110 and a second computing apparatus 120. The electronic device or system 100 may also include processor 10, memory 20, and communications interface 30. In an example, the first computing apparatus 110 and/or the second computing apparatus 120 may be respective separate (or remote) devices in the electronic system 100, or they may be internal electronic apparatuses of the electronic device 100, noting that all below respective references to hardware and operations of the electronic apparatus are also applicable to such electronic devices. In an example, a processor 10 of the electronic device 100 may be configured to perform any one or any combination of operations or methods described herein, and/or a processor of the first computing 110 and/or processor of the second computing apparatus 120 may performed the respective compression and/or decompression operations described herein, using any one or any combination of memories or buffers of the first computing apparatus 110, the second computing apparatus 120, and/or memory 20 of the electronic device 100. The processor 10 of the electronic device m100 may also be configured to control other operations or functionalities, in addition to compression and decompression operations. The terms “decoding”, “decompression,” and “decompressing” may be used interchangeably below.

The first computing apparatus 110 may compress data including symbols using a variable length coding (VLC) scheme. For example, the first computing apparatus 110 may compress the data using Huffman coding without losing information about the symbols. The Huffman coding may be a scheme of compressing the data about the symbols using a binary tree or a Huffman tree for the symbols configured based on frequency of appearance of the symbols. Metadata about the binary tree may be associated with the compressed data to decode the compressed data. The compressed data for each of the symbols may be referred to as a codeword(s).

The second computing apparatus 120 may decode the compressed data using the metadata. For example, the second computing apparatus 120 may decode the compressed data by reading the compressed data (or the codewords) on a bit-by-bit basis and sequentially determining symbols matching the read bits using the metadata.

As the lengths of codewords for each of the symbols are different in the VLC scheme, lengths of codewords for combinations of different symbols may be different even if a number of the symbols is the same. In some examples, length information for the lengths of the codewords for the symbols may not be included in the compressed data or the metadata. In an instance where there is no length information of the codewords for the symbols, bits of the compressed data may be processed only in series. When the length information of the codewords for the symbols does exist in association with the compressed data or the metadata, the codewords are divided into a plurality of groups and the divided groups may be processed in parallel.

A method of processing compressed data in parallel will be described in detail below with reference to FIGS. 2 to 13. A method of generating compressed data will be described with reference to FIGS. 2 to 5B, and a method of decoding compressed data will be described with reference to FIGS. 6 to 13.

FIG. 2 illustrates an example electronic apparatus for compressing data according to one or more embodiments.

An electronic apparatus 200 may include a communications interface 210, a processor 220, and a memory 230. For example, the electronic apparatus 200 may be the first computing apparatus 110 described above with reference to FIG. 1. The communication interface 30 of FIG. 1 may have same configurations of any or all of the communication systems described herein.

Referring to FIG. 2, in a non-limiting example, the communications interface 210 may be connected to the processor 220 and the memory 230 and transmit and receive data to and from the processor 220 and the memory 230. The communications interface 210 may be connected to another external apparatus and transmit and receive data to and from the external apparatus. Hereinafter, transmitting and receiving “A” may refer to transmitting and receiving “information or data of or indicating A”.

The processor 220 may be configured to execute computer-readable instructions, which when executed by the processor 220, to configure the processor 220 to perform one or more or all operations and/or methods involving the compressing of data or perform any one or any combination of the operations and/or methods described herein. The processor 220 may include any one or a combination of two or more of, for example, a central processing unit (CPU), a processor core, a multi-core processor, a multiprocessor, an application-specific integrated circuit (ASIC), and a field-programmable gate array (FPGA). The processor 220 may also execute other computer-readable instructions, programs, or applications to control other functionalities of the electronic device.

The memory 230 may be configured to store, or store, any of the computer-readable instructions described herein. The memory 230 may be a volatile or nonvolatile memory.

