Data processing method and system and chip

By dividing the logical memory block into multiple physical memory blocks and performing data interleaving processing, the problems of high trace density and low data processing efficiency in on-chip storage systems are solved, and a storage system design with high bandwidth and low complexity is realized.

CN119091937APending Publication Date: 2024-12-06TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202310665871.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-06
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Under the demand for ultra-large read and write bandwidth, the high trace density and low data processing efficiency caused by the splicing of multiple memory banks in on-chip storage systems.

Method used

By dividing logical memory blocks into multiple physical memory blocks and adopting a hierarchical aggregation strategy, data interleaving is performed to reduce wiring complexity and improve data processing efficiency.

Benefits of technology

The physical cabling friendly high-bandwidth on-chip storage system design is realized, reducing the wiring complexity and improving data processing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091937A_ABST
    Figure CN119091937A_ABST
Patent Text Reader

Abstract

The invention discloses a data processing method and system and a chip. The method comprises the steps that a first data processing request sent by first main equipment is received, the first data processing request comprises first address information used for specifying at least one logic storage block in the data processing system, and the data processing system comprises a plurality of storage block groups which are set to be routed and converged by a plurality of logic storage blocks in advance, and detecting that the first data processing request is valid in response to the first storage block group, respectively acquiring first data and second data from the first physical storage block and the second physical storage block, and performing interleaving processing on the first data and the second data to generate first target data. The data processing method and device can be applied to the field of data storage including but not limited to the field of data storage based on artificial intelligence, and solve the technical problems of low data processing efficiency and high wiring complexity of an on-chip storage system in related technologies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular, to a data processing method, system and chip. Background Art

[0002] At present, under the demand for ultra-large read and write bandwidth, multiple storage banks need to be spliced ​​in the on-chip storage system. The read data may come from any storage bank in the system, so the aggregation logic of all storage banks to the read data end is formed.

[0003] In scenarios where the read and write data width is large, for example, a storage body needs to provide a 512-bit read and write data interface, which will cause the wiring density of the wiring space above the storage body and the channel to be very large. If there are multiple master device interfaces in the system that need to access the storage system in parallel, the wiring pressure will be greatly increased, resulting in low data processing efficiency and high complexity of on-chip storage system wiring.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a data processing method, system and chip to at least solve the technical problems of low data processing efficiency and high wiring complexity of on-chip storage system in related technologies.

[0006] According to one aspect of an embodiment of the present application, a data processing method is provided, comprising: receiving a first data processing request sent by a first master device, wherein the first data processing request comprises first address information, the first address information is used to specify at least one logical storage block in a data processing system, the data processing system comprises a plurality of storage block groups, each of the storage block groups is configured to be pre-routed by a plurality of logical storage blocks; in response to the first storage block group detecting that the first data processing request is valid, obtaining first data and second data from a first physical storage block and a second physical storage block respectively, wherein the first storage block group comprises a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth for parallel data processing, the plurality of physical storage blocks comprising the first physical storage block and the second physical storage block; interleaving the first data and the second data to generate first target data, wherein the first target data represents the data that the first data processing request expects to process.

[0007] According to another aspect of an embodiment of the present application, a data processing system is also provided, including: a first master device, connected to a data storage device, and used to send a first data processing request, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in the data processing system, the data storage device includes multiple storage block groups, each of the storage block groups is set to be pre-routed by multiple logical storage blocks, and the multiple storage block groups include a first storage block group; the first storage block group is used to respond to the first storage block group to detect that the first data processing request is valid, and obtain first data and second data from a first physical storage block and a second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, the multiple physical storage blocks include the first physical storage block and the second physical storage block, the first data and the second data are used for interleaving processing to generate first target data, and the first target data represents the data that the first data processing request expects to process.

[0008] Optionally, the system also includes: a reading unit connected to the data storage device, and used to determine a first access address and a first data length according to a first data processing request sent by a first master device after receiving the first data processing request, wherein the first access address is used to indicate a starting access address of the first target data, and the first data length represents a data length of the first target data; obtaining a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of the multiple storage block groups, the number of storage blocks in each of the storage block groups, and the data bandwidth of each storage block; a computing unit connected to the reading unit, and used to determine the first address information according to the first access address and the first data length, and determine that the first data processing request is valid when the first address information falls within the first storage address range.

