摘要:
A content addressable memory (CAM). A data portion of the CAM array includes word data storage. Each word line includes CAM cells (dynamic or static) in the data portion and a common word match line. An error correction (e.g., parity) portion of the CAM array contains error correction cells for each word line. Error correction cells at each word line are connected to an error correction match line. A match on an error correction match line enables precharging a corresponding data match line. Only data on word lines with a corresponding match on an error correction match line are included in a data compare. Precharge power is required only for a fraction (inversely exponentially proportional to the bit length of error correction employed) of the full array.
摘要:
A content addressable memory (“CAM”) system includes a plurality of segments arranged in an array, wherein each of the plurality of segments includes a plurality of CAM cells, each of the plurality of CAM cells includes a wordline, a matchline and a sinkline, the wordline being shared by all of the cells in the same row, the matchline and sinkline being shared by all of the cells in the same segment; and a corresponding method of searching within a CAM system includes providing an input word to the CAM system, comparing a portion of the input word in a segment of the CAM system, and propagating a mismatch to obviate the need for comparison in other segments of the CAM system.
摘要:
A hierarchical power supply noise monitoring device and system for very large scale integrated circuits. The noise-monitoring device is fabricated on-chip to measure the noise on the chip. The noise-monitoring system includes a plurality of on-chip noise-monitoring devices distributed strategically across the chip. A noise-analysis algorithm analyzes the noise characteristics from the noise data collected from the noise-monitoring devices, and a hierarchical noise-monitoring system maps the noise of each core to the system on chip.
摘要:
A three-dimensional (3D) chip stack structure and method of fabricating the structure thereof are provided. The 3D chip stack structure includes a plurality of vertically stacked chips which are interconnected and bonded together, wherein each of the vertically stacked chips include one or more IC device strata. The 3D chip stack structure further includes an air channel interconnect network embedded within the chip stack structure, and wherein the air channel interconnect network is formed in between at least two wafers bonded to each other of the vertically stacked wafers and in between at least two bonded wafers of the vertically stacked wafers at a bonding interface thereof. In addition, the 3D chip stack structure further includes one or more openings in a peripheral region of the chip stack structure that lead into and out of the air channel interconnect network, so that air can flow into and out of the air channel interconnect network through the one or more openings to remove heat from the chip stack structure.
摘要:
Improved transistor array device performance is obtained by use of bias voltage regulation which tracks with a fraction of a monitor transistor threshold voltage. The circuitry and methods are especially useful for improving the performance of transistor array devices such as DRAM and embedded DRAM. These benefits are obtained especially when at least two bias voltages normally supplied to the array are regulated by tracking with a fraction of an actual threshold voltage of at least one monitor transistor. Performance improvements include improved reliability, wider operational bias conditions, reduced power consumption and (in the case of memory cells) improved retention time.
摘要:
In accordance with the present invention, a method for producing at least two different chips with a controlled total chip thickness such that when these chips are placed into a corresponding pocket of a plurality of pockets located in a wafer chip carrier wherein each of the plurality of pockets have a total pocket depth (Tdp) at least substantially equal to one another, a substantially planarized top surface of said wafer chip carrier is achieved. The method comprises forming at least a first chip on a first dummy carrier and at least a second chip different from the first chip on a separate second dummy carrier using partial wafer bonding and partial wafer dicing. The method further includes using a chip thickness control mechanism in conjunction with said partial wafer bonding and partial wafer dicing in forming the at least a first chip and at least second chip different from the first chip, such that the at least first chip and the at least second different chip formed from each carrier each have a final total chip thickness (FTC) which is substantially equal to one another, and an FTC which is substantially equal to a total pocket depth (Tdp) of each of the uniform pockets of said wafer chip carrier, minus the final thickness of an attaching material (FTG) used within said each respective pocket.
摘要:
A method and system are disclosed herein for determining optimum power level settings for a transmitter and receiver pair of a communication system having a plurality of transmitter and receiver pairs, as determined with respect to bit error rate. In the method disclosed herein, the power levels of a transmitter and a receiver pair coupled to communicate over a duplex communication link are set to initial values. The bit error rate is then determined over the link. Then, the power level of the transmitter, the receiver, or both, is altered, incrementally, and the effect upon the bit error rate is determined. When an improvement appears in the bit error rate at an altered power level, the power level of the transmitter, the receiver or both, are set to the altered power level at which the improvement is found. The steps of incrementally altering power levels, determining the bit error rate, and establishing new power level settings when there is an improvement are repeated until power levels are determined at which the bit error rate is optimized.
摘要:
A data transmitter and transmitting method are provided in which an adaptive finite impulse response (FIR) driver has a plurality of taps to which coefficients having updateable values are applied. The FIR driver has a transfer function between an input stream of data bits and an output stream of data bits such that each data bit output from the FIR driver has an amplitude adjusted as a function of the values of a plurality of data bits of the input stream, and the values of the coefficients. The data transmitter includes a rewriteable non-volatile storage, operable to be rewritten with control information representing the values of the coefficients updated during operation of the FIR driver.
摘要:
Integrated circuits having combined memory and logic functions are provided. In one aspect, an integrated circuit is provided. The integrated circuit comprises: a substrate comprising a silicon layer over a BOX layer, wherein a select region of the silicon layer has a thickness of between about three nanometers and about 20 nanometers; at least one eDRAM cell comprising: at least one pass transistor having a pass transistor source region, a pass transistor drain region and a pass transistor channel region formed in the select region of the silicon layer; and a capacitor electrically connected to the pass transistor.
摘要:
A converter circuit and methods for operating the same. The converter circuit includes a supply voltage, a capacitor, an inductor, and four stacked switching elements. Each switching element is adjustable from a low resistance state to a high resistance state by a control signal. The inductor outputs current to a circuit load. The circuit may be operated in a first mode such that the output is adjustable between the supply voltage and half the supply voltage. Alternatively, in a second mode of operation, the output is adjustable from half the supply voltage to a ground voltage.