The memory 230 may include, for example, random-access memory (RAM), dynamic random-access memory (DRAM), static random-access memory (SRAM), or other types of non-volatile memory known in the art.

The communication interface 210 may include a network interface or other wired and/or hardware interfaces (e.g., I/O hardware and/or transceivers) for connecting to a network and/or provide a data transfer channel with a mobile storage medium (e.g., the memory 230 and/or memory 30 of FIG. 1). The communicator 210 may be implemented as circuitry in the electronic apparatus 200. For example, the communicator 210 may include an internal bus and an external bus. FIG. 3 illustrates an example method of generating compressed data according to one or more embodiments.

In an example, the operations 310 to 350 of FIG. 3 may be performed by an electronic apparatus, such as the electronic apparatus 200 described above with reference to FIG. 2, as a non-limiting example.

Referring to FIG. 3, in a non-limiting example, the electronic apparatus 200 may obtain data about a plurality of symbols before operation 310 is performed. For example, a symbol may be an alphanumeric character or a special character, but it is not limited thereto. For example, the data about the plurality of symbols may be bit data based on Unicode or an ASCII code used in the electronic apparatus.

In operation 310, the electronic apparatus may divide the plurality of symbols into a plurality of k blocks. For example, when a number of the plurality of symbols is 40 and k is 4, 40 symbols may be divided into four blocks such that a first block includes 10 symbols, a second block includes 10 symbols, a third block includes 10 symbols, and a fourth block includes 10 symbols. In the above-described example, each block includes the same number of symbols, but examples are not limited thereto, and a plurality of symbols may be divided such that the number of symbols included in each of the plurality of k blocks is different from each other depending on an implementation example.

In an example, the value k may correspond to a number of decoding engines for decoding compressed data, which will be described in greater detail below with reference to FIG. 6.

In operation 320, the electronic apparatus may generate first codewords by compressing symbols of the first block among the plurality of k blocks. For example, the electronic apparatus may generate the first codewords by processing the symbols of the first block using Huffman coding.

The electronic apparatus may generate codewords for each of the other blocks as well as the first block. Operation 320 may be repeatedly performed to generate the codewords for each of the plurality of k blocks.

In operation 330, the electronic apparatus may generate first partial metadata for the compression of the symbols of the first block. The first partial metadata may include binary tree information or Huffman tree information about the Huffman coding for the symbols of the first block.

The electronic apparatus may generate partial metadata for each of the other blocks as well as the first block. Operation 330 may be repeatedly performed to generate partial metadata for each of the plurality of k blocks.

In operation 340, the electronic apparatus may generate first count information indicating a length of the first codewords. The length of the first codewords may be a number of bits representing the first codewords.

The electronic apparatus may generate count information indicating a length of codewords for each of the other blocks as well as the first codewords of the first block. Operation 340 may be repeatedly performed to generate the count information for each of the plurality of k blocks.

In operation 350, the electronic apparatus may generate compressed data for the plurality of symbols based on the first codewords, the first partial metadata, and the first count information.

A method of generating compressed data for a plurality of symbols based on first codewords, first partial metadata, and first count information (e.g., operation 350) will be described in greater detail below with reference to FIG. 4.

FIG. 4 illustrates an example method of generating compressed data based on a compressed data stream, metadata, and count information according to one or more embodiments.

According to an example, operation 350 described above with reference to FIG. 3 may include operations 410 to 440. Operations 410 to 440 may be performed by an electronic apparatus, such as the electronic apparatus 200 described above with reference to FIG. 2, as a non-limiting example.

Referring to FIG. 4, in a non-limiting example, in operation 410, the electronic apparatus may generate a compressed data stream for a plurality of symbols based on first codewords. For example, when the plurality of symbols is divided into four blocks, the compressed data stream may include the first codewords for a first block, second codewords for a second block, third codewords for a third block, and fourth codewords for a fourth block.