[0009] Optionally, the system also includes: a starting storage block grouping determination unit, connected to the reading unit, and used to determine the number of storage block groups that need to be accessed by the first data processing request according to the first data length, wherein the number of storage block groups is determined by the ratio of the first data length to a data bandwidth pre-set for the storage block; determining a starting storage block group according to the first access address, wherein the first storage block group includes the starting storage block group; a data processing unit, connected to the starting storage block grouping determination unit, and used to obtain the first data from the first physical storage block according to the starting storage block group and the number of storage block groups.

[0010] Optionally, the data processing unit includes: a termination storage block group calculation subunit, used to determine the termination storage block group based on the starting storage block group and the number of storage block groups; a group sequence number comparison subunit, used to compare the starting group sequence number of the starting storage block group with the termination group sequence number of the termination storage block group; a circular shift subunit, used to determine that the multiple storage block groups are wrapped when the termination group sequence number is less than the starting group sequence number, and perform a circular shift operation on the multiple storage block groups to move the data corresponding to the first storage block group to the low bit of the corresponding register, wherein the starting storage block group is arranged after the termination storage block group; a data acquisition subunit, used to obtain the first data from the first physical storage block corresponding to the first storage block group when the data corresponding to the first storage block group is moved to the low bit of the corresponding register.

[0011] Optionally, the system also includes: a second master device, connected to the data storage device, and used to send a second data processing request, wherein the second data processing request includes second address information, the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device; a second storage block group, used to respond to the second storage block group detecting that the second data processing request is valid, and obtain third data and fourth data from a third physical storage block and a fourth physical storage block respectively, wherein the multiple storage block groups include a second storage block group, the second storage block group includes a second logical storage block, the second logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, the multiple physical storage blocks include the third physical storage block and the fourth physical storage block, the third data and the fourth data are used for interleaving processing to generate second target data, and the second target data represents the data that the second data processing request expects to process.

[0012] Optionally, the system also includes: a priority information acquisition unit, used to acquire first priority information and second priority information when the first data processing request and the second data processing request are received, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; a detection unit, used to detect whether the first data processing request is valid when the first priority information is higher than the second priority information; detect whether the second data processing request is valid when the first priority information is lower than the second priority information; and randomly detect whether the first data processing request or the second data processing request is valid when the first priority information is the same as the second priority information.

[0013] Optionally, the system also includes: a first multiplexer, used to obtain the first data and the second data from the first logical storage block, and obtain the fifth data and the sixth data from the third logical storage block in response to the first storage block group detecting that the first data processing request is valid, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request; a second multiplexer, used to obtain the third data and the fourth data from the second logical storage block, and obtain the seventh data and the eighth data from the fourth logical storage block in response to the second storage block group detecting that the second data processing request is valid, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

[0014] Optionally, the first logical storage block includes: N physical storage blocks, wherein the first logical storage block includes the N physical storage blocks, and when the preset data width of the first logical storage block is M, the data width of each of the N physical storage blocks is M / N, and N and M are both positive integers.

[0015] According to another aspect of the embodiment of the present application, a data processing chip is provided, characterized in that the data processing chip includes the above-mentioned data processing system.

[0016] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned data processing method when running.

[0017] According to another aspect of the embodiments of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above data processing method.

[0018] According to another aspect of the embodiments of the present application, there is further provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the data processing method through the computer program.

[0019] In an embodiment of the present application, a first data processing request sent by a first master device is received, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, and the data processing system includes multiple storage block groups, and each of the storage block groups is configured to be pre-routed by multiple logical storage blocks; in response to the first storage block group detecting that the first data processing request is valid, first data and second data are obtained from a first physical storage block and a second physical storage block respectively, wherein the first storage block group includes a first logical storage block, and the first logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, and the multiple physical storage blocks include the first physical storage block and the first physical storage block. Two physical storage blocks; interleaving the first data and the second data to generate first target data, wherein the first target data represents the data that the first data processing request expects to process, by pre-routing and converging multiple logical storage blocks to form the above-mentioned multiple storage block groups, and then the multiple storage block groups constitute a data processing system to achieve a physical wiring-friendly high-bandwidth on-chip storage system design, by distributing ultra-large bit-width logical storage blocks to multiple physical storage blocks, and adopting a hierarchical convergence strategy, the local routing complexity of the wiring is greatly reduced, and it is beneficial to the convergence of the timing, thereby achieving the technical effect of improving data processing efficiency and reducing wiring complexity, and then solving the technical problems of low data processing efficiency and high wiring complexity of the on-chip storage system in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0021] Figure 1 is a schematic diagram of an application environment of an optional data processing method according to an embodiment of the present application;

