Hung C. Ngo

A 32-bit PowerPC system-on-a-chip with support for dynamic voltage scaling and dynamic frequency scaling

A PowerPC system-on-a-chip processor which makes use of dynamic voltage scaling and on-the-fly frequency scaling to adapt to the dynamically changing performance demands and power consumption constraints of high-content, battery powered applications is described. The PowerPC core and caches achieve frequencies as high as 380 MHz at a supply of 1.8 V and active power consumption as low as 53 mW at a supply of 1.0 V. The system executes up to 500 MIPS and can achieve standby power as low as 54 /spl mu/W. Logic supply changes as fast as 10 mV//spl mu/s are supported. A low-voltage PLL supplied by an on-chip regulator, which isolates the clock generator from the variable logic supply, allows the SOC to operate continuously while the logic supply voltage is modified. Hardware accelerators for speech recognition, instruction-stream decompression and cryptography are included in the SOC. The SOC occupies 36 mm/sup 2/ in a 0.18 /spl mu/m, 1.8 V nominal supply, bulk CMOS process.

Design methodology for a 1.0 GHz microprocessor

This paper describes the design methodology used to build an experimental 1.0 GigaHertz PowerPC integer microprocessor at IBMs Austin Research Laboratory. The high frequency requirements dictated the chip composition to be almost entirely custom macros using dynamic circuit techniques. The methodology presented will cover design and verification tools as well as circuit constraints and microarchitecture philosophy. The microarchitecture, circuits and tools were defined by the high frequency requirements of the processor as well as the aggressive design schedule and size of the design team.

A 1 MB Cache Subsystem Prototype With 1.8 ns Embedded DRAMs in 45 nm SOI CMOS

We describe a single voltage supply, 1 MB cache subsystem prototype that integrates 2 GHz embedded DRAM (eDRAM) macros with on-chip word-line voltage supply generation , a 4 Kb one-time-programmable read-only memory (OTPROM) for redundancy and repair control, on-chip OTPROM programming voltage generation, clock generation and distribution, array built-in self-test circuitry (ABIST), user logic and pervasive logic. The eDRAM employs a programmable pipeline, achieving 1.8 ns latency, and features concurrent refresh capability.

Limited switch dynamic logic circuits for high-speed low-power circuit design

This paper describes a new circuit family--limited switch dynamic logic (LSDL). LSDL is a hybrid between a dynamic circuit and a static latch that combines the desirable properties of both circuit families. The paper also describes many enhancements and extensions to LSDL that increase its logical capability. Finally, it presents the results of two multiplier designs, one fabricated in 130- nm technology and one in 90-nm technology. The 130- and 90-nm designs respectively reach speeds up to 2.2 GHz and 8 GHz.

Controlled-load limited switch dynamic logic circuit

Limited switch dynamic logic (LSDL), a high performance logic circuit, has been modified by introducing a pseudo-nMOS style load. The resultant circuit consumes less power, primarily due to the reduction of capacitance on the clock network. The controlled-load LSDL is shown to be more robust to noise and power rail bounce. A 64-bit rotator circuit was used in the analysis. The effect of process variation on circuit performance is also evaluated.

Circuit Techniques Utilizing Independent Gate Control in Double-Gate Technologies

Independent gate control in double-gate (DG) devices enhances circuit performance and robustness while substantially reducing leakage and chip area. In this paper, we describe circuit techniques which take advantage of the independent biasing properties of symmetrical and asymmetrical DG devices in design. DG circuits at the 25-nm node are analyzed via mixed-mode numerical simulations using Taurus MEDICI. In dynamic circuits, we give examples of conditional keepers, charge sharing prevention scheme, and static keepers. A conditional keeper can dynamically achieve the optimal strength ratio between keeper and evaluation devices by utilizing the front- and backchannel currents in DG devices. A charge sharing mitigation scheme utilizing the back-gate of a logic transistor is then described. Static data retention scheme in dynamic circuits is proposed. A case study for analog applications using a voltage controlled oscillator (VCO) illustrates the specific advantages of DG devices.

Wide limited switch dynamic logic circuit implementations

Wide circuit implementation of limited switch dynamic logic (LSDL), a high performance logic circuit, with a modified pseudo-nMOS style load has been studied in this paper. A conventional two levels of 8-way mux implementation and a wide 64-way mux implementation of the 64-bit rotator circuit were used in the analysis. The resultant wide circuit implementation consumes less power, maintains robustness to noise and power rail bounce and reduces the pipeline depth to a single stage. The effect of process variation on circuit performance is evaluated. Based on the analysis, a new circuit style known as hybrid LSDL has been proposed.

Low-Power High-Performance Asymmetrical Double-Gate Circuits Using Back-Gate-Controlled Wide-Tunable-Range Diode Voltage

This paper presents a new power-reduction scheme using a back-gate-controlled asymmetrical double-gate device with robust data-retention capability for high-performance logic/SRAM power gating or variable/dynamic supply applications. The scheme reduces the transistor count, area, and capacitance in the header/footer device and provides a wide range of virtual ground (GND) or supply voltage. Physical analysis and numerical mix-mode device/circuit-simulation results confirm that the proposed scheme can be applied to low-power high-performance circuit applications in 65-nm technology node and beyond. Variable/dynamic supply or GND voltage using the proposed scheme improves read and write margins in scaled SRAM without degrading read and write performance.

A low-overhead virtual rail technique for SRAM leakage power reduction

We propose a virtual supply rail control technique that reduces SRAM leakage. This method encompasses a cell-based image, serial tiling, pitch matching, small drive device overhead, and controlled power-on currents while incurring small circuit overhead. A virtual rail cell contains both the sleep transistor fingers and input/output drive transistors. The usual overhead associated with the drive circuit that controls the sleep transistors is significantly reduced due to reduced wire load and improved drive efficiency. This technique provides gradual power-on characteristics and good signal slews while effectively mitigating leakage current, maintaining read/write speed and achieving power-on latency compatible with high-performance designs.

TCAD/Physics-Based Analysis of High-Density Dual-BOX FD/SOI SRAM Cell With Improved Stability

This paper presents a new SRAM cell using a global back-gate bias scheme in dual buried-oxide (BOX) FD/SOI CMOS technologies. The scheme uses a single global back-gate bias for all cells in the entire columns or subarray, thereby reducing the area penalty. The scheme improves 6T SRAM standby leakage, read stability, write ability, and read/write performance. The basic concept of the proposed scheme is discussed based on physical analysis/equation to facilitate device parameter optimization for SRAM cell design in back-gated FD/SOI technologies. Numerical 2-D mixed-mode device/circuit simulation results validate the merits and advantages of the proposed scheme.