In operation 420, the electronic apparatus may generate metadata about the compressed data stream based on respective partial metadata for each block. For example, when the plurality of symbols is divided into four blocks, the metadata may include the first partial metadata about the first block, second partial metadata about the second block, third partial metadata about the third block, and fourth partial metadata about the fourth block.

In operation 430, the electronic apparatus may generate count information about a length of codewords for each of the plurality of k blocks based on first count information. For example, when the plurality of symbols is divided into four blocks, the count information may include the first count information about a length of the first codewords of the first block, second count information about a length of the second codewords of the second block, third count information about a length of the third codewords of the third block, and fourth count information about a length of the fourth codewords of the fourth block.

In operation 440, the electronic apparatus may generate compressed data based on the compressed data stream, the metadata, and the count information. For example, the compressed data may include the compressed data stream, the metadata, and the count information.

FIG. 5A illustrates an example method of generating a compressed data stream, metadata, and count information according to one or more embodiments.

Referring to FIG. 5A, in a non-limiting example, the electronic apparatus 200 described above with reference to FIG. 2, as a non-limiting example, may include a frequency counter and sorter 510, a Huffman tree/table builder 520, an encoder 530, an input buffer 540, an output buffer 550, and a counter 560. The frequency counter and sorter 510, the Huffman tree/table builder 520, the encoder 530, the input buffer 540, the output buffer 550, and the counter 560 may each be implemented through hardware or through hardware and computer-readable instructions. As a non-limiting example, the frequency counter and sorter 510, the Huffman tree/table builder 520, the encoder 530, the input buffer 540, the output buffer 550, and the counter 560 may be implemented based on hardware resources of the communications interface 210, the processor 220, and the memory 230 of the electronic apparatus 200.

The frequency counter and sorter 510 may receive data about a plurality of symbols to be compressed. The frequency counter and sorter 510 may divide the plurality of symbols into a plurality of blocks. The frequency counter and sorter 510 may calculate frequency of appearance of the symbols included in each of the blocks.

The Huffman tree/table builder 520 may generate Huffman tree information about a corresponding block based on the calculated frequency of appearance of the symbols. The Huffman tree information for each of the blocks may be different from each other. The Huffman tree/table builder 520 may generate partial meta data about the blocks based on the Huffman tree information. The Huffman tree/table builder 520 may generate metadata 572 that includes the partial metadata.

The input buffer 540 may store data 501 about the plurality of symbols to be compressed. The data 501 stored in the input buffer 540 may be provided to the encoder 530.

The encoder 530 may compress the data 501 based on the Huffman tree information about the blocks generated by the Huffman tree/table builder 520. For example, the Huffman tree/table builder 520 may generate codewords by compressing the data 501. For example, the Huffman tree/table builder 520 may generate first codewords for a first block and second codewords for a second block. In an example, a processor (e.g., the processor 220 of FIG. 2) may be configured to perform any or all operations of compression or encoding by encoding described herein, or the encoder 530 may also be representative of a processor configured to perform any or all operations of compression or encoding described herein, or representative of the processor and a memory also represented by the encoder 530, where the execution of computer-readable instructions stored in the memory may configure the processor to perform such operations.

The counter 560 may generate count information 576 by counting a length of codewords generated by the encoder 530. For example, the count information 576 may include first count information obtained by counting a length of the first codewords and second count information obtained by counting a length of the second codewords.

The output buffer 550 may store the codewords generated by the encoder 530. The codewords stored in the output buffer 550 may be a compressed data stream 574.

FIG. 5B illustrates an example compressed data including a compressed data stream, metadata, and count information according to one or more embodiments.

Referring to FIG. 5B, in a non-limiting example, the compressed data 570 may include the metadata 572, the compressed data stream 574, and the count information 576. The count information 576 may include first count information 582, second count information 584, and k-th count information 586.

FIG. 6 illustrates an example electronic apparatus for decoding a compressed data stream according to one or more embodiments.

