Abstract
Process scaling has resulted in an exponential increase of the number of transistors available to designers. Meanwhile, global interconnect has not scaled nearly as well, because global wires scale only in one dimension instead of two, resulting in fewer, high-resistance routing tracks. This paper evaluates the use of three-dimensional (3D) integration to reduce global interconnect by adding multiple layers of silicon with vertical connections between them using through-silicon vias (TSVs). Because global interconnect can be millimeters long, and silicon layers tend to be only tens of microns thick in 3D stacked processes, the power and performance gains by using vertical interconnect can be substantial. To address the thermal issues that arise with 3D integration, this paper also evaluates the use of near-threshold computing---operating the system at a supply voltage just above the threshold voltage of the transistors.
Specifically, we will discuss the design and test of Centip3De, a large-scale 3D-stacked near-threshold chip multiprocessor. Centip3De uses Tezzaron's 3D stacking technology in conjunction with Global Foundries' 130 nm process. The Centip3De design comprises 128 ARM Cortex-M3 cores and 256MB of integrated DRAM. Silicon measurements are presented for a 64-core version of the design.<!-- END_PAGE_1 -->
- ARM Cortex-A9. http://www.arm.com/products/processors/cortex-a/cortex-a9.php.Google Scholar
- ARM Cortex-M3. http://www.arm.com/products/CPUs/ARM_Cortex-M3.html.Google Scholar
- Dennard, R., Gaensslen, F., Rideout, V., Bassous, E., LeBlanc, A. Design of ion-implanted mosfet's with very small physical dimensions. IEEE J. Solid State Circ. 9, 5 (Oct. 1974), 256--268,.Google ScholarCross Ref
- Dreslinkski, R.G., Zhai, B., Mudge, T., Blaauw, D., Sylvester, D. An energy efficient parallel architecture using near threshold operation. In Proceedings of the Parallel Architecture and Compilation Techniques (2007). Google ScholarDigital Library
- Fick, D., Dreslinski, R., Giridhar, B., Kim, G., Seo, S., Fojtik, M., Satpathy, S., Lee, Y., Kim, D., Liu, N., et al. Centip3de: A 3930 dmips/w configurable near-threshold 3d stacked system with 64 arm cortex-m3 cores. In ISSCC 2012 (2012).Google Scholar
- Hanson, S., et al. Ultralow-voltage, minimum-energy CMOS. IBM J. Res. Dev. 50, 4/5 (Jul/Sep 2006), 469--490. Google ScholarDigital Library
- Kim, T.H., Persaud, R., Kim, C. Silicon odometer: an on-chip reliability monitor for measuring frequency degradation of digital circuits. IEEE J. Solid State Circ. 43, 4 (Apr. 2008), 874--880.Google ScholarCross Ref
- Knickerbocker, J.U., et al. Three-dimensional silicon integration. IBM J. Res. Dev. 52, 6 (Nov. 2008), 553--569. Google ScholarDigital Library
- Loh, G. 3d-stacked memory architectures for multi-core processors. In ACM SIGARCH Computer Architecture News (2008), volume 36. IEEE Computer Society, 453--464. Google ScholarDigital Library
- Qazi, M., Stawiasz, K., Chang, L., Chandrakasan, A. A 512 kb 8t sram macro operating down to 0.57 μv with an ac-coupled sense amplifier and embedded data-retention-voltage sensor in 45 nm soi cmos. IEEE J. Solid State Circ. 46, 1 (Jan. 2011), 85--96.Google ScholarCross Ref
- Rusu, S., et al. A 45 nm 8-Core Enterprise Xeon Processor. In Proceedings of ISSCC (Feb. 2009).Google Scholar
- Soeleman, H., Roy, K. Ultra-low power digital subthreshold logic circuits. In Proceedings of the 1999 International Symposium on Low Power Electronics and Design (1999), 94--96. Google ScholarDigital Library
- Swanson, R., Meindl, J. Ion-implanted complementary mos transistors in low-voltage circuits. IEEE J. Solid State Circ. 7, 2 (Apr. 1972), 146--153.Google ScholarCross Ref
- Tezzaron Semiconductor FaStack® Technology. http://tezzaron.com/technology/FaStack.htm.Google Scholar
- Tezzaron Semiconductor Octopus DRAM. http://www.tezzaron.com/memory/Octopus.html.Google Scholar
- Vittoz, E., Fellrath, J. Cmos analog integrated circuits based on weak inversion operations. IEEE J. Solid State Circ. 12, 3 (Jun. 1977), 224--231.Google ScholarCross Ref
- Wang, A., Chandrakasan, A. A 180 mv fft processor using subthreshold circuit techniques. In IEEE International Solid-State Circuits Conference 2004, ISSCC 2004. Digest of Technical Papers (15--19 Feb. 2004), volume 1, 292--529,.Google Scholar
- Zhai, B., Blaauw, D., Sylvester, D., Hanson, S. A sub-200 mv 6t sram in 0.13 μm cmos. In IEEE International Solid-State Circuits Conference, 2007. ISSCC 2007. Digest of Technical Papers (Feb. 2007), 332--606.Google Scholar
- Zhai, B., Nazhandali, L., Olson, J., Reeves, A., Minuth, M., Helfand, R., Pant, S., Blaauw, D., Austin, T. A 2.60pj/inst subthreshold sensor processor for optimal energy efficiency. In IEEE VLSI Technology and Circuits (2006).Google Scholar
Index Terms
- Centip3De: a many-core prototype exploring 3D integration and near-threshold computing
Recommendations
Centip3De: A 64-Core, 3D Stacked Near-Threshold System
Centip3De uses the synergy between 3D integration and near-threshold computing to create a reconfigurable system that provides both energy-efficient operation and techniques to address single-thread performance bottlenecks. The original Centip3De design ...
Analysis of power-performance for ultra-thin-body GeOI logic circuits
ISLPED '11: Proceedings of the 17th IEEE/ACM international symposium on Low-power electronics and designThis work analyzes the power-performance of the emerging Ultra-Thin-Body (UTB) GeOI devices for logic circuit applications. The impacts of temperature and Vdd scaling on the leakage/delay are studied. Compared with the subthreshold leakage dominated SOI ...
Impact of channel length, gate insulator thickness, gate insulator material, and temperature on the performance of nanoscale FETs
Aggressive technology scaling as per Moore's law has led to elevated power dissipation levels owing to an exponential increase in subthreshold leakage power. Short channel effects (SCEs) due to channel length reduction, gate insulator thickness change, ...
Comments