Keith A. Bowman
ABSTRACT
A statistical performance simulator is developed to explore the impact of die-to-die (D2D) and within-die (WID) parameter variations on the distributions of maximum clock frequency (FMAX) and throughput for multi-core processors in a future 22nm technology.allThe simulator integrates a compact analytical throughput model, which captures the key dependencies of multi-core processors, into a statistical simulation framework that models the effects of D2D and WID parameter variations on critical path delays across a die. The salient contributions from this paper are: (1) Product-level variation analysis for multi-core processors must focus on throughput, rather than just FMAX, and (2) Multi-core processors are inherently more variation tolerant than single-core processors due to the larger impact of memory latency and bandwidth on overall throughput. To elucidate these two points, multi-core and single-core processors have a similar chip-level FMAX distribution (mean degradation of 9% and standard deviation of 5%) for multi-threaded applications. In contrast to single-core processors, memory latency and bandwidth constraints significantly limit the throughput dependency on FMAX in multi-core processors, thus reducing the throughput mean degradation and standard deviation by 50%. Since single-threaded applications running on a multi-core processor can execute on the fastest core, mean FMAX and throughput gains of 4% are achieved from the nominal design target.
- P. Gelsinger, "Microprocessors for the New Millennium: Challenges, Opportunities, and New Frontiers," in IEEE ISSCC, Feb. 2001, pp. 22--25.Google Scholar
- M. Horowitz and W. Dally, "How Scaling Will Change Processor Architecture," in IEEE ISSCC, Feb. 2004, pp. 132--133.Google Scholar
- M. Horowitz, et al., "Scaling, Power, and the Future of CMOS," in IEEE IEDM, Dec. 2005, pp. 11--17.Google Scholar
- K. Bowman, S. Duvall, and J. Meindl, "Impact of Die-to-Die and Within-Die Parameter Fluctuations on the Maximum Clock Frequency Distribution for Gigascale Integration," IEEE JSSC, pp. 183--190, Feb. 2002.Google Scholar
- International Technology Roadmap for Semiconductors (ITRS), Semiconductor Industry Association, www.itrs.net, 2006.Google Scholar
- S. Borkar, et al., "Parameter Variations and Impact on Circuits and Microarchitecture," in 40th DAC Proc., June 2003, pp. 338--342. Google ScholarDigital Library
- S. Duvall, "Statistical Circuit Modeling and Optimization," in 5th Intl. Workshop Statistical Metrology, June 2000, pp. 56--63.Google Scholar
- H. Masuda, et al., "Challenge: Variability Characterization and Modeling for 65- to 90-nm Processes," in IEEE CICC, Sept. 2005, pp. 593--600.Google Scholar
- S. Samaan, "The Impact of Device Parameter Variations on the Frequency and Performance of Microprocessor Circuits," in IEEE ISSCC Microprocessor Circuit Design Forum, Feb. 2004.Google Scholar
- J. Tschanz, et al., "Adaptive Body Bias for Reducing Impacts of Die-to-Die and Within-Die Parameter Variations on Microprocessor Frequency and Leakage," IEEE JSSC, pp. 1396--1402, Nov. 2002.Google Scholar
- Y. Abulafia and A. Kornfeld, "Estimation of FMAX and ISB in Microprocessors," IEEE Trans. VLSI Syst., pp. 1205--1209, Oct. 2005. Google ScholarDigital Library
- HSPICE User's Manual, Meta-Software Inc., Mar. 1995.Google Scholar
- P. Bai, et al., "A 65nm Logic Technology Featuring 35nm Gate Lengths, Enhanced Channel Strain, 8 Cu Interconnect Layers, Low-k ILD and 0.57 mm2 SRAM Cell," in IEEE IEDM, Dec. 2004, pp. 657--660.Google Scholar
- M. Eisele, et al., "The Impact of Intra-Die Device Parameter Variations on Path Delays and on the Design for Yield of Low Voltage Digital Circuits," IEEE Trans. VLSI Syst., pp. 360--368, Dec. 1997. Google ScholarDigital Library
- J. Schutz and C. Webb, "A Scalable X86 CPU Design for 90nm Process," in IEEE ISSCC, Feb. 2004, pp. 62--63.Google Scholar
- P. Dubey, "A Platform 2015 Model: Recognition, Mining and Synthesis Moves Computers to the Era of Tera," Intel Corp., Feb. 2005, download.intel.com/technology/computing/archinnov/platform2015/download/RMS.pdf.Google Scholar
- A. Alameldeen, Using Compression to Improve Chip Multiprocessor Performance, PhD Thesis, University of Wisconsin, 2006. Google ScholarDigital Library
- T. Karkhanis and J. Smith, "A First-Order Superscalar Processor Model," in Proc. of 31st Annual Intl. Symp. Comp. Arch., June 2004, pp. 338--349. Google ScholarDigital Library
- J. Emer, et al., "ASIM: A Performance Model Framework," IEEE Computer, pp. 68--76, Feb. 2002. Google ScholarDigital Library
- Microprocessor Quick Reference Guide, Intel Corp., www.intel.com/pressroom/kits/quickrefyr.htm. Google ScholarDigital Library
- SPEC Benchmarks, Systems Performance Evaluation Corp., www.spec.org.Google Scholar
- J. Doweck, "Inside the CORETM Microarchitecture," in HotChips, Aug. 2006.Google Scholar
- S. Rusu, et al., "A Dual-Core Multi-Threaded Xeon Processor with 16MB L3 Cache," in IEEE ISSCC, Feb. 2006, pp. 102--103.Google Scholar
Index Terms
- Impact of die-to-die and within-die parameter variations on the throughput distribution of multi-core processors
Recommendations
Impact of die-to-die and within-die parameter variations on the clock frequency and throughput of multi-core processors
A statistical performance simulator is developed to explore the impact of parameter variations on the maximum clock frequency (FMAX) and throughput distributions of multicore processors in a future 22 nm technology. The simulator captures the effects of ...
Time-borrowing multi-cycle on-chip interconnects for delay variation tolerance
ISLPED '06: Proceedings of the 2006 international symposium on Low power electronics and designInsertion of time-borrowing (TB) flip-flops in multi-cycle repeater-based on-chip interconnects enables significant improvements in mean performance and energy by averaging systematic and random within-die (WID) delay variations across multiple ...
Reducing random-dopant fluctuation impact using footer transistors in many-core systems
Process variation creates core-speed discrepancy among the core in a many-core platforms. Random variation is one of the important components that contributes into core-speed discrepancy. In this paper, we propose a novel technique that uses footer ...
Comments