Abstract
With constant scaling of process technologies, chip design is becoming increasingly difficult due to process variations. The FPGA community has only recently started focusing on the effects of variations. In this work we present a statistical analysis to compare the effects of variations on designs mapped to FPGAs and ASICs. We also present CAD and architecture techniques to mitigate the impact of variations. First we present a variation-aware router that optimizes statistical criticality. We then propose a modification to the clock network to deliver programmable skews to different flip-flops. Finally, we combine the two techniques and the result is a 9x reduction in yield loss that translates to a 12% improvement in timing yield. When the desired timing yield is set to 99%, our combined statistical routing and skew assignment technique results in a delay improvement of about 10% over a purely deterministic approach.
- Alpert, C., Hu, J., Sapatnekar, S., and Villarubia, P. 2001. A practical methodology for early buffer and wire resource allocation. In Proceedings of the Design Automation Conference. 189--194. Google Scholar
Digital Library
- Berkeley-Device-Group. Predictive technology model. http://www.eas.asu.edu/ptm.Google Scholar
- Betz, V., Rose, J., and Marquardt, A. 1999. Architecture and CAD for Deep-Submicron FPGAs. Kluwer Academic Publishers. Google Scholar
Digital Library
- Capo. A large scale fixed-die placer from UCLA. http://vlsicad.ucsd.edu/GSRC/bookshelf/Slots/ Placement.Google Scholar
- Chakraborty, A., Duraisami, K., Sathanur, A., Sithambaram, P., Benini, L., Macii, A., Macii, E., and Poncino, M. 2006. Dynamic thermal clock skew compensation using tunable delay buffers. In Proceedings of the International Symposium on Low Power Electronics and Design. Google Scholar
Digital Library
- Chang, H. and Sapatnekar, S. 2005. Statistical timing analysis under spatial correlations. IEEE Trans. Computer-Aid. Des. Integ. Circ. Syst. 24, 9, 1467--1482. Google Scholar
Digital Library
- Chopra, K., Shah, S., Srivastava, A., Blaauw, D., and Sylvester, D. 2005. Parametric yield maximization using gate sizing based on efficient statistical power and delay gradient computation. In Proceedings of the International Conference on Computer-Aided-Design. Google Scholar
Digital Library
- Clark, C. 1961. The greatest of a finite set of random variables. Operations Research 9, 85--91.Google Scholar
Digital Library
- Cline, B., Chopra, K., Blaauw, D., and Cao, Y. 2006. Analysis and modeling of CD variation for statistical timing. In Proceeding of the International Conference on Computer-Aided Design. Google Scholar
Digital Library
- Froedberg, P., Cao, Y., Cain, J., Wang, R., Rabaey, J., and Spanos, C. 2005. Modeling within-die spatial correlation effects for process-design co-optimization. In Proceedings of the International Symposium on Quality of Electronic Design. 516--521. Google Scholar
Digital Library
- Guthaus, M., Venkateswaran, N., Viswesariah, C., and Zolotov, V. 2005. Gate sizing using incremental parameterized statistical timing analysis. In Proceedings of the International Conference on Computer-Aided-Design. Google Scholar
Digital Library
- Krishnan, G. Low-cost easypath FPGAs offer promise to ASSP companies. www.xilinx.com/ publications/xcellonline/xcell_53/xc_easyassp53.htm.Google Scholar
- Lin, Y. and Hei, L. 2007. Stochastic physical synthesis for FPGAs with pre-routing interconnect uncertainty and process variation. In Proceedings of the International Symposium on Field Programmable Gate Arrays. Google Scholar
Digital Library
- Lin, Y., Hutton, M., and Hei, L. 2006. Placement and timing for FPGAs considering variations. In Proceedings of the International Conference on Field Programmable Logic and Applications.Google Scholar
- Mishchenko, A., Chatterjee, S., Brayton, R., and Ciesielski, M. 2005. An integrated technology mapping environment. In Proceedings of the International Workshop on Logic Synthesis.Google Scholar
- Moghal, H., Qian, H., Sapatnekar, S., and Bazargan, K. 2007. Clustering based pruning for statistical criticality computation under process variations. In Proceedings of the International Conference on Computer-Aided Design. Google Scholar
