Abstract: Methodology enabling a reduction of edge and strap cell size, and the resulting device are disclosed. Embodiments include: providing first and second NW regions on a substrate; providing first and second RX regions on the first and second NW regions, respectively; providing a contact on the substrate connecting the first and second RX regions; and providing a dummy PC on the substrate connecting the first and second RX regions. Other embodiments include: determining an RX region of an IC design; determining a PPLUS mask region extending along a horizontal direction and being on an entire upper surface of the RX region; determining a NW region extending along a vertical direction and separated from the RX region; and comparing an area of an overlap of the NW region and PPLUS mask region to a threshold value.

Abstract: Methodology enabling a reduction of edge and strap cell size, and the resulting device are disclosed. Embodiments include: providing first and second NW regions on a substrate; providing first and second RX regions on the first and second NW regions, respectively; providing a contact on the substrate connecting the first and second RX regions; and providing a dummy PC on the substrate connecting the first and second RX regions. Other embodiments include: determining an RX region of an IC design; determining a PPLUS mask region extending along a horizontal direction and being on an entire upper surface of the RX region; determining a NW region extending along a vertical direction and separated from the RX region; and comparing an area of an overlap of the NW region and PPLUS mask region to a threshold value.

Abstract: Hierarchical layout versus schematic comparison with extraneous device elimination is provided. This includes obtaining a hierarchical layout netlist for a circuit design, the hierarchical layout netlist grouping arrayed devices of the circuit design into blocks repeated at a top level of a hierarchy of the hierarchical layout netlist. A modified hierarchical layout netlist defining active devices and connections thereof to top level pads of the circuit design is generated, in which extraneous devices are selectively removed from the obtained hierarchical layout netlist. The modified hierarchical layout netlist is verified against an input schematic netlist defining active devices of the circuit design and connections thereof to pads of the circuit design.

Abstract: A method includes designating a cell mismatch parameter of a memory cell including a plurality of transistors and an initial value of a transistor mismatch parameter for each of the plurality of transistors. A critical current sensitivity parameter is determined for each of the plurality of transistors based on the transistor mismatch parameters in a computing apparatus. The cell mismatch parameter is distributed across the plurality of transistors in the computing apparatus to update the individual transistor mismatch parameters for each of the plurality of transistors based on the critical current sensitivity parameters and the cell mismatch parameter. The memory cell is simulated based on the individual transistor mismatch parameters to generate a simulation result.

Abstract: A network-on-chip router which includes an input buffer, an input controller connected to said input buffer, an arbiter connected to said input controller, a crossbar connected to said arbiter and said input buffer, and an output buffer connected to said crossbar. The network-on-chip router minimizes propagation time of data through the router by ensuring that the propagation delay of data through an input buffer is less than the combined propagation delay of data through an input controller and arbiter.

Abstract: A high performance memory based computation system comprises an array of memory cells. Each memory cell stores a logic data corresponding to a chosen combination of inputs based on a specific logic function. For improved performance, the memory cell array can be divided into sub-blocks; and the sub-blocks can be serially disposed or juxtaposed. The performance of the memory based computation system can further be improved by removing the repeated memory cell rows, column, and/or sub-arrays.

Abstract: A logic device implementing configurations for ROM based logic uses arrays of memory cells to provide outputs based on inputs received at the logic device. The logic device stores values in the memory cells that are accessed when an input is received. The memory cells are transistors that provide values of ‘1’ or ‘0.’ Various configurations reduce the number of transistors while implementing the memory block by utilizing a single bitline or a dynamic precharge implementation.

Abstract: A multivalued memory device which includes a first multivalued memory transistor and a second multivalued memory transistor, wherein each transistor has a channel made from at least one carbon nanotube or nanowire, wherein data is stored by varying the number of carbon nanotubes or nanowires used in the channel, wherein the channel is the at least one carbon nanotube or nanowire which allows current to flow through it.

Abstract: A network-on-chip router which includes an input buffer, an input controller connected to said input buffer, an arbiter connected to said input controller, a crossbar connected to said arbiter and said input buffer, and an output buffer connected to said crossbar. The network-on-chip router minimizes propagation time of data through the router by ensuring that the propagation delay of data through an input buffer is less than the combined propagation delay of data through an input controller and arbiter.

