Edson et al.
Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 2012. 2012. p. 9-15 6380684 (Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP).

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

TY - GEN

T1 - Buffer depth and traffic influence on 3D NoCs performance

AU - Ghidini, Yan

AU - Webber, Thais

AU - Moreno, Edson

AU - Grando, Fernando

AU - Fagundes, Rubem

AU - Marcon, César

PY - 2012

Y1 - 2012

N2 - 3D NoC-based architectures have emerged to reduce the network latency, the energy consumption and total area in comparison to 2D NoC topologies. However, they are characterized by various trade-offs with regard to the three dimensional structure and its performance specifications. In this paper, we present a 3D NoC mesh architecture called Lasio, whose latency and the throughput achieved, for both network and application, are evaluated considering two types of traffic patterns, varied buffer depth and a range of packet sizes. Cycle-accurate simulations demonstrated that there is a high impact of buffer depth and packet size on the NoC latency and on the application latency. Applying an appropriate buffer depth, for several sizes of packets, the application latency is reduced and throughput is increased.

AB - 3D NoC-based architectures have emerged to reduce the network latency, the energy consumption and total area in comparison to 2D NoC topologies. However, they are characterized by various trade-offs with regard to the three dimensional structure and its performance specifications. In this paper, we present a 3D NoC mesh architecture called Lasio, whose latency and the throughput achieved, for both network and application, are evaluated considering two types of traffic patterns, varied buffer depth and a range of packet sizes. Cycle-accurate simulations demonstrated that there is a high impact of buffer depth and packet size on the NoC latency and on the application latency. Applying an appropriate buffer depth, for several sizes of packets, the application latency is reduced and throughput is increased.

KW - 3D NoC

KW - Buffer depth

KW - Latency

KW - Throughput

UR - https://www.scopus.com/pages/publications/84873465537

U2 - 10.1109/RSP.2012.6380684

DO - 10.1109/RSP.2012.6380684

M3 - Conference contribution

AN - SCOPUS:84873465537

SN - 9781467327862

T3 - Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP

SP - 9

EP - 15

BT - Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping

T2 - 23rd IEEE International Symposium on Rapid System Prototyping, RSP 2012

Y2 - 11 October 2012 through 12 October 2012

ER -

Ghidini Y, Webber T, Moreno E, Grando F, Fagundes R, Marcon C. Buffer depth and traffic influence on 3D NoCs performance. In Proceedings of the 2012 23rd IEEE International Symposium on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 2012. 2012. p. 9-15. 6380684. (Proceedings - IEEE International Symposium on Rapid System Prototyping, RSP). doi: 10.1109/RSP.2012.6380684

Buffer depth and traffic influence on 3D NoCs performance

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this