Signature Codes for Energy-Efficient Data Movement in On-chip Networks

On-chip networks provide a scalable infrastructure for moving data among cores in many-core systems. In future technologies, significant amounts of dynamic energy are dissipated for data movement on on-chip links. This paper proposes Sig-NoC, a predictable approach via signature encoding and transition signaling to reduce switching activity on links. Also, we show that link energy dominates in future technologies. Sig-NoC makes the switching activity proportional to the number of 1s per data in the source by using transition signaling. We estimate the energy of each packet at the source of routing in NoC. Therefore, we reduce the number of 1s for high energy packets through signature coding in the source. Sig-NoC mechanism encodes the packets once at the source and decodes them at the destination only; therefore, it has virtually no impact on performance. Simulation results on NAS and Phoenix benchmark suits on 4X4 NoC indicate that Sig-NoC achieves an average of 28% reduction in the overall NoC energy.

[1]  Christoforos E. Kozyrakis,et al.  Phoenix rebirth: Scalable MapReduce on a large-scale shared-memory system , 2009, 2009 IEEE International Symposium on Workload Characterization (IISWC).

[2]  Ali El-Moursy,et al.  Optimization for traffic-based virtual channel activation low-power NoC , 2015, 5th International Conference on Energy Aware Computing Systems & Applications.

[3]  L. Benini,et al.  Xpipes: a network-on-chip architecture for gigascale systems-on-chip , 2004, IEEE Circuits and Systems Magazine.

[4]  Bharathi Subramaniam,et al.  Crosstalk minimization in network on chip (NoC) links with dual binary weighted code CODEC , 2020 .

[5]  Kunle Olukotun,et al.  The Future of Microprocessors , 2005, ACM Queue.

[6]  Tobias Bjerregaard,et al.  A survey of research and practices of Network-on-chip , 2006, CSUR.

[7]  Marjan Asadinia,et al.  On-Fly-TOD: an efficient mechanism for crosstalk fault reduction in WNoC , 2020, The Journal of Supercomputing.

[8]  David H. Bailey,et al.  The NAS parallel benchmarks summary and preliminary results , 1991, Proceedings of the 1991 ACM/IEEE Conference on Supercomputing (Supercomputing '91).

[9]  Onur Mutlu,et al.  Comparative evaluation of FPGA and ASIC implementations of bufferless and buffered routing algorithms for on-chip networks , 2015, Sixteenth International Symposium on Quality Electronic Design.

[10]  Mircea R. Stan,et al.  Bus-invert coding for low-power I/O , 1995, IEEE Trans. Very Large Scale Integr. Syst..

[11]  William J. Dally,et al.  Principles and Practices of Interconnection Networks , 2004 .

[12]  Niraj K. Jha,et al.  Reducing Wire and Energy Overheads of the SMART NoC Using a Setup Request Network , 2016, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[13]  Mahdi Nazm Bojnordi,et al.  STFL: Energy-Efficient Data Movement with Slow Transition Fast Level Signaling , 2019, 2019 56th ACM/IEEE Design Automation Conference (DAC).

[14]  Antonia Zhai,et al.  Energy-Efficient Time-Division Multiplexed Hybrid-Switched NoC for Heterogeneous Multicore Systems , 2014, 2014 IEEE 28th International Parallel and Distributed Processing Symposium.

[15]  Yuan Xie,et al.  Hybrid Drowsy SRAM and STT-RAM Buffer Designs for Dark-Silicon-Aware NoC , 2016, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[16]  Jun Yang,et al.  Frequent value encoding for low power data buses , 2004, TODE.

[17]  Norman P. Jouppi,et al.  Optimizing NUCA Organizations and Wiring Alternatives for Large Caches with CACTI 6.0 , 2007, 40th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 2007).

[18]  Sharath Krishnamurthy,et al.  Data Encoding Techniques for Reducing Energy Consumption in Network-on-Chip , 2015 .

[19]  Jose Renau,et al.  ESESC: A fast multicore simulator using Time-Based Sampling , 2013, 2013 IEEE 19th International Symposium on High Performance Computer Architecture (HPCA).

[20]  Jun Yang,et al.  Tunable and Energy Efficient Bus Encoding Techniques , 2009, IEEE Transactions on Computers.

[21]  Vincenzo Catania,et al.  An encoding scheme to reduce power consumption in Networks-on-Chip , 2009, 2009 International Conference on Computer Engineering & Systems.

[22]  Sparsh Mittal,et al.  A survey of encoding techniques for reducing data-movement energy , 2019, J. Syst. Archit..

[23]  Onur Mutlu,et al.  Carpool: a bufferless on-chip network supporting adaptive multicast and hotspot alleviation , 2017, ICS.