Referring to FIG. 6, in a non-limiting example, an electronic apparatus may include a communications interface 610, a processor 620, a memory 630, and display device. For example, the electronic apparatus 600 may be the second computing apparatus 120 described above with reference to FIG. 1, as a non-limiting example. In addition, the configurations and operation of the communications interface 210, processor 220, and memory 230 of the electronic apparatus 200 of FIG. 2 may be same and/or respectively applicable with respect to the communications interface 610, the processor 620, and the memory 630 of the electronic apparatus 600, so similar descriptions regarding the same are therefore omitted.

The processor 620 may be configured to execute computer-readable instructions that, when executed by the processor 620, configure the processor 620. For example, the memory 630 may be configured to store or stores the computer-readable instructions to perform one or more or all operations and/or methods involving the decoding of compressed data, and/or perform any one or any combinations of the operations and/or methods described herein.

FIG. 7 illustrates an example method of decoding a compressed data stream according to one or more embodiments.

Referring to FIG. 7, in a non-limiting example, operations 710 to 750 may be performed by an electronic apparatus, such as the electronic apparatus 600 described above with reference to FIG. 6, as a non-limiting example.

In operation 710, the electronic apparatus may obtain target data to be decompressed. For example, the electronic apparatus may obtain the target data by receiving compressed data (e.g., the compressed data 570 described above with reference to FIG. 5B) from another apparatus, such as the electronic apparatus 200 of FIG. 2 as a non-limiting example. In an example, the target data may include a compressed data stream (e.g., the compressed data stream 574 of FIG. 5B) for a plurality of symbols divided into a plurality of k blocks, metadata (e.g., the metadata 572 of FIG. 5B) about the compressed data stream, and count information (e.g., the count information 576 of FIG. 5B) about a length of codewords for each of the plurality of k blocks. In an example, the target data may be other compressed data that includes another compressed data stream for a plurality of symbols that were divided into a plurality of k blocks, other metadata about the another compressed data stream, and other count information about a length of the codewords for the other plurality of k blocks.

In operation 720, the electronic apparatus may store the compressed data stream of the target data in an input buffer of the electronic apparatus. For example, the input buffer may store codewords of the compressed data stream on a n-bit basis. N bits may be a size of an array of a buffer.

In operation 730, the electronic apparatus may generate k sub-compressed data streams based on the count information of the target data and the compressed data stream stored in the input buffer. For example, each of the k sub-compressed data streams may include codewords in multiples of n bits.

A method of generating k sub-compressed data streams will be described in greater detail below with reference to FIGS. 8 and 9.

In operation 740, the electronic apparatus may generate k sub-symbols by processing each of the k sub-compressed data streams using k decoding engines. For example, a first decoding engine may generate first sub-symbols based on a first sub-compressed data stream provided to a first decoding engine. For example, when a total number of the plurality of symbols is 40 and k is 4, a number of the first sub-symbols may be 10. In an example, the k sub-compressed data streams are decoded using the k decoding engines to arrive at a k number of sub-symbols.

A method of generating k sub-symbols will be described in greater detail below with reference to FIGS. 9 and 10.

In operation 750, the electronic apparatus may generate an output data stream corresponding to the plurality of symbols based on the k sub-symbols. Each of the plurality of symbols may be an alphanumeric character or a special character, but is not limited thereto. For example, an output data stream corresponding to the plurality of symbols may be bit data based on Unicode or an ASCII code.

FIG. 8 illustrates an example method of generating k sub-compressed data streams according to one or more embodiments.

Referring to FIG. 8, in a non-limiting example, an operation (e.g., operation 730 described above with reference to FIG. 7) may include operations 810 and 820 to be described hereinafter. Operations 810 and 820 may be performed by an electronic apparatus (e.g., electronic apparatus 600 of FIG. 6, as a non-limiting example).

