Abstract: Scalable Coherent Apparatus and Method have been disclosed. In one implementation a dual directory approach is used to implement scalable coherent accesses in a heterogeneous system.

Abstract: Scalable Coherent Apparatus and Method have been disclosed. In one implementation a dual directory approach is used to implement scalable coherent accesses in a heterogeneous system. A transaction identification mapping for coherent RapidIO memory transactions between a plurality of external hardware processing elements is used. Source transaction identification encoding is a combination of bits from two advanced extensible interface identifications. Target transaction identification is decoded into a combination of bits for two advanced extensible interface identifications.

Abstract: A Method and Apparatus for Flexible Coherent and Scale-out Computing Architecture have been disclosed. In one implementation a plurality of integrated photonics are used to effect communications.

Abstract: A method and apparatus provides a parameter estimation processor configured to estimate parameters used to compress data for transmission over a serial data link. The parameter estimation processor includes a processor. The processor includes user programmable inputs. The user programmable inputs set an input data packet length, a target compression ratio, and a resampling factor and allow filter parameters to be set. Input data information is received from an input data buffer of a data sample compressor. The processor performs a function that: (a) adjusting a target compression ratio by a first compression ratio to determine a remaining compression ratio when the resampling operation is enabled; (b) estimating a set of compression parameters that are used to achieve the remaining compression ratio, the set of compression parameters includes an attenuation value, filter order, a type of encoding; and (c) sends the set of compression parameters to the data sample compressor.

Abstract: A method and apparatus for compressing data in a communication system by receiving uncompressed packet at a compressor of the communication system, segmenting the packet into a plurality of packet segments, estimating packet segment compression parameters for each of the plurality of packet segments and compressing the packet segments utilizing one or more of the estimated packet segment compression parameters that are estimated and adjusted based upon the signal characteristics of the uncompressed packet and a desired compression ratio.

Abstract: The method and apparatus of the present invention provides for the compression and decompression of data bursts wherein the propagation of synchronization errors is limited to a desired number of signal samples and the start of a burst boundary is identified. In accordance with the present invention, a method and apparatus are provided for compressing data in a communication system by receiving data bursts comprising a plurality of uncompressed data packets at a compressor of the communication system, generating a start of burst parameter and a packet size parameter for each of the uncompressed data packets and compressing the data packets. At the decompressor, the compressed data packets are received and when a synchronization error occurs, the packet size parameter is used to limit the propagation of the error to a desired number of samples and to restore the data burst utilizing the start of burst parameter.

Abstract: Methods and Apparatus for IO, Processing and Memory Bandwidth Optimization for Analytics Systems has been disclosed. In one implementation a plurality of cards each having one or more GPU+CPU are interconnected via RapidIO.

Abstract: Methods and Apparatus for Efficient Network Analytics and Computing Card have been disclosed. In one implementation a plurality of cards each having one or more GPU+CPU are interconnected via RapidIO.

Abstract: A dynamic context resource module measures a compression performance level of a most recent compressed data packet of each of a plurality of compressed signal streams to generate a signal stream compression performance level for each signal stream. Dynamic compression performance indicators are calculated from the measured signal stream compression performance levels and are stored in a dynamic context resource table. A compression parameter estimation module reads the dynamic compression performance indicators and determines if each signal stream exhibits a desired performance level. If a signal stream does not exhibit the desired performance level, the compression parameters for the signal stream are adjusted. A compressed packet generator compresses a next data packet of the signal stream based upon the adjusted compression parameters for the signal stream or the unadjusted compression parameters for the signal stream.

Abstract: A method and apparatus provides a parameter estimation processor configured to estimate parameters used to compress data for transmission over a serial data link. The parameter estimation processor includes a processor. The processor includes user programmable inputs. The user programmable inputs set an input data packet length, a target compression ratio, and a resampling factor and allow filter parameters to be set. Input data information is received from an input data buffer of a data sample compressor. The processor performs a function that: (a) adjusting a target compression ratio by a first compression ratio to determine a remaining compression ratio when the resampling operation is enabled; (b) estimating a set of compression parameters that are used to achieve the remaining compression ratio, the set of compression parameters includes an attenuation value, filter order, a type of encoding; and (c) sends the set of compression parameters to the data sample compressor.

Abstract: The method and apparatus of the present invention provides for reduced power consumption and cost while supporting wide bandwidth signals from a large number of antennas, as is required by next generation systems. In accordance with the present invention, a method and apparatus are provided for processing data in a baseband unit in which compressed signals from remote radio units are sent directly to the switch instead of to a separate conversion device. Thereby, the local input and output bottleneck within processing devices is removed in the data path of the signals between the remote radio units and the baseband unit. This also reduces the number of ports in the switch and bandwidth requirement for the switch.

