Adapative Error Protection for Energy Efficiency

With dramatic scaling in feature sizes, noise resilience is becomingone of the most important design parameters, similar to performanceand energy efficiency. Noise resilience is particularly problematicin long on-chip buses of complex single chip systems suchas on-chip multiprocessors. While one might opt to employ a verypowerful error protection scheme, this may not be very energy efficientas noise behavior typically varies over time. In this paper, wepropose an adaptive error protection scheme for energy efficiency,where the type of the coding scheme is modulated dynamically.The idea behind this strategy is to monitor the dynamic variationsin noise behavior and use the least powerful (and hence the mostenergy efficient) error protection scheme required to maintain theerror rates below a pre-set threshold. Our detailed experimental resultsobtained through simulation show that this adaptive strategyachieves the same level of error protection as the most powerfulstrategy experimented, without experiencing the latter's energy inefficiency.Based on our results, we recommend system designersto adopt adaptive protection schemes in environments where bothenergy and reliability are important.

[1]  John F. Wakerly,et al.  Digital design - principles and practices , 1990, Prentice Hall Series in computer engineering.

[2]  K. L. Shepard,et al.  Noise in deep submicron digital design , 1996, ICCAD 1996.

[3]  Anoop Gupta,et al.  The SPLASH-2 programs: characterization and methodological considerations , 1995, ISCA.

[4]  Naraig Manjikian.  Multiprocessor enhancements of the SimpleScalar tool set , 2001, CARN.

[5]  Luca Benini,et al.  Low power error resilient encoding for on-chip data buses , 2002, Proceedings 2002 Design, Automation and Test in Europe Conference and Exhibition.

[6]  Sujit Dey,et al.  Fault modeling and simulation for crosstalk in system-on-chip interconnects , 1999, 1999 IEEE/ACM International Conference on Computer-Aided Design. Digest of Technical Papers (Cat. No.99CH37051).

[7]  Naresh R. Shanbhag,et al.  Reliable low-power design in the presence of deep submicron noise (embedded tutorial session) , 2000, ISLPED '00.

[8]  Sung-Mo Kang,et al.  Noise-aware power optimization for on-chip interconnect , 2000, ISLPED'00: Proceedings of the 2000 International Symposium on Low Power Electronics and Design (Cat. No.00TH8514).

[9]  Naresh R. Shanbhag,et al.  Toward achieving energy efficiency in presence of deep submicron noise , 2000, IEEE Trans. Very Large Scale Integr. Syst..

[10]  Giovanni De Micheli,et al.  An adaptive low-power transmission scheme for on-chip networks , 2002, 15th International Symposium on System Synthesis, 2002..

[11]  William J. Dally,et al.  Digital systems engineering , 1998 .

[12]  Naresh R. Shanbhag,et al.  Soft digital signal processing , 2001, IEEE Trans. Very Large Scale Integr. Syst..