One embodiment provides for a general-purpose graphics processor comprising a multisample antialiasing compression module to perform planar multi-sample anti-aliasing, the multisample antialiasing compression module to analyze color data for a set of sample locations of a first pixel; determine a first plane to allocate for the first pixel, wherein the first plane is a lowest order plane to be allocated for the first pixel; and merge a plane allocation for the first pixel with a plane allocation for a second pixel in response to a determination that the first plane is the lowest order plane to be allocated for the second pixel.

A device that includes a master boss and a slave boss. The slave boss is configured to iteratively send execute and output commands to a first plurality of nodes implemented by a node engine identified in a first boss table in response to receiving an execute command from the master boss. The master boss is configured to iteratively send execute and output commands to the slave boss and a second plurality of nodes implemented by the node engine identified in a second boss table. Each node is configured to receive a first correlithm object, fetch a second correlithm object based on the first correlithm object in response to receiving an execute command, and output the second correlithm object in response to receiving an output command.

Data weighted averaging of a thermometric coded input signal is accomplished by controlling the operation of a crossbar switch matrix to generate a current cycle of a data weighted averaging output signal using a control signal generated in response to feedback of a previous cycle of the data weighted averaging output signal. The control signal specifies a bit location for a beginning logic transition of the data weighted averaging output signal in the current cycle based on detection of an ending logic transition of the data weighted averaging output signal in the previous cycle.

An electronic device has a transmit circuit and a processing circuit. The processing circuit outputs a first portion of compressive sensing (CS) samples corresponding to a signal segment to another electronic device via the transmit circuit, and selectively outputs a second portion of the CS samples corresponding to the signal segment to another electronic device via the transmit circuit according to a response of another electronic device. In this way, a balance between the compression ratio and the reconstruction quality/speed can be achieved. Moreover, the signal reconstruction performed at the processing circuit may employ a multi-resolution/multi-scale reconstruction scheme to achieve a balance between the dictionary size and the reconstruction quality/speed, and/or may employ a multi-stage reconstruction scheme to achieve a balance between the reconstruction algorithm control setting and the reconstruction quality/speed. In addition, dictionary weighting, online dictionary update, and/or point constraints may be used to improve the reconstruction quality.

In various embodiments, a system comprising a network interface, a processor, and a non-transient memory medium operatively coupled to the processor is disclosed. The memory medium is configured to store a plurality of instructions configured to program the processor to receive a digital bit stream, transform the digital bit stream to an encoded digital bit stream. The encoded digital bit stream comprises at least one of a gateway channel, a composite channel, or a data channel, and any combination thereof, and provides the encoded digital bit stream to the network interface for transmission. A non-transitory computer-readable memory medium and a computer-implemented method also are disclosed.

Providing efficient lossless compression for small data blocks in processor-based systems is provided. In one aspect, a method comprises receiving a plurality of input words. Each mask of a plurality of masks is applied to each unassigned input word to generate a corresponding plurality of patterns. For each mask, if a most frequently occurring pattern exists among the plurality of patterns, the most frequently occurring pattern and an uncompressed data portion of each unassigned input word are stored in association with a prefix associated with the mask. The prefix is also assigned to each unassigned input word corresponding to the most frequently occurring pattern. A compressed output block is generated, comprising prefixes assigned to the plurality of input words, the most frequently occurring patterns associated with the assigned prefixes, and uncompressed data portions corresponding to one or more input words of the plurality of input words.

Disclosed herein are techniques for improving compression ratio for dictionary-based data compression. A method includes receiving a data block to be compressed, selecting an initial compression dictionary from a plurality of initial compression dictionaries based on a characteristic of the data block, loading the initial compression dictionary into an adaptive compression dictionary in a buffer, and compressing the data block using the adaptive compression dictionary. The method also includes updating the adaptive compression dictionary based on data in the data block that has been compressed, while compressing the data block.

A hardware compression architecture including a shift register including: a plurality of sequentially coupled stages and a window stage coupled at an output end of the shift register, the shift register configured to receive an uncompressed data stream at an input end and output the uncompressed data from the window stage; a plurality of comparators each coupled to receive a data value held in a corresponding stage of the shift register and a data value held in the window stage, each of the comparators being configured to output a comparison result indicating whether the received stage value and the window stage data value match; logic, coupled to the comparators to receive the comparison results, to selectively compute one or more indexes based on the comparisons; and an encoder coupled to receive the one or more indexes and output, based on the one or more indexes, a position of a matching data value and a length of a matching sequence of data values.

A method performed in real-time includes receiving and storing time-based data over a specific time period and dividing the specific time period into a plurality of time windows. The method further includes determining that data associated with two or more proximate time windows are within a predetermined variance of one another and responsive to the determination: generating a mathematical function representative of the data associated with the two or more proximate time windows, deleting the data associated with the two or more proximate time windows, and generating a representation of the deleted data from the mathematical function. In certain embodiments, the data comprises empirical network telemetry data.

A data compression apparatus is described which has an encoder configured to receive an input data item and to compress the data item into an encoding comprising a plurality of numerical values. The numerical values are grouped at least according to whether they relate to content of the input data item or style of the input data item. The encoder has been trained using a plurality of groups of training data items grouped according to the content and where training data items within individual ones of the groups vary with respect to the style. The encoder has been trained using a training objective which takes into account the groups.