[0022] Figure 2 is a flowchart of an optional data processing method according to an embodiment of the present application;

[0023] Figure 3 is a schematic diagram of an optional data processing method according to an embodiment of the present application;

[0024] Figure 4 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0025] Figure 5 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0026] Figure 6 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0027] Figure 7 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0028] Figure 8 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0029] Fig. 9 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0030] Fig.10 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0031] Fig.11 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0032] Fig.12 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0033] Fig.13 is a schematic diagram of the structure of an optional data processing system according to an embodiment of the present application;

[0034] Fig.14 is a schematic structural diagram of an optional data processing device according to an embodiment of the present application;

[0035] Fig.15 is a schematic diagram of the structure of an optional data processing product according to an embodiment of the present application;

[0036] Fig.16 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretation:

[0040] Bank: A block of memory in physical implementation.

[0041] BGU: bank group unit, consisting of multiple banks.

[0042] interleave: interleaving, usually refers to the storage of different domain segments of multiple groups of data mixed together.

[0043] mux: multiplexer, multiplexer.

[0044] The present application is described below in conjunction with embodiments:

[0045] According to one aspect of an embodiment of the present application, a data processing method is provided. Optionally, in this embodiment, the data processing method can be applied to Figure 1 In the hardware environment composed of the server 101 and the terminal device 103 shown in FIG. Figure 1As shown, the server 101 is connected to the terminal 103 via a network, and can be used to provide services for the terminal device or the application installed on the terminal device. The application can be a video application, an instant messaging application, a browser application, an educational application, a game application, etc. A database 105 may be set up on the server or independently of the server to provide data storage services for the server 101, for example, a game data storage server. The above-mentioned network may include, but is not limited to, a wired network and a wireless network, wherein the wired network includes, for example, a local area network, a metropolitan area network, and a wide area network; the wireless network includes, for example, Bluetooth, WIFI, and other networks that implement wireless communication; the terminal device 103 may be a terminal configured with an application program, and may include, but is not limited to, at least one of the following: a mobile phone (such as an Android phone, an iOS phone, etc.), a laptop computer, a tablet computer, a PDA, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a mixed reality (MR) terminal, and other computer devices; the above-mentioned server may be a single server, a server cluster consisting of multiple servers, or a cloud server.

[0046] Combination Figure 1 As shown, the above data processing method can be implemented in the terminal device 103 through the following steps:

[0047] S1, receiving a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, the data processing system includes a plurality of storage block groups, and each storage block group is configured to be pre-routing converged by a plurality of logical storage blocks;

[0048] S2, in response to the first storage block group detecting that the first data processing request is valid, obtaining the first data and the second data from the first physical storage block and the second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a first physical storage block and a second physical storage block;

[0049] S3, interleaving the first data and the second data to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

[0050] Optionally, in this embodiment, the above data processing method can also be implemented by a server, for example, Figure 1 It is implemented in the server 101 shown; or it is implemented by the terminal device and the server together.

[0051] The above is only an example and is not specifically limited in this embodiment.

[0052] Optionally, as an optional implementation, as Figure 2 As shown, the above data processing method includes:

[0053] S202, receiving a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, wherein the data processing system includes a plurality of storage block groups, and each storage block group is configured to be pre-routing converged by a plurality of logical storage blocks;

[0054] Optionally, in this embodiment, the above-mentioned first main device may include but is not limited to devices that allow sending data processing requests, for example, mobile phones (such as Android phones, iOS phones, etc.), laptops, tablets, PDAs, MIDs (Mobile Internet Devices), PADs, desktop computers, smart TVs, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, virtual reality (VR) terminals, augmented reality (AR) terminals, mixed reality (MR) terminals and other computer devices.

[0055] Specifically, the above-mentioned first main device may include but is not limited to an AI (Artificial Intelligence) processor device.

