Examples herein relate to decoding tokens using speculative decoding operations to decode tokens at an offset from a token decoded by a sequential decoding operation. At a checkpoint, a determination is made as to whether tokens to be decoded by the sequential and speculative decoding operations align. If there is alignment, the speculatively decoded tokens after a discard window are committed and made available for access. If there is not alignment, the speculatively decoded tokens are discarded. A miss in alignment and a fullness level of a buffer that stores speculatively decoded tokens are assessed to determine a next offset level for a start of speculative decoding. A size of a discard window can be set using a relationship based on the offset level to improve buffer utilization and to attempt to improve changes of alignments.

An input control unit collectively sends a plurality of pieces of variable length code data. A primary analysis processing unit processes a first piece of variable length code data, recognizes zero run information, group number information, and overhead bit information relating thereto, and outputs them to a frequency conversion unit. A continuous analysis processing unit processes one or more pieces of variable length code data subsequent to the first piece of variable length code data. In the case of variable length code data targeted for a predetermined process, the continuous analysis processing unit recognizes zero run information, group number information, and overhead bit information relating thereto, and output them to the frequency conversion unit. In the case of not the variable length code data targeted for the predetermined process, the continuous analysis processing unit discards this and subsequent pieces of data.

The application provides an operation method and device. Quantized data is looked up to realize an operation, which simplifies the structure and reduces the computation energy consumption of the data, meanwhile, a plurality of operations are realized.

A predictive model utilizes a set of coefficients for processing received input data. To reduce memory usage storing the coefficients, a compression circuit compresses the set of coefficients prior to storage by generating a cumulative count distribution of the coefficient values, and identifying a distribution function approximating the cumulative count distribution. Function parameters for the determined function are stored in a memory and used by a decompression circuit to apply the function the compressed coefficients to determine the decompressed component values. Storing the function parameters may consume less memory in comparison to storing a look-up table for decompression, and may reduce an amount of memory look-ups required during decompression.

Lossless content-aware compression and decompression techniques are provided for floating point data, such as seismic data. A minimum-length compression technique exploits an association between an exponent and a length of the significand, which corresponds to the position of the least significant bit of the significand. A reduced number of bits from the significand can then be stored. A prediction method is also optionally previously applied, so that residual values with shorter lengths are compressed instead of the original values. An alignment compression technique exploits repetition patterns in the floating point numbers when they are aligned to the same exponent. Floating point numbers are then split into integral and fractional parts. The fractional part is separately encoded using a dictionary-based compression method, while the integral part is compressed using a delta-encoding method. A prediction method is also optionally previously applied to the integral part, thereby increasing the compression ratio.

A graph compression system includes a memory unit to store graph data, and an electronic hardware controller in signal communication with the memory unit. The electronic hardware controller determines a distribution of a set of vertices in a graph, and encodes each vertex included in the set of vertices as a variable length integer (VLI) that includes a variable number of bytes. The variable number of bytes of each vertex is based on the determined distribution. Accordingly, the memory unit stores each vertex having been encoded according to the distribution of the set of vertices in the graph.

In one embodiment, an apparatus comprises a decompression engine to determine a plurality of tokens used to encode a block of data; populate a lookup table with at least two of the tokens in order of increasing token length; disable a first portion of the lookup table and enable a second portion of the lookup table based on a value of a payload of the block of data; and search for a match between a token and the payload in the second portion of the lookup table.

The application provides an operation method and device. Quantized data is looked up to realize an operation, which simplifies the structure and reduces the computation energy consumption of the data, meanwhile, a plurality of operations are realized.

Systems, apparatuses, and methods for implementing a parallel Huffman decoding scheme are disclosed herein. A system with a plurality of execution units receives a Huffman encoded bitstream. The system partitions the encoded bitstream into a plurality of chunks. Each execution unit is assigned to decode a separate chunk of the encoded bitstream as well as an extra portion of an adjacent chunk. With this approach, the decoding of the bitstream overlaps for a programmable amount of data at each chunk boundary since each execution unit, excluding the first execution unit decoding the first chunk of the bitstream, will likely decode a certain number of symbols incorrectly at the beginning of the chunk since the chunk boundaries will not be aligned with symbol boundaries. The system determines, from the decoded extra portion at each chunk boundary, where incorrectly decoded ends and where correctly decoded data begins for each decoded chunk.

Modifying a digital data stream that includes immediately consecutive code words of different length by segmenting, based on a certain block grid, the digital data stream. Each block of the block grid includes a fixed number of bits. It is determined whether all bits of the last block associated with the digital data stream are occupied by data of the digital data stream. If not all bits of the last block are occupied, the unoccupied bits of the last block are padded with bits of an end-of-record (EOR) indicator. If all bits of the last block are occupied, attaching an EOR indicator to the digital data stream is skipped.