An Energy and Performance Exploration of Network-on-Chip Architectures

作者: A. Banerjee , P.T. Wolkotte , R.D. Mullins , S.W. Moore , G.J.M. Smit

DOI: 10.1109/TVLSI.2008.2011232

关键词:

摘要: In this paper, we explore the designs of a circuit-switched router, wormhole quality-of-service (QoS) supporting virtual channel router and speculative accurately evaluate energy-performance tradeoffs they offer. Power results from placed routed in 90-nm CMOS process show that all architectures dissipate significant idle state power. The additional energy required to route packet through is then shown be dominated by data path. This leads key result that, if trend continues, use more elaborate control can justified will not immediately limited budget. A performance analysis also shows dynamic resource allocation lowest network latencies, while static may used meet QoS goals. Combining power figures allows an energy-latency product calculated judge efficiency each networks. was have very similar providing better performance, its for general purpose designs. Finally, area metrics are presented allow comparison implementation costs.

参考文章(17)
Daniel Greenfield, Arnab Banerjee, Jeong-Gun Lee, Simon Moore, Implications of Rent's Rule for NoC Design and Its Fault-Tolerance networks on chips. pp. 283- 294 ,(2007) , 10.1109/NOCS.2007.26
Nikolay Krasimirov Kavaldjiev, A run-time reconfigurable network-on-chip for streaming DSP applications University of Twente. ,(2006)
Hangsheng Wang, L.-S. Peh, S. Malik, A Technology-Aware and Energy-Oriented Topology Exploration for On-Chip Networks design, automation, and test in europe. ,vol. 2, pp. 1238- 1243 ,(2005) , 10.1109/DATE.2005.40
P.T. Wolkotte, G.J.M. Smit, G.K. Rauwerda, L.T. Smit, An energy-efficient reconfigurable circuit-switched network-on-chip international parallel and distributed processing symposium. pp. 155- ,(2005) , 10.1109/IPDPS.2005.95
Jinwen Xi, Peixin Zhong, A Transaction-Level NoC Simulation Platform with Architecture-Level Dynamic and Leakage Energy Models Proceedings of the 16th ACM Great Lakes symposium on VLSI - GLSVLSI '06. pp. 341- 344 ,(2006) , 10.1145/1127908.1127986
Xuning Chen, Li-Shiuan Peh, Leakage power modeling and optimization in interconnection networks Proceedings of the 2003 international symposium on Low power electronics and design - ISLPED '03. pp. 90- 95 ,(2003) , 10.1145/871506.871531
Kangmin Lee, Se-Joong Lee, Hoi-Jun Yoo, Low-power network-on-chip for high-performance SoC design IEEE Transactions on Very Large Scale Integration Systems. ,vol. 14, pp. 148- 160 ,(2006) , 10.1109/TVLSI.2005.863753
Robert Mullins, Andrew West, Simon Moore, The design and implementation of a low-latency on-chip network Proceedings of the 2006 conference on Asia South Pacific design automation - ASP-DAC '06. pp. 164- 169 ,(2006) , 10.1145/1118299.1118348
N. Jouppi, The Future Evolution of High-Performance Microprocessors international symposium on microarchitecture. pp. 155- 155 ,(2005) , 10.1109/MICRO.2005.34
Karam S. Chatha, Praveen Vellanki, Nilanjan Banerjee, A power and performance model for network-on-chip architectures design, automation, and test in europe. ,vol. 2, pp. 21250- ,(2004) , 10.5555/968879.969210