In operation 810, the electronic apparatus may generate a first sub-compressed data stream based on codewords of first regions of an input buffer in which first codewords for a first block among a plurality of k blocks are stored. For example, when a size of an array of an input buffer is n bits and lengths of the first codewords is greater than or equal to n, the first regions may include a plurality of arrays in which the first codewords are stored. For example, when the size of the array of the input buffer is 16 bits, and the lengths of the first codewords is 34, the first regions may include a first array, a second array, and a third array which are consecutive arrays. In the above-described example, when the first array includes only one codeword, the first regions may further include a fourth array including one last codeword.

The electronic apparatus may generate an entire group of data represented by a plurality of arrays in which the first codewords are stored as the first sub-compressed data stream. The first sub-compressed data stream may include bits in multiples of n.

In operation 820, the electronic apparatus may generate a second sub-compressed data stream based on codewords of second regions of an input buffer in which second codewords for a second block among the plurality of k blocks are stored. For example, when a size of an array of an input buffer is n bits and lengths of the second codewords is greater than or equal to n, the second regions may include a plurality of arrays in which the second codewords are stored.

When the first codewords and the second codewords are consecutive, a last codeword of the first codewords and a starting codeword of the second codewords may be consecutive bits. When the last codeword of the first codewords is not included in an array (e.g., the third array) of the input buffer, the last codeword of the first codewords and the starting codeword of the second codewords may be included in a same array. In this example, the first regions of the input buffer and the second regions of the input buffer may include at least one same overlap region (e.g., the third array).

The electronic apparatus may generate an entire group of data represented by a plurality of arrays in which the second codewords are stored as the second sub-compressed data stream. The second sub-compressed data stream may include bits in multiples of n. The description of operation 820 may also apply to a description of a generation of each of a third sub-compressed data stream to a k-th sub-compressed data stream.

FIG. 9 illustrates an example of input buffer in which a compressed data stream is stored according to one or more embodiments.

Referring to FIG. 9, in a non-limiting example, an input buffer 900 may store a plurality of k codewords for a plurality of k blocks. For example, the input buffer 900 may have an address defined by a width X and a depth Y and may include storage spaces for storing n bits. A number of storage spaces may be X×Y. For example, the input buffer 900 may be static random access memory (SRAM). The input buffer may be a memory an electronic apparatus (e.g., the electronic apparatus 600 of FIG. 6).

The storage spaces of the input buffer 900 may be represented as arrays. For example, the input buffer 900 may include a first array 901 through a twentieth array 920. Each array may include n bits.

For example, first codewords may be stored in the first array 901 to a sixth array 906. When a starting codeword of the first codewords is stored in a start position of the first array 901, the first array 901 may include only bits of the first codewords. When the first codewords are not in multiples of n, the sixth array 906 may include at least a portion including a last codeword of the first codewords and at least a portion including a starting codeword of the second codewords. In an example, do may denote a number of bits of the first codewords included in the sixth array 906. The electronic apparatus may calculate do based on first count information (e.g., the first count information 582 of FIG. 5B) of count information (e.g., the count information 576 of FIG. 5B) and transmit the calculated do to a first decoding engine. In an example, do may be considered first aberration information.

For example, second codewords may be stored in the sixth array 906 to a twelfth array 912. When a starting codeword of the second codewords is stored in an intermediate position of the sixth array 906, the sixth array 906 may include bits of the first codewords and bits of the second codewords. In an example, d₁may denote a number of bits of second codewords included in the twelfth array 912. The electronic apparatus may calculate d₁based on second count information (e.g., the second count information 584 of FIG. 5B) of the count information (e.g., the count information 576 of FIG. 5B) and transmit the calculated d₁to a second decoding engine. In an example, d₁may be second aberration information.