[0056] It should be noted that artificial intelligence is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0057] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0058] Optionally, in this embodiment, the above-mentioned first data processing request may include but is not limited to sending a read data request, a write data request, etc., and the above-mentioned first address information may include but is not limited to the storage address of the data that needs to be processed by the above-mentioned first data processing request, and the first address information may include but is not limited to being composed of a starting address and a data length.

[0059] Optionally, in this embodiment, the data processing system may include but is not limited to a data processing system used by a data processing chip, a data storage device, etc. The data processing system implements data aggregation by a pre-set gather logic.

[0060] Figure 3 is a schematic diagram of an optional data processing method according to an embodiment of the present application, such as Figure 3 As shown, under the demand for ultra-large read and write bandwidth, multiple storage bodies need to be spliced ​​in the on-chip storage system. The data processed by the data processing request may come from any storage body in the system, so a gathering logic (gather logic) from all storage bodies to the read data end is formed. For scenarios with small read and write data bit widths, such as CPU, too many storage bodies will not cause too much problem. Generally, the area occupied by the storage body is relatively large, and the channel above or beside the storage body has enough wiring space to distribute the gathering lines. However, if the storage body read and write data width is too large, for example, a storage body needs to provide a 512-bit read and write data interface (that is, 512 pins need to be set, one pin corresponds to one bit), then the routing density of the routing space above the storage body and the channel will be very large. If there are multiple master (master device) interfaces in the system that need to access the storage system in parallel, the routing pressure will be greatly increased. Therefore, by dividing the storage body into multiple storage block groups, and then setting each storage block group to be pre-routing converged by multiple logic storage blocks, the storage blocks are grouped by adopting a hierarchical aggregation method. First, the data of each storage block in the storage block group is aggregated, and then the data between the storage block groups is aggregated, and the long lines are cut into short lines to reduce the difficulty of timing convergence.

[0061] S204, in response to the first storage block group detecting that the first data processing request is valid, obtaining first data and second data from the first physical storage block and the second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a first physical storage block and a second physical storage block;

[0062] Optionally, in this embodiment, the above-mentioned first storage block grouping can be understood as a storage block grouping composed of multiple logical storage blocks divided by the data processing system. For example, the data processing system includes 64 logical storage blocks, and the 64 logical storage blocks can be divided into 16 storage block groups according to preset rules. Each storage block group includes 4 logical storage blocks, and each logical storage block can be evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel. For example, the data width of a logical storage block is pre-defined as 512 bits. When it needs to be divided into 2 physical storage blocks, the data bandwidth of the first physical storage block is 256 bits, and the data bandwidth of the second physical storage block is 256 bits. The first physical storage block and the second physical storage block are interleaved in units of 256 bits.

[0063] Figure 4 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Figure 4 As shown, by pre-routing and converging multiple logical storage blocks (storage block 1, storage block 2, storage block 3, storage block 4) to form the above-mentioned multiple storage block groups (storage block group 1, storage block group 2, storage block group 3, storage block group 4), and then the data processing system is formed by multiple storage block groups to realize the design of physical wiring friendly high-bandwidth on-chip storage system, by distributing the ultra-large bit width logical storage blocks to multiple physical storage blocks, and adopting the strategy of hierarchical convergence, the local routing complexity of the wiring is greatly reduced, and it is conducive to the convergence of the timing, wherein, taking the 512KByte on-chip cache data processing system as an example, it is logically divided into 64 storage blocks, the size of each storage block is set to 512KB / 64=8KB, the width of each storage block is defined as 512bit, and a storage block group is equivalent to the above-mentioned one storage body, then the depth of each storage block is 128 (also called bit width). Here, every 4 storage blocks are divided into a group, and the specific number of storage blocks and the storage block grouping method can be adapted according to different scenarios and can be arbitrarily expanded.

[0064] Optionally, in this embodiment, the first data and the second data may be understood as data read from a first physical storage block and data read from a second physical storage block.

[0065] S206, interleaving the first data and the second data to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

[0066] In an exemplary embodiment, Figure 5 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Figure 5 As shown, for a bank with a logical width of 512 bits (total_width), its width is too large and the local routing congestion is high. This application introduces a strategy of splitting multiple slices (physical storage blocks). Each slice is completely isomorphic. Each slice includes 64 storage blocks. The width of each storage block is 512 bits / N (N is the total number of slices). Here, N=2 is used as an example for explanation. The logical storage block with a width of 512 bits is divided into two physical storage blocks with a width of 256 bits. The upper 256 bits are distributed in slice[1] (corresponding to the first physical storage block mentioned above), and the lower 256 bits are distributed in slice[0] (corresponding to the second physical storage block mentioned above). In this way, the local wiring complexity is reduced by half, and except for the output data that needs to be interleaved between the slices, there is no interaction between the remaining logics, which is conducive to layout and routing.

