2010
DOI: 10.1007/978-3-642-15672-4_37
|View full text |Cite
|
Sign up to set email alerts
|

A Methodology for Design of Unbuffered Router Microarchitecture for S-Mesh NoC

Abstract: International audienceCurrently, most of Network on-Chip (NoC) architectures have some limitation in routing decisions. And it makes router nodes overloaded, and sequentially forms deadlock, livelock and congestion. A simple unbuffered router microarchitecture for S-mesh NoC architecture is proposed in this paper. Unbuffered router transforms message without making routing decision. Simulation results showed that S-mesh could get optimal performance in message latency compared with 2D-mesh, Butterfly and Octag… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(2 citation statements)
references
References 20 publications
0
2
0
Order By: Relevance
“…Since DVFS with in-situ timing-error monitoring is a very effective method for designing a dependable low-power circuit, many implementations of this method have been reported. [9][10][11][12] The error tolerance of these implementations is mainly based on duplicated logic circuits and comparisons of results obtained by using them. An error is considered to have occurred when two results are different.…”
Section: In-situ Timing-error Monitoringmentioning
confidence: 99%
See 1 more Smart Citation
“…Since DVFS with in-situ timing-error monitoring is a very effective method for designing a dependable low-power circuit, many implementations of this method have been reported. [9][10][11][12] The error tolerance of these implementations is mainly based on duplicated logic circuits and comparisons of results obtained by using them. An error is considered to have occurred when two results are different.…”
Section: In-situ Timing-error Monitoringmentioning
confidence: 99%
“…Some implementations comprise duplicated computational logic; the hardware overheads of the duplicated logic circuits are considerable. Such methods can include the employment of approximation circuits, 9,13) noise-tolerant algorithms, 14) or the TEAtime method, 15) etc.…”
Section: In-situ Timing-error Monitoringmentioning
confidence: 99%