Digital Library
- OpenAccess. Library core of EDA algorithms and infrastructure for openaccess. http://www. openedatools.org/projects/oagear.Google Scholar
- Poon, K. K. W., Wilton, S. J. E., and Yan, A. 2005. A detailed power model for field-programmable gate arrays. ACM Trans. Des. Automat. Electron. Systs. 10, 2, 279--302. Google Scholar
Digital Library
- Raj, S., Vrudhala, S., and Wang, J. 2004. A methodology to improve timing yield in the presence of process variations. In Proceedings of the Design Automation Conference. Google Scholar
Digital Library
- Rebonato, R., and Jackel, P. 1999. The most genereral methodology to create a valid correlation matrix for risk management and pricing purposes. J. Risk 2, 2.Google Scholar
- Sapatnekar, S. 2004. Timing. Springer-Verlag, New York. Google Scholar
Digital Library
- Sedcole, P. and Cheung, P. Y. K. 2006. Within-die delay variability in 90nm FPGAs and beyond. In Proceedings of the IEEE International Conference on Field Programmable Technology.Google Scholar
- Singh, D. and Brown, S. 2002. Constrained clock shifting for field programmable gate arrays. In Proceedings of the International Symposium on Field Programmable Gate Arrays. Google Scholar
Digital Library
- Sivaswamy, S. and Bazargan, K. 2007a. Statistical generic and chip-specific skew assignment for improving timing yield of FPGAs. In Proceedings of the International Conference on Field Programmable Logic and Its Applications.Google Scholar
- Sivaswamy, S. and Bazargan, K. 2007b. Variation-aware routing for FPGAs. In Proceedings of the International Symposium on Field Programmable Gate Arrays. Google Scholar
Digital Library
- Tsai, J., Baik, D., Chen, C., and Saluja, K. 2004. An yield improvement methodology using pre- and post-silicon statistical clock scheduling. In Proceedings of the International Conference on Computer Aided Design. Google Scholar
Digital Library
- Visweswariah, C., Ravindran, K., Kalafala, K., Walker, S., and Narayan, S. 2004. First-order incremental block-based statistical timing analysis. In Proceedings of the Design Automation Conference. Google Scholar
Digital Library
- Wang, L. 2005. An energy-efficient skew compensation technique for high-speed skew sensitive signaling. In Proceedings of the International Symposium on Circuits and Systems.Google Scholar
- Wong, H. Y., Cheng, L., Lin, Y., and Hei, L. 2005. FPGA device and architecture evalutaion considering process variations. In Proceedings of the International Conference on Computer-Aided-Design. Google Scholar
Digital Library
- Yeh, C.-Y. and Marek-Sadowska, M. 2005. Skew programmable clock design for FPGA and skew-aware placement. In Proceedings of the International Symposium on Field Programmable Gate Arrays. Google Scholar
Digital Library
Index Terms
Statistical Analysis and Process Variation-Aware Routing and Skew Assignment for FPGAs
Recommendations
Variation-aware routing for FPGAs
FPGA '07: Proceedings of the 2007 ACM/SIGDA 15th international symposium on Field programmable gate arraysChip design in the nanometer regime is becoming increasingly difficult due to process variations. ASIC designers have adopted statistical optimization techniques to mitigate the effects of variations. The FPGA community on the other hand, has only ...
Statistical clock skew analysis considering intradie-process variations
With shrinking cycle times, clock skew has become an increasingly difficult and important problem for high performance designs. Traditionally, clock skew has been analyzed using case-files which cannot model intradie-process variations and hence result ...
Statistical clock tree routing for robustness to process variations
ISPD '06: Proceedings of the 2006 international symposium on Physical designAdvances in VLSI technology make clock skew more susceptible to process variations. Notwithstanding efficient zero skew routing algorithms, clock skew still limits post-manufacturing performance. Process-induced skew presents an ever-growing limitation ...






Comments