For example, k-th codewords may be stored in a fifteenth array 915 to the twentieth array 920. When a starting codeword of the k-th codewords is stored in an intermediate position of the fifteenth array 915, the fifteenth array 915 may include bits of k−1-th codewords and bits of the k-th codewords. In an example, d_k-2may denote a number of the bits of the k−1-th codewords included in the fifteenth array 915. In an example, d_k-1may denote a number of the bits of the k-th codewords included in the twentieth array 920. The electronic apparatus may calculate d_k-1based on k-th count information (e.g., the k-th count information 586 of FIG. 5B) of the count information (e.g., the count information 576 of FIG. 5B) and transmit the calculated d_k-1to a k-th decoding engine. In an example, d_k-1may be k-th aberration information.

In an example, when an address of a predetermined array of the input buffer 900 is defined as {i, j} (i is an integer from 0 to X−1, and j is an integer from 0 to Y−1), A_e-1denoting an address of a starting array for e-th codewords (e is any one of 1 to k) may be calculated by Equation 1 below, and A_edenoting an address of a last array may be calculated by Equation 2 below.

$\begin{matrix} A_{e - 1} = {i_{e - 1}, j_{e - 1}} = {((C_{0} + \dots + C_{e - 2}) \mod n) rem X, ((C_{0} + \dots + C_{e - 2}) \mod n) \mod X} & Equation 1 \end{matrix}$

$\begin{matrix} A_{e} = {i_{e}, j_{e}} = {((C_{0} + \dots + C_{e - 1}) \mod n) rem X, ((C_{0} + \dots + C_{e - 1}) \mod n) \mod X}; & Equation 2 \end{matrix}$

When A₀is calculated by Equation 1 and Equation 2, (C₀+ . . . +C_e-2) may be regarded as 0. In an example, mod may denote a modulo operation, and rem may denote a remainder operation.

In an example, e-th aberration information may be calculated by Equation 3 below.

$\begin{matrix} d_{e - 1} = (C_{0} + \dots + C_{e - 1}) rem n & Equation 3 \end{matrix}$

According to an example, e-th count information about the e-th codewords may indicate a cumulative length of the first to e-th codewords. For example, the first count information may indicate a length of the first codewords, the second count information may indicate a sum of the length of the first codewords and a length of the second codewords, and the k-th count information may indicate a sum of the length of the first codewords to a length of the k-th codewords. In the above-described example, Equation 1, Equation 2, and Equation 3 may be changed to Equation 4, Equation 5, and Equation 6, respectively.

$\begin{matrix} A_{e - 1} = {i_{e - 1}, j_{e - 1}} = {((C_{e - 2}) \mod n) rem X, ((C_{e - 2}) \mod n) \mod X} & Equation 4 \end{matrix}$

$\begin{matrix} A_{e} = {i_{e}, j_{e}} = {((C_{e - 1}) \mod n) rem X, ((C_{e - 1}) \mod n) \mod X} & Equation 5 \end{matrix}$

$\begin{matrix} d_{e - 1} = (C_{e - 1}) rem n & Equation 6 \end{matrix}$

Each of k decoding engines may generate sub-symbols based on a sub-compressed data stream and aberration information. For example, the first decoding engine may generate first sub-symbols based on a first sub-compressed data stream and first aberration information.

A method of generating first sub-symbols by a first decoding engine based on a first sub-compressed data stream and first aberration information will be described in greater detail below with reference to FIG. 10.

FIG. 10 illustrates an example method of generating first sub-symbols according to one or more embodiments.

Referring to FIG. 10, in a non-limiting example, an operation (e.g., operation 740 described above with reference to FIG. 7) may include operations 1010 and 1020 to be described hereinafter. Operations 1010 and 1020 may be performed by an electronic apparatus (e.g., electronic apparatus 600 described above with reference to FIG. 6, as a non-limiting example). For example, operations 1010 and 1020 may be performed by a first decoding engine that may be implemented with hardware resources or hardware and computer-readable instruction resources of the electronic apparatus 600.

