The very large scale integration (VLSI) community has utilized aggressive complementary metal-oxide semiconductor (CMOS) technology scaling to meet the ever-increasing performance requirements of computing systems. However, as we enter the nanoscale regime, the prevalent process variation effects degrade the CMOS device reliability. Hence, it is increasingly essential to explore emerging technologies which are compatible with the conventional CMOS process for designing highly-dense memory/logic circuits. Memristor technology is being explored as a potential candidate in designing non-volatile memory arrays and logic circuits with high density, low latency and small energy consumption. In this thesis, we present the detailed functionality of multi-bit 1-Transistor 1-memRistor (1T1R) cell-based memory arrays. We present the performance and energy models for an individual 1T1R memory cell and the memory array as a whole. We have considered TiO2- and H fOchi-based memristors, and for these technologies there is a sub-10% difference between energy and performance computed using our models and HSPICE simulations. Using a performance-driven design approach, the energy-optimized TiO2-based RRAM array consumes the least write energy (4.06 pJ/bit) and read energy (188 fJ/bit) when storing 3 bits/cell for 100 nsec write and 1 nsec read access times. Similarly, H fOchi-based RRAM array consumes the least write energy (365 fJ/bit) and read energy (173 fJ/bit) when storing 3 bits/cell for 1 nsec write and 200 nsec read access times.;On the logic side, we investigate the use of equalization techniques to improve the energy efficiency of digital sequential logic circuits in sub-threshold regime. We first propose the use of a variable threshold feedback equalizer circuit with combinational logic blocks to mitigate the timing errors in digital logic designed in sub-threshold regime. This mitigation of timing errors can be leveraged to reduce the dominant leakage energy by scaling supply voltage or decreasing the propagation delay. At the fixed supply voltage, we can decrease the propagation delay of the critical path in a combinational logic block using equalizer circuits and, correspondingly decrease the leakage energy consumption. For a 8-bit carry lookahead adder designed in UMC 130 nm process, the operating frequency can be increased by 22.87% (on average), while reducing the leakage energy by 22.6% (on average) in the sub-threshold regime. Overall, the feedback equalization technique provides up to 35.4% lower energy-delay product compared to the conventional non-equalized logic. We also propose a tunable adaptive feedback equalizer circuit that can be used with sequential digital logic to mitigate the process variation effects and reduce the dominant leakage energy component in sub-threshold digital logic circuits. For a 64-bit adder designed in 130 nm our proposed approach can reduce the normalized delay variation of the critical path delay from 16.1% to 11.4% while reducing the energy-delay product by 25.83% at minimum energy supply voltage. In addition, we present detailed energy-performance models of the adaptive feedback equalizer circuit. This work serves as a foundation for the design of robust, energy-efficient digital logic circuits in sub-threshold regime.
展开▼
