A High Throughput GALS Wrapper with Virtual Channels

Abstract:

A novel high throughput globally asynchronous locally synchronous wrapper is proposed to improve the network interface performance of network on chip. To prevent the occurrence of data sampling error, both the sender and receiver wrapper use the pausable clocking scheme to stop the clock when data come. The VC is based on Muller pipeline and chosen by the VC selector in the circuit. This wrapper can avoid the comparison between read pointer and write pointer in conventional design scheme and can increase the throughput. Simulations were based on SMIC 0.18um CMOS technology, and the sender wrapper and receiver wrapper have the throughput as high as 810Mflit/s and 820Mflit/s respectively.

• Periodical: Key Engineering Materials (Volumes 474-476)
• Pages: 2171-2176
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.474-476.2171
• Online since: April 2011
• Authors: Shi Tong Shang, Xu Guang Guan, Yin Tang Yang
• Keywords: Asynchronous Circuit, GALS, NoC, VC

References

[1] E. G. Friedman, Clock distribution networks in synchronous digital integrated circuits, Proc. IEEE, Vol. 89 (2001), p.665–692.

DOI: 10.1109/5.929649

[2] D. M. Chapiro, Globally Asynchronous Locally Synchronous Systems. PhD Dissertation, Stanford University (1984).

[3] Thonnart Yvain, Beigné Edith, Vivet Pascal. Design and Implementation of a GALS Adapter for ANoC based Architectures. In Pro. 15th International Symposium on Asynchronous Circuits and Systems, Chapel Hill, North Carolina, 2009, pp.13-22.

DOI: 10.1109/async.2009.13

[4] T. Chelcea, S. M. Nowick, Robust Interfaces for Mixed-Timing Systems, in IEEE Transactions on Very Large Scale Integration Systems, Vol. 12 (2004), pp.857-873.

DOI: 10.1109/tvlsi.2004.831476

[5] I. Miro-Panades, A. Greiner, Bi-Synchronous FIFO for Synchronous Circuit Communication Well Suited for Network-on-Chip in GALS Architectures", in Proceedings of the 1st International Symposium on Networks-on-Chip (NOCS, 07), pp.83-92.

DOI: 10.1109/nocs.2007.14

[6] R. Apperson, Z. Yu, M. J. Meeuwsen, T. Mohsenin, B. M. Baas, A Scalable Dual-Clock FIFO for Data Transfers Between Arbitrary and Haltable Clock Domains, in IEEE Transactions on VLSI Systems, Vol. 15 (2007), pp.1125-1134.

DOI: 10.1109/tvlsi.2007.903938

[7] E. Beigne, P. Vivet, Design of On-chip and Off-chip Interfaces for a GALS NoC Architecture, in Proceedings of the 12th International Symposium on Advanced Research in Asynchronous Circuits and Systems (ASYNC'06), pp.172-181.

DOI: 10.1109/async.2006.16

[8] E. Beigne, F. Clermidy, S. Miermont, P. Vivet, Dynamic Voltage and Frequency Scaling Architecture for Units Integration within a GALS NoC, in Proceedings of the 2nd IEEE International Symposium on Networks-on-Chip (NOCS'08), pp.129-138.

DOI: 10.1109/nocs.2008.4492732