Abstract: A method and apparatus provides a parameter estimation processor configured to estimate parameters used to compress data for transmission over a serial data link. The parameter estimation processor includes a processor. The processor includes user programmable inputs. The user programmable inputs set an input data packet length, a target compression ratio, and a resampling factor and allow filter parameters to be set. Input data information is received from an input data buffer of a data sample compressor. The processor performs a function that: (a) adjusting a target compression ratio by a first compression ratio to determine a remaining compression ratio when the resampling operation is enabled; (b) estimating a set of compression parameters that are used to achieve the remaining compression ratio, the set of compression parameters includes an attenuation value, filter order, a type of encoding; and (c) sends the set of compression parameters to the data sample compressor.

Abstract: The method and apparatus of the present invention provides for reduced power consumption and cost while supporting wide bandwidth signals from a large number of antennas, as is required by next generation systems. In accordance with the present invention, a method and apparatus are provided for processing data in a radio unit of a communication system by receiving compressed data at one or more interfaces or internal resources of a distributed switch of a radio unit, operating the distributed switch to match the interface bandwidth to the resource bandwidth and distributing the compressed data received at the one or more interfaces or internal resources to the one or more internal resources or interfaces of the radio unit through the distributed switch.

Abstract: A method and apparatus that compresses signal data to generate compressed signal data having a low latency jitter while maintaining an acceptable compression ratio and reasonable degradation, as is required by next generation systems. A method and apparatus for compressing data in a communication system by receiving uncompressed data at a compressor of the communication system, analyzing the uncompressed data at the compressor to estimate at least one compression parameter to be applied during the compression of the uncompressed data, compressing the uncompressed data utilizing the at least one estimated compression parameter, monitoring the performance level of the compressed data packet and adjusting the estimated compression parameter used to compress the uncompressed data if the compressed data packet does not exhibit a desired performance level.

Abstract: A method and apparatus for a configurable packet routing, buffering and scheduling scheme to optimize throughput with deadlock prevention in SRIO-to-PCIe Bridges have been described. A single level enqueue method with dynamic buffering and dequeuing based on packet re-ordering is disclosed. Single level packet routing and scheduling to meet SRIO and PCIe rules to enqueue packets based on FType/TType is disclosed. Backpressure based on ingress watermarks for different packet types is disclosed. Use of a circular-reorder queue (CRQ) for both ingress and egress allows packet reordering and packet passing.

Abstract: A method and apparatus provides a parameter estimation processor configured to estimate parameters used to compress data for transmission over a serial data link. The parameter estimation processor includes a processor. The processor includes user programmable inputs. The user programmable inputs set an input data packet length, a target compression ratio, and a resampling factor and allow filter parameters to be set. Input data information is received from an input data buffer of a data sample compressor. The processor performs a function that: (a) adjusting a target compression ratio by a first compression ratio to determine a remaining compression ratio when the resampling operation is enabled; (b) estimating a set of compression parameters that are used to achieve the remaining compression ratio, the set of compression parameters includes an attenuation value, filter order, a type of encoding; and (c) sends the set of compression parameters to the data sample compressor.

Abstract: The method and apparatus of the present invention provides for the compression of signal data having a low latency jitter while maintaining a target compression ratio and reasonable degradation, as is required by next generation systems. In accordance with the present invention, a method and apparatus are provided for compressing data in a communication system by receiving uncompressed packet at a compressor of the communication system, segmenting the packet into a plurality of packet segments, calculating signal sample bit-removal information for each of the plurality of packet segments and compressing the packet segments utilizing the signal sample bit-removal information that is calculated based upon the signal characteristics of the uncompressed packet and a desired target compression ratio.

Abstract: A method and apparatus for a configurable packet routing, buffering and scheduling scheme to optimize throughput with deadlock prevention in SRIO-to-PCIe Bridges have been described. A single level enqueue method with dynamic buffering and dequeuing based on packet re-ordering is disclosed. Single level packet routing and scheduling to meet SRIO and PCIe rules to enqueue packets based on FType/TType is disclosed. Backpressure based on ingress watermarks for different packet types is disclosed. Use of a circular-reorder queue (CRQ) for both ingress and egress allows packet reordering and packet passing.

Abstract: A Method and Apparatus for Channel Equalization in High Speed S-RIO based Communication Systems have been disclosed. By adjusting equalizer coefficients based on 8B10B error counts and an error threshold, a receiver may be dynamically adjusted. By adjusting transmitter pre-emphasis based on 8B10B error counts and an error threshold, a transmitter may be dynamically adjusted. Both the transmitter and receiver may be adjusted dynamically based on 8B10B error counts and different error thresholds.

Abstract: A method and apparatus for a configurable packet routing, buffering and scheduling scheme to optimize throughput with deadlock prevention in SRIO-to-PCIe Bridges have been described. A single level enqueue method with dynamic buffering and dequeuing based on packet re-ordering is disclosed. Single level packet routing and scheduling to meet SRIO and PCIe rules to enqueue packets based on FType/TType is disclosed. Backpressure based on ingress watermarks for different packet types is disclosed. Use of a circular-reorder queue (CRQ) for both ingress and egress allows packet reordering and packet passing.