113 results about "EDRAM" patented technology

A combined low-voltage, low-power band-gap reference and temperature sensor circuit is provided for providing a band-gap reference parameter and for sensing the temperature of a chip, such as an eDRAM memory unit or CPU chip, using the band-gap reference parameter. The combined sensor circuit is insensitive to supply voltage and a variation in the chip temperature. The power consumption of both circuits, i.e., the band-gap reference and the temperature sensor circuits, encompassing the combined sensor circuit is less than one μW. The combined sensor circuit can be used to monitor local or global chip temperature. The result can be used to (1) regulate DRAM array refresh cycle time, e.g., the higher the temperature, the shorter the refresh cycle time, (2) to activate an on-chip or off-chip cooling or heating device to regulate the chip temperature, (3) to adjust internally generated voltage level, and (4) to adjust the CPU (or microprocessor) clock rate, i.e., frequency, so that the chip will not overheat. The combined band-gap reference and temperature sensor circuit of the present invention can be implemented within battery-operated devices having at least one memory unit. The low-power circuits of the sensor circuit extend battery lifetime and data retention time of the cells of the at least one memory unit.

A method and embedded dynamic random access memory (EDRAM) circuit for implementing a physically unclonable function (PUF), and a design structure on which the subject circuit resides are provided. An embedded dynamic random access memory (EDRAM) circuit includes a first EDRAM memory cell including a memory cell true storage capacitor and a second EDRAM memory cell including a memory cell complement storage capacitor. The memory cell true storage capacitor and the memory cell complement storage capacitor include, for example, trench capacitors or metal insulator metal capacitors (MIM caps). A random variation of memory cell capacitance is used to implement the physically unclonable function. Each memory cell is connected to differential inputs to a sense amplifier. The first and second EDRAM memory cells are written to zero and then the first and second EDRAM memory cells are differentially sensed and the difference is amplified to consistently read the same random data.

A circuit and method for testing an eDRAM through a test controller with direct access (DA) mode logic is provided. The circuit and method of the present invention allows the testing of eDRAMs with a conventional memory tester. The present invention provides a semiconductor device including an embedded dynamic random access memory (eDRAM) for storing data, the eDram including a plurality of memory cells, and a test controller for testing the plurality of memory cells to determine if the cells are defective, the test controller including built-in self-test (BIST) logic circuitry for performing tests and for interfacing to a logic tester, and direct access mode logic circuitry for interfacing the eDRAM with an external memory tester. The test controller further comprises a multiplexer for multiplexing data, commands, and addresses from the BIST logic circuitry and the direct access mode logic circuitry to the eDRAM.

A memory refresh requestor, a memory request interpreter, a cache memory, and a cache controller on a single chip. The cache controller configured to receive a memory access request, the memory access request for a memory address range in the cache memory, detect that the cache memory located at the memory address range is available, and send the memory access request to the memory request interpreter when the memory address range is available. The memory request interpreter configured to receive the memory access request from the cache controller, determine if the memory access request is a request to refresh a contents of the memory address range, and refresh data in the memory address range when the memory access request is a request to refresh memory.

A method of fabricating an integrated circuit chip having a first region of a first layout rule and a second region of a second layout rule. The method includes using a first material to establish a first hard mask pattern in only the first region and using a second material to establish a second hard mask pattern on top of the first hard mask pattern. The second material is additionally used to establish a third hard mask pattern in the second region.

A method and apparatus for controlling marking store updates in a central electronic complex with a plurality of core processors and eDRAM cache and interconnect bus to a service processor for loading memory controller firmware to dual-channel DDR3 memory controllers with an internal marking store. Loaded firmware of the memory controllers is responsible for tracking of ECC errors using a ECC decoder control whereby said marking store is written by a slow ECC decoder, and read by a fast ECC decoder for every read operation of said memory controllers to provide a blocking mechanism for notifying marking store firmware when the marking store has been updated and which guarantees that marking store firmware cannot write to the marking store until the marking store firmware has seen updates without causing the marking store hardware to time out.