Abstract: Embodiments of the present invention disclose a memory architecture for optimizing memory performance and size. Memory optimization is realized by configuring the memory to a particular logic state; that is, restricting memory data storage to either logic “0” or “1.” The opposite logic state, “1” or “0,” can be available through initialization and, therefore, may be presumed. Accordingly, the presumed, initialized logic state is available unless the configured logic state in memory changes the initialized data during memory access. Memory size reduction is realized by restricting physical memory to contain only cells that store data. Memory size can be further reduced by eliminating redundant data rows and columns. By reducing memory size, processing speed can be enhanced and power consumption reduced relative to conventional memory structures.

Abstract: A multivalued memory device which includes a first multivalued memory transistor and a second multivalued memory transistor, wherein each transistor has a channel made from at least one carbon nanotube or nanowire, wherein data is stored by varying the number of carbon nanotubes or nanowires used in the channel, wherein the channel is the at least one carbon nanotube or nanowire which allows current to flow through it.

Abstract: A logic device implementing configurations for ROM based logic uses arrays of memory cells to provide outputs based on inputs received at the logic device. The logic device stores values in the memory cells that are accessed when an input is received. The memory cells are transistors that provide values of ‘1’ or ‘0.’ Various configurations reduce the number of transistors while implementing the memory block by utilizing a single bitline or a dynamic precharge implementation.

Abstract: A high performance memory based computation system comprises an array of memory cells. Each memory cell stores a logic data corresponding to a chosen combination of inputs based on a specific logic function. For improved performance, the memory cell array can be divided into sub-blocks; and the sub-blocks can be serially disposed or juxtaposed. The performance of the memory based computation system can further be improved by removing the repeated memory cell rows, column, and/or sub-arrays.

Abstract: A high performance memory based computation system comprises an array of memory cells. Each memory cell stores a logic data corresponding to a chosen combination of inputs based on a specific logic function. For improved performance, the memory cell array can be divided into sub-blocks; and the sub-blocks can be serially disposed or juxtaposed. The performance of the memory based computation system can further be improved by removing the repeated memory cell rows, column, and/or sub-arrays.

Abstract: Embodiments of the present invention disclose a memory architecture for optimizing memory performance and size. Memory optimization is realized by configuring the memory to a particular logic state; that is, restricting memory data storage to either logic “0” or “1.” The opposite logic state, “1” or “0,” can be available through initialization and, therefore, may be presumed. Accordingly, the presumed, initialized logic state is available unless the configured logic state in memory changes the initialized data during memory access. Memory size reduction is realized by restricting physical memory to contain only cells that store data. Memory size can be further reduced by eliminating redundant data rows and columns. By reducing memory size, processing speed can be enhanced and power consumption reduced relative to conventional memory structures.

Abstract: A high performance logic circuit optimizes a digital logic function by dividing the function into smaller blocks. Thus, the logic circuit is divided into smaller blocks. The smaller blocks are implemented with read-only memory (ROM), in which outputs corresponding to input combination are pre-stored. Inputs to each of the smaller blocks are used as an address to access the ROM.

Abstract: A CNT transistor has source extension 36a and drain extension 36b that shunt electrical current and reduce the effective CNT resistance and allow significant reductions in fringe capacitances 28 30. The extensions 36a 36b are electrically conductive, and are electrically connected to the source electrode 22 and drain electrode 24. The extensions each span a portion of gaps 35a 35b. Consequently, the source and drain can be located relatively far from the gate electrode 26, thereby reducing the fringe capacitances 28 30. Nanotube 20 is a semiconducting single-walled carbon nanotube, and the extensions 36a 36b comprise metallic-conducting nanotubes surrounding and coaxial with the nanotube 20. The nanotube 20 and extensions 36a 36b are fabricated from a multiwalled nanotube by selectively removing outer nanotubes in a region near the gate electrode. Alternatively, the extensions 36a 36b can comprise metal deposited on peripheral portions of the semiconducting CNT 20.