In operation 1010, the electronic apparatus may ascertain or find a first valid sub-compressed data stream in a first sub-compressed data stream. In the example described above with reference to FIG. 9, the electronic apparatus may ascertain the first valid sub-compressed data stream including first codewords based on first aberration information (e.g., do of FIG. 9) among bits (e.g., 6×n bits) included in the first array 901 to the sixth array 906 of the first regions of the input buffer 900.

Similar to operation 1010, the electronic apparatus may ascertain or find a second valid sub-compressed data stream in a second sub-compressed data stream. In the example described above with reference to FIG. 9, the electronic apparatus may find the second valid sub-compressed data stream including second codewords based on the first aberration information (e.g., d₀of FIG. 9) and second aberration information (e.g., d₁of FIG. 9) among bits (e.g., 6×n bits) included in the sixth array 906 to the twelfth array 912 of the second regions of the input buffer 900. Among the bits included in the sixth array 906, bits corresponding to the first aberration information may not be included in the second valid sub-compressed data stream. Among the bits included in the twelfth array 912, bits corresponding to the second aberration information may not be included in the second valid sub-compressed data stream.

In operation 1020, the electronic apparatus may generate first sub-symbols by processing the first valid sub-compressed data stream using a first decoding engine. For example, the first decoding engine may generate the first sub-symbols corresponding to the first codewords based on the first valid sub-compressed data stream and first partial metadata. Data about the generated first sub-symbols may be stored in an output buffer of the first decoding engine. For example, the data about the first sub-symbols may be bit data based on Unicode or an ASCII code used in the electronic apparatus.

Each of k decoding engines may generate sub-symbols corresponding to an individual decoding engine by performing operations 1010 and 1020. In an example, k sub-symbols generated by the k decoding engines may form a plurality of symbols. In an example, data about the k sub-symbols generated by the k decoding engines may be same as the data 501 about the plurality of symbols described above with reference to FIG. 5A.

FIG. 11 illustrates an example method of generating a data stream decoded based on a compressed data stream, metadata, and count information according to one or more embodiments.

Referring to FIG. 11, in a non-limiting example, an electronic apparatus (e.g., the electronic apparatus 600 described above with reference to FIG. 6, as a non-limiting example) may include an input data supplier 1110 and a group 1120 of decoding engines. For example, the input data supplier 1110 may include an input buffer 1111, search logic 1112, and a metadata buffer 1113. For example, the group 1120 of decoding engines may include k decoding engines including a first decoding engine 1130. The first decoding engine 1130 may include a parallel to serial converter 1131, a 1132, and an output buffer 1133. The input buffer 111, the search logic 1112, the metadata buffer 1113, the parallel to serial converter 1131, the decoder 1132, and the output buffer 1133 may be implemented through hardware or through hardware and computer-readable instructions stored on memory (e.g., processor 620 and memory 630) of the electronic apparatus 600. For example, the input buffer 1111, the search logic 1112, the metadata buffer 1113, the parallel to serial converter 1131, the decoder 1132, and the output buffer 1133 may be implemented based on hardware resources of the communications interface 610, the processor 620, and the memory 630 of the electronic apparatus 600.

The input buffer 1111 may store a compress data stream (e.g., the compressed data stream 574 described above with reference to FIG. 5B).

In an example, the search logic 1112 may generate a plurality of sub-compressed data streams based on the compressed data stream 574 stored in the input buffer 1111 using the count information 576 described above with reference to FIG. 5B. For example, the search logic 1112 may generate a first sub-compressed data stream based on codewords of first regions of the input buffer 1111, generate a second sub-compressed data stream based on codewords of second regions, and a k-th sub-compressed data stream based on codewords of k-th regions. The input data supplier 1110 may provide each of k sub-compressed data streams generated by the search logic 1112 to a corresponding decoding engine. For example, the first sub-compressed data stream may be provided to the first decoding engine 1130.