[0067] It should be noted that the interleaving process does not require processing logic during physical implementation, but only requires rearranging the signal lines. Therefore, the wiring area consumed is very small, which can solve the problem of high wiring complexity of the above-mentioned on-chip storage system.

[0068] For example, a 512-bit wide logical storage block can be divided into two 256-bit wide physical storage blocks, where the high-order 256 bits are distributed in the first physical storage block and the low-order 256 bits are distributed in the second physical storage block. Since slice[0] stores the low-order data in the original logical storage block and slice[1] stores the high-order data in the original logical storage block, the two parts of data need to be interleaved when finally output. The interleaving process is as follows: Figure 5 Assume that the read data width of the request source is 2Kbit, slice[0] outputs 1Kbit of read data, slice[1] outputs 1Kbit of read data, and they are interleaved in units of 256 bits.

[0069] Through this embodiment, a first data processing request sent by a first master device is received, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, and the data processing system includes multiple storage block groups, and each of the storage block groups is set to be pre-routed by multiple logical storage blocks; in response to the first storage block group detecting that the first data processing request is valid, first data and second data are obtained from a first physical storage block and a second physical storage block respectively, wherein the first storage block group includes a first logical storage block, and the first logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, and the multiple physical storage blocks include the first physical storage block and the second physical storage block; interleaving the first data and the second data to generate first target data, wherein the first target data represents the data that the first data processing request expects to process, by pre-routing and converging multiple logical storage blocks to form the above-mentioned multiple storage block groups, and then forming a data processing system with multiple storage block groups to achieve a physical wiring-friendly high-bandwidth on-chip storage system design, by distributing ultra-large bit-width logical storage blocks to multiple physical storage blocks, and adopting a hierarchical convergence strategy, the local routing complexity of the wiring is greatly reduced, and it is beneficial to the convergence of the timing, thereby achieving the technical effect of improving data processing efficiency and reducing wiring complexity, thereby solving the technical problems of low data processing efficiency and high wiring complexity of the on-chip storage system in the related technology.

[0070] As an optional scheme, after receiving the first data processing request sent by the first master device, the method also includes: determining a first access address and a first data length according to the first data processing request, wherein the first access address is used to indicate the starting access address of the first target data, and the first data length represents the data length of the first target data; obtaining a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of multiple storage block groups, the number of storage blocks in each storage block group, and the data bandwidth of each storage block; determining first address information according to the first access address and the first data length, and determining that the first data processing request is valid when the first address information falls within the first storage address range.

[0071] Optionally, in this embodiment, the above-mentioned first data processing request includes relevant information for reading and writing data of at least one storage block group, and the first data processing request includes at least two parameters: start_addr and req_len, wherein start_addr represents the starting address of memory access (corresponding to the above-mentioned first access address), and req_len represents requestlength (corresponding to the above-mentioned first data length), that is, the request length.

[0072] It should be noted that the above-mentioned first data processing request can be a continuous data request. The above-mentioned two parameters can be used to determine the storage block group currently being accessed, and then the request selection module corresponding to the data request outputs the request to all storage block groups. For the storage block groups that need to be accessed, their input sub-requests are valid, and for the storage block groups that do not need to be accessed, their input sub-requests are invalid.

[0073] In an exemplary embodiment, Figure 6 is a schematic diagram of another optional data processing method according to an embodiment of the present application, in which the logical storage blocks are divided as follows Figure 6 As shown, the address increments corresponding to each storage block are as indicated by the arrows. The width of each logical storage block is 512 bits, that is, 64 bytes, and the address unit is Byte, so addresses 0 to 63 are in the first row of BGU[0].bank0, addresses 64 to 127 are in the first row of BGU[1].bank0, and so on. Addresses 960 to 1023 are in the first row of BGU

[15] .bank0; addresses 1024 to 1087 are in the second row of BGU[0].bank0..., and so on.

