Joint Hybrid Frequent Value Cache and Multi-Coding for Data Bus Energy Saving

In the deep submicron technology domain, the on-chip buses consume considerable amount of total energy of embedded multi-core chip. Lots of techniques have been produced to reduce the bus energy consumption. FVE (Frequent Value Encoding) and FV-MSB (Frequent Value-Most Significant Bit) which exploit abundant value locality on the data buses, are effective methods for reducing data bus energy consumption. In this paper, we propose a method that exploits more value locality that is overlooked by the FVE and FV-MSB. We found that a significant amount of non-frequent values and low-order bits of partial frequent values, not captured by the FVE and FV-MSB, produced large number of switching activity. Therefore, we produce an bus energy saving method based on frequent values and multi-coding which can be used to further reduce the on-chip data bus switching activity. The simulation results show that our method can reduce the ratio of switching activity by 18.7% on the data bus lines, and obtain the maximum ratio of energy saving by 17.76% and the average ratio about 16.91%, with 70nm technology when the coupling factor λ is 5. And the results also show that the method can still play a role when the technology size is further reduced in the future.

[1]  Anne Rogers,et al.  Supporting dynamic data structures on distributed-memory machines , 1995, TOPL.

[2]  Ahmad Khademzadeh,et al.  Data Encoding Techniques for Reducing Energy Consumption in Network-on-Chip , 2014, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[3]  Hui Zhang,et al.  Low-swing interconnect interface circuits , 1998, Proceedings. 1998 International Symposium on Low Power Electronics and Design (IEEE Cat. No.98TH8379).

[4]  Taewhan Kim,et al.  Bus-invert coding for low-power I/O - a decomposition approach , 2000, Proceedings of the 43rd IEEE Midwest Symposium on Circuits and Systems (Cat.No.CH37144).

[5]  Mahmut T. Kandemir,et al.  Optimizing bus energy consumption of on-chip multiprocessors using frequent values , 2004, 12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings..

[6]  A. Bahuman,et al.  A low-energy adaptive bus coding scheme , 2001, Proceedings IEEE Computer Society Workshop on VLSI 2001. Emerging Technologies for VLSI Systems.

[7]  Chih-Peng Fan,et al.  Novel low-power bus invert coding methods with crosstalk detector , 2011 .

[8]  Sung Kyu Lim,et al.  Design, packaging, and architectural policy co-optimization for DC power integrity in 3D DRAM , 2015, 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC).

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

[10]  Jun Yang,et al.  FV-MSB: A Scheme for Reducing Transition Activity on Data Buses , 2003, HiPC.

[11]  Min Chen,et al.  A novel pre-cache schema for high performance Android system , 2016, Future Gener. Comput. Syst..

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

[13]  John L. Henning.  SPEC CPU2006 benchmark descriptions , 2006, CARN.

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

[15]  Weisong Shi,et al.  Application configuration selection for energy-efficient execution on multicore systems , 2016, J. Parallel Distributed Comput..

[16]  Jun Yang,et al.  Frequent value compression in data caches , 2000, MICRO 33.

[17]  Chi-Ying Tsui,et al.  Dynamic reconfigurable bus encoding scheme for reducing the energy consumption of deep sub-micron instruction bus , 2004, 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512).

[18]  Spyros Tragoudas,et al.  On-Chip Codeword Generation to Cope With Crosstalk , 2014, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[19]  C CarlisleMartin,et al.  Supporting dynamic data structures on distributed-memory machines , 1995 .

[20]  Brajesh Kumar Kaushik,et al.  Bus encoder design for reduced crosstalk, power and area in coupled VLSI interconnects , 2013, Microelectron. J..