The metadata buffer 1113 may store metadata (e.g., the metadata 572 described above with reference to FIG. 5B). The input data supplier 1110 may provide each of a plurality of pieces of metadata stored in the metadata buffer 1113 to a corresponding decoding engine. For example, first partial metadata may be provided to the first decoding engine 1130.

The input data supplier 1110 may calculate aberration information for each of the plurality of sub-compressed data streams based on a size (e.g., n) of an array of the input buffer 1111 and provide the calculated aberration information to a decoding engine. For example, first aberration information 1114 may be provided to the first decoding engine 1130.

In an example, the first decoding engine 1130 may serialize the first sub-compressed data stream through the parallel to serial converter 1131. The first decoding engine 1130 may determine a first valid sub-compressed data stream in the first sub-compressed data stream based on the first aberration information. The first valid sub-compressed data stream may correspond to first codewords representing first sub-symbols.

In an example, the parallel to serial converter 1131 may be an n:1 converter. In another example, the parallel to serial converter 1131 may be an n:n converter.

The decoder 1132 may generate the first sub-symbols corresponding to the first codewords based on the first valid sub-compressed data stream and the first partial metadata. The decoder 1132 may store data about the first sub-symbols in the output buffer 1133. The data about the first sub-symbols may be bit data based on Unicode or an ASCII code used in the electronic apparatus 600.

According to an example, a storage capacity of the output buffer may be a size capable of, or sufficient to, storing data about the first sub-symbols. For example, when a length of Unicode for a symbol is 16 bits and a number of first sub-symbols is 10, the storage capacity of the output buffer 1133 may be 160 bits.

Each of the k decoding engines of the group 1120 of decoding engines may output data about sub-symbols provided to each of the decoding engines. The k decoding engines may operate independently or in parallel with each other.

The group 1120 of decoding engines may output decoded data 1140 based on the data about the plurality of sub-symbols output by the k decoding engines. The decoded data 1140 may be a data stream corresponding to a plurality of symbols.

FIG. 12 illustrates an example method of scheduling k decoding engines of an electronic apparatus according to one or more embodiments.

in an example, and although k decoding engines start decoding target data simultaneously, a compression rate of a sub-compressed data stream or a valid sub-compressed data stream provided to each of the k decoding engines may be different from each other, and a time point when each of the k decoding engines completes decoding may be different from each other.

In an example, although some of the decoding engines may not finish decoding the target data, when there is additional target data that needs, or is desired, to be decoded in the electronic apparatus 600, the electronic apparatus 600 may provide an additional sub-compressed data stream of the additional target data to a decoding engine in an idle state, thereby increasing operation rates of the k decoding engines.

Referring to FIG. 12, in a non-limiting example, operations 1210 to 1240 may be performed after operation 720 described above with reference to FIG. 7 is performed. Operations 1210 to 1240 may be performed an electronic apparatus (e.g., by the electronic apparatus 600 described above with reference to FIG. 6).

In operation 1210, the electronic apparatus may obtain state information from the k decoding engines. For example, the state information may indicate whether a corresponding decoding engine has completed its decoding. For example, the state information may indicate whether the corresponding decoding engine is in an idle state (e.g., a complete decoding).

In operation 1220, in an example, the electronic apparatus may obtain additional target data. The additional target data may include an additional compressed data stream for a plurality of additional symbols divided into a plurality of k blocks, additional metadata about the additional compressed data stream, and additional count information about a length of additional codewords for each of the plurality of k blocks. The descriptions of the compressed data stream, the metadata, and the count information of the target data described above with reference to FIG. 7 may also apply to descriptions of the additional compressed data stream, the additional metadata, and the additional count information of the additional target data.

In operation 1230, the electronic apparatus may store the additional compressed data stream of the additional target data in an additional input buffer. For example, the electronic apparatus may store the additional target data in the additional input buffer different from an input buffer in which the target data is stored (e.g., in the memory 630). An example of the additional input buffer will be described in greater detail below with reference to FIG. 13.