[0074] For example, taking the first access address as 62 and the first data length as 256Byte, the first address information represents 63 to 318. Therefore, the first address information falls into the first row of BGU[1].bank0, the first row of BGU[2].bank0, the first row of BGU[3].bank0, and the first row of BGU[4].bank0. Then, BGU[1] to BGU[4] determine that the first data processing request is valid, and the remaining BGUs determine that the first data processing request is invalid.

[0075] As an optional solution, Figure 7 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Figure 7 As shown, in response to the first storage block group detecting that the first data processing request is valid, obtaining the first data from the first physical storage block includes:

[0076] S702, determining the number of storage block groups that need to be accessed by the first data processing request according to the first data length, wherein the number of storage block groups is determined by a ratio of the first data length to a data bandwidth preset for the storage block;

[0077] S704, determining a starting storage block group according to the first access address, wherein the first storage block group includes the starting storage block group;

[0078] S706: Acquire first data from the first physical storage block according to the starting storage block group and the number of storage block groups.

[0079] Optionally, in this embodiment, the number of the above-mentioned storage block groups is determined by the ratio of the first data length to the data bandwidth pre-set for the storage block, which can be understood as first obtaining the data bandwidth pre-set for the non-storage block. For example, the data bandwidth pre-set for the logical storage block is set to 512 bits, that is, 64 bytes, and the address unit is Byte. When the first data length is 256 bytes, it can be understood that the number of the above-mentioned storage block groups is 256 / 64=4. At this time, if the first access address bit is 64 bytes, the above-mentioned BGU[1] is the above-mentioned starting storage block group.

[0080] It should be noted that the above-mentioned acquisition of the first data from the first physical storage block according to the starting storage block grouping and the number of storage block groups can be understood as dividing the logical storage block in BGU[0] into a first physical storage block and a second physical storage block, wherein the first physical storage block is used to store low-order data and the second physical storage block is used to store high-order data. The above-mentioned first data can be understood as the low-order data stored in the first physical storage block divided by the logical storage blocks in BGU[0].

[0081] As an optional solution, Figure 8 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Figure 8 As shown, obtaining first data from a first physical storage block according to a starting storage block group and a storage block group quantity includes:

[0082] S802, determining an ending storage block grouping according to a starting storage block grouping and the number of storage block groups;

[0083] S804, comparing the start group sequence number of the start storage block group and the end group sequence number of the end storage block group;

[0084] S806, when the sequence number of the terminating group is less than the sequence number of the starting group, determining that a plurality of storage block groups are wrapped, performing a circular shift operation on the plurality of storage block groups to move data corresponding to the first storage block group to a lower bit of a corresponding register, wherein the starting storage block group is arranged after the terminating storage block group;

[0085] S808: Acquire first data from a first physical storage block corresponding to the first storage block group.

[0086] Optionally, in this embodiment, the above determination of the terminating storage block grouping according to the starting storage block grouping and the number of storage block groups can be understood as starting with the above starting storage block grouping and determining the terminating storage block grouping after the above number of storage block groups.

[0087] Exemplarily, the starting storage block group is BGU[1], and the number of storage block groups is 4, then the above-mentioned ending storage block group is BGU[4].

[0088] Optionally, in this embodiment, the comparison of the start group sequence number of the start storage block group and the end group sequence number of the end storage block group can be understood as comparing the size relationship between the storage block group sequence numbers to determine whether the storage block group has a wrapping phenomenon.

[0089] For example, taking the starting storage block grouping as BGU[1] and the ending storage block grouping as BGU[4] as an example, since 4 is greater than 1, multiple storage block groups do not wrap around; taking the starting storage block grouping as BGU

[15] and the ending storage block grouping as BGU[2] as an example, since 2 is less than 15, multiple storage block groups wrap around.

[0090] Optionally, in this embodiment, if the above-mentioned multiple storage block groups are wrapped, a circular shift operation needs to be performed on the multiple storage block groups. The circular shift operation can be understood as converting a numerical value into binary and then performing a circular shift operation. When using the circular shift operation, the bits of a binary number are moved to the left or right by a certain number of bits, which is achieved through a bit operator.

[0091] It should be noted that the purpose of the above circular shift operation is to move the data corresponding to the first storage block group to the lower bits of the corresponding register, so as to facilitate the subsequent acquisition of the first data from the first physical storage block corresponding to the first storage block group.

[0092] In an exemplary embodiment, the BGU outputs the read data to the cyclic shift unit for cyclic shift processing, which is used to select the final read data. For example, the preset BGU is 16, and the maximum parallel bit width of each channel is 16*512bit=8Kbit. For the first data processing request, the read data is restored according to its first address information, including two situations: BGU wrapping occurs and BGU wrapping does not occur.

[0093] Fig. 9 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Fig. 9 As shown, for the case where there is no BGU wrap (corresponding to the aforementioned situation where no BGU wrap occurs), assuming that a certain access start address falls on BGU[9], its req_len=256Byte, that is, its access spans 4 BGUs, so the BGUs it accesses are 9~12, and the circular shift unit shifts the BGU[9] read data corresponding to its first address to the lowest bit.

[0094] Fig.10is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Fig.10 As shown, for the case of BGU wrapping (corresponding to the aforementioned BGU wrapping), assuming that a certain access start address falls on BGU

[14] , its req_len=256Byte, that is, its access spans 4 BGUs, so the BGUs it accesses are 14~15 and 0~1, and the circular shift unit shifts the BGU

[14] read data corresponding to its first address to the lowest bit. After the circular shift operation, the read data of BGU[0] is arranged behind BGU

[15] .

[0095] As an optional solution, the above method also includes:

[0096] receiving a second data processing request sent by a second master device, wherein the second data processing request includes second address information, and the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device;

[0097] In response to the second storage block group detecting that the second data processing request is valid, obtaining third data and fourth data from a third physical storage block and a fourth physical storage block respectively, wherein the second storage block group includes a second logical storage block, the second logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a third physical storage block and a fourth physical storage block;

[0098] The third data and the fourth data are interleaved to generate second target data, wherein the second target data represents data that the second data processing request expects to process.

[0099] Optionally, in this embodiment, the second main device and the first main device are different devices, for example, mobile phones (such as Android phones, iOS phones, etc.), laptops, tablet computers, PDAs, MIDs (Mobile Internet Devices), PADs, desktop computers, smart TVs, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, virtual reality (VR) terminals, augmented reality (AR) terminals, mixed reality (MR) terminals and other computer devices.

[0100] Optionally, in this embodiment, the above-mentioned second data processing request may include but is not limited to sending a read data request, a write data request, etc., and the above-mentioned second address information may include but is not limited to the storage address of the data that needs to be processed by the above-mentioned second data processing request, and the second address information may include but is not limited to being composed of a starting address and a data length.

[0101] Optionally, in this embodiment, the above-mentioned second storage block grouping can be understood as a storage block grouping composed of multiple logical storage blocks divided by the data processing system. For example, the data processing system includes 64 logical storage blocks, and the 64 logical storage blocks can be divided into 16 storage block groups according to preset rules. Each storage block group includes 4 logical storage blocks, and each logical storage block can be evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel. For example, the data width of a logical storage block is pre-defined as 512 bits. When it needs to be divided into 2 physical storage blocks, the data bandwidth of the third physical storage block is 256 bits, and the data bandwidth of the fourth physical storage block is 256 bits. The third physical storage block and the fourth physical storage block are interleaved in units of 256 bits.

[0102] Optionally, in this embodiment, the third data and the fourth data may be understood as data read from the third physical storage block and data read from the fourth physical storage block.

[0103] It should be noted that the request broadcasting unit of each storage block group can complete the broadcasting of the first data processing request and the second data processing request. Since each request input may access any bank in the BGU, the request needs to be broadcasted to the four banks. Each bank corresponds to a req sel (request select) unit for request selection.

[0104] As an optional scheme, the above method also includes: in case of receiving the first data processing request and the second data processing request, obtaining the first priority information and the second priority information, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; in case the first priority information is higher than the second priority information, detecting whether the first data processing request is valid; in case the first priority information is lower than the second priority information, detecting whether the second data processing request is valid; in case the first priority information is the same as the second priority information, randomly detecting whether the first data processing request or the second data processing request is valid.

[0105] Optionally, in this embodiment, the above request selection is based on two aspects: one is to determine whether the request currently broadcast to the bank is valid, and the judgment basis is whether the request address falls within the address range of the bank; the second is that if there are multiple requests (the first data processing request and the second data processing request) are valid, then select one from them to execute, and the selection basis can be selected according to the priority of the request, and the specific selection strategy can be determined according to the needs of the application scenario.

[0106] It should be noted that when a request is selected, the request address addr is latched into the register. If it is a write operation, the write data also needs to be latched. Then the request is sent to the storage block. Since the access to the storage block requires a clock cycle, the domain segment identifying the storage block in the address needs to be tapped for subsequent use in data selection between storage blocks.

[0107] As an optional solution, the above method also includes:

[0108] In response to the first storage block group detecting that the first data processing request is valid, using a first multiplexer to obtain first data and second data from a first logical storage block, and to obtain fifth data and sixth data from a third logical storage block, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request;

[0109] In response to the second storage block group detecting that the second data processing request is valid, a second multiplexer is used to obtain third data and fourth data from the second logical storage block, and to obtain seventh data and eighth data from the fourth logical storage block, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

[0110] Optionally, in this embodiment, the data processing system is further provided with a plurality of multiplexers, one multiplexer corresponds to one data processing request. Fig.11 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Fig.11 As shown, 4 muxes are included to select read access data of different request sources (corresponding to the aforementioned first data processing request and second data processing request, etc.) from the storage block. The specific number of muxes can be determined according to the maximum number of parallel read requests of the system. The data selected by mux is stored in d0~3 registers, corresponding to the request sources of master0~3 (corresponding to the aforementioned first master device and second master device, etc.).

[0111] As an optional solution, in response to the first storage block group detecting that the first data processing request is valid, before respectively acquiring the first data and the second data from the first physical storage block and the second physical storage block, the method further includes:

[0112] When the preset data width of the first logical storage block is M, the first logical storage block is divided into N physical storage blocks, wherein the data width of each physical storage block is M / N, and N and M are both positive integers.

[0113] In an exemplary embodiment, taking the preset data width of 512 bits as an example, due to its excessive width and high local routing congestion, a strategy of splitting multiple slices (multiple physical storage blocks) is introduced. Each slice is completely isomorphic, and each slice includes 64 banks, except that the width of each bank is 512 bits / N (N is the total number of slices). Here, N=2 is used as an example for explanation. The 512-bit wide logical storage block is divided into two 256-bit wide physical storage blocks, with the upper 256 bits distributed in slice[1] and the lower 256 bits distributed in slice[0]. In this way, the local wiring complexity is reduced by half, and except for the output data that needs to be interleaved between the slices, there is no interaction between the remaining logics, which is conducive to layout and routing.

[0114] Fig.12 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Fig.12 As shown in the figure, since slice[0] stores the low-order data in the original logical bank and slice[1] stores the high-order data in the original logical bank, these two parts of data need to be interleaved when finally output. Assuming that the read data width of the request source is 2Kbit, slice[0] outputs 1Kbit of read data and slice[1] outputs 1Kbit of read data. They are interleaved in units of 256 bits.

[0115] The following is a further explanation of this application with reference to specific examples:

[0116] In the architecture of modern processors, the design of the storage system is a key part. Since some computing scenarios require a lot of calculations, such as convolution operations, fully connected layer operations, pooling operations, etc. in each layer of the neural network, these operations are characterized by a very large amount of calculation and a certain degree of parallelism, which is more suitable for the processor architecture of single instruction multiple data.

[0117] The computing requirements of ultra-large parallelism place high demands on the bandwidth of the data storage unit in the processor. In common scalar or superscalar CPU processors, the data width is usually only 32 bits, while in AI processors, the data width can reach 1024 bits or even wider, which requires the memory access system to provide a read and write interface of corresponding width to match the high computing power of the computing unit in the AI ​​processor.

[0118] However, as the bandwidth of data processing systems continues to increase, the physical implementation of data processing systems faces huge challenges: the wiring complexity is too high, the number of connections per unit area is too large, and physical implementation has become a bottleneck in the design of AI processors. Under the demand for ultra-high bandwidth, how to design a data processing system that meets the timing and layout and wiring requirements is the key to AI processor design.

[0119] The present application proposes a physical wiring-friendly high-bandwidth on-chip storage system design, which distributes ultra-large bit-width logical storage blocks into multiple physical storage blocks and adopts a hierarchical aggregation strategy, which greatly reduces the local routing complexity of the wiring and is conducive to timing convergence.

[0120] The present