Apparatus and methods for signal margin centering for single-ended eDRAM sense amplifier. A plurality of DRAM cells is connected to an input side of a multiplexer by a first bitline. A single-ended sense amplifier is connected to an output side of the multiplexer by a second bitline. The single-ended sense amplifier has a switch voltage. The second bitline is precharged to a selected voltage level. The multiplexer passes a signal voltage from a selected one of the plurality of DRAM cells to the second bitline. The selected voltage level is selected such that reception of the signal voltage of a first type adjusts a voltage of the second bitline in a first direction and reception of the signal voltage of a second type adjusts the voltage of the second bitline in a second direction opposite from the first direction, centering the signal voltage around the switch voltage.

A memory refresh system and method. The inventive system includes a mechanism for selectively refreshing elements of a memory array in response to signals from a conventional memory management system. In the illustrative application, the memory is dynamic random access memory and the inventive system is adapted to provide for selective refresh of those DRAM memory elements to which data has been or will be stored. This allows for the use of advantageous DRAM memory elements while minimizing the power consumption thereof. Consequently, the utility of DRAM memory elements is extended to a variety of power sensitive applications including cellular telephony and mobile computing.

A method of forming a silicon-on-insulator (SOI) semiconductor structure in a substrate having a bulk semiconductor layer, a buried oxide (BOX) layer and an SOI layer. During the formation of a trench in the structure, the BOX layer is undercut. The method includes forming a dielectric material on the upper wall of the trench adjacent to the undercutting of the BOX layer and then etching the dielectric material to form a spacer. The spacer fixes the BOX layer undercut and protects it during subsequent steps of forming a bottle-shaped portion of the trench, forming a buried plate in the deep trench; and then forming a trench capacitor. There is also a semiconductor structure, preferably an SOI eDRAM structure, having a spacer which fixes the undercutting in the BOX layer.

The invention belongs to the technical field of dynamic random access memories and in particular relates to an EDRAM (Enhanced Dynamic Random Access Memory) unit of a gain unit, a memory and an operating method thereof. Based on a writing MOS (Metal Oxide Semiconductor) transistor, a reading MOS transistor, a writing word line, a writing bit line, a reading word line, a reading bit line and an equivalent parasitic capacitor, the EDRAM unit of the gain unit provided by the invention is obtained by increasing a coupling complementary MOS transistor and a shared bit line which is connected to a fixed voltage such that the EDRAM unit of the gain unit has the characteristics of long data retention time and low refreshing frequency; and the memory formed by the EDRAM unit of the gain unit has the characteristics of high reading speed and low power consumption.

A method of operating a memory includes performing a write operation and a read operation on a memory cell. The write operation includes starting a first global bit line (GBL) pre-charge on a GBL; and after the first GBL pre-charge is started, enabling a word line to write into the memory cell, wherein the steps of starting the first GBL pre-charge and enabling the word line have a first time interval. The read operation includes starting a second GBL pre-charge on the GBL; and after the second GBL pre-charge is started, enabling the word line to read from the memory cell, wherein the steps of starting the second GBL pre-charge and enabling the word line have a second time interval. The first time interval is greater than the second time interval.

A method for forming a memory device is provided by first forming at least one trench in a semiconductor substrate. Next, a lower electrode is formed in the at least one trench, and thereafter a conformal dielectric layer is formed on the lower electrode.

An upper electrode is then formed on the conformal dielectric layer. The forming of the upper electrode may include a conformal deposition of metal nitride layer, and a non-conformal deposition of an electrically conductive material atop the metal nitride layer, in which the electrically conductive material encloses the at least one trench.

Disclosed herein is a graphics-processing unit (GPU) having an internal memory for general-purpose use and applications thereof. Such a GPU includes a first internal memory, an execution unit coupled to the first internal memory, and an interface configured to couple the first internal memory to a second internal memory of an other processing unit. The first internal memory may comprise a stacked dynamic random access memory (DRAM) or an embedded DRAM. The interface may be further configured to couple the first internal memory to a display device. The GPU may also include another interface configured to couple the first internal memory to a central processing unit. In addition, the GPU may be embodied in software and/or included in a computing system.