摘要:
A computer-implemented method of automated contact list determination can include detecting a collaborative event in real time and, responsive to detecting the collaborative event, identifying an owner of an electronic message and at least one contact specified by the electronic message, wherein the electronic message is associated with the collaborative event. The contact can be added to a collaborative contact list for the owner. The method can include determining a collaborative ranking for each contact in the collaborative contact list according to a collaborative history between the owner and that contact, selecting a plurality of contacts from the collaborative contact list according to collaborative ranking, and including each of the plurality of contacts within a dynamic address book of the owner.
摘要:
A transition encoded dynamic bus includes an encoder circuit at the input to the bus and a decoder circuit at the output to the bus. The encoder circuit generates a signal indicative of a transition at the input to the bus rather than the actual value at the input. The decoder circuit decodes the transition encoded information to track the appropriate value to be output from the bus.
摘要:
An interconnect architecture is provided to reduce power consumption. A first driver may drive signals on a first interconnect and a second driver may drive signals on a second interconnect. The first driver may be powered by a first voltage and the second driver may be powered by a second voltage different than the first voltage.
摘要:
A system is provided that includes a clocking circuit to provide two repeater clock signals and a flop repeater circuit to receive the two repeater clock signals and an input data signal. The flop repeater circuit to provide an output data signal based on the two repeater clock signals. The flop repeater circuit including a plurality of transistors and inverters coupled together to function as a flip-flop circuit that passes data without any full transmission gates.
摘要:
A register file contains a local bit trace and a driving signal trace as well as a plurality of data cells coupled to the local bit trace. A device is coupled to the driving signal trace and the local bit trace to intelligently charge and float the local bit trace. The intelligent charging and floating is facilitated by determination of a selection of one of the data cells.
摘要:
A current mirror multi-channel leakage monitor circuit and method measures die leakage and generates digital keeper control bits to control a process compensated dynamic circuit. The leakage monitor enables high resolution on-chip leakage measurements in multiple locations on a die, thereby saving test time and enabling both die to die and within die process compensation.
摘要:
A noise canceling circuit is provided in a dynamic circuit that includes a high fan-in domino gate. The noise canceling circuit decouples noise from neighboring wires in the dynamic circuit that is injected into a wire that controls the domino gate.
摘要:
A method and apparatus to improve register file performance. In various embodiments, a shadow bitline runs parallel to a local bitline in a register file, and the shadow bitline is coupled to a subset of the data cells to which the local bitline is coupled. In operation, a static keeper holds the local bitline in a condition complementary to the condition of the shadow bitline, when appropriate.
摘要:
A double data rate dynamic logic gate in which an evaluation phase is performed for each phase of a clock signal. In one embodiment, an nMOSFET pull-down logic unit is clocked by two nMOSFETs switched in complementary fashion, and dynamic latches provide the output signals. In another embodiment, two nMOSFET pull-down logic units are employed, each clocked by an nMOSFET in complementary fashion, and a static logic unit provides the output signals.
摘要:
The present disclosure is directed to systems and methods of implementing a neural network using in-memory, bit-serial, mathematical operations performed by a pipelined SRAM architecture (bit-serial PISA) circuitry disposed in on-chip processor memory circuitry. The on-chip processor memory circuitry may include processor last level cache (LLC) circuitry. The bit-serial PISA circuitry is coupled to PISA memory circuitry via a relatively high-bandwidth connection to beneficially facilitate the storage and retrieval of layer weights by the bit-serial PISA circuitry during execution. Direct memory access (DMA) circuitry transfers the neural network model and input data from system memory to the bit-serial PISA memory and also transfers output data from the PISA memory circuitry to system memory circuitry. Thus, the systems and methods described herein beneficially leverage the on-chip processor memory circuitry to perform a relatively large number of vector/tensor calculations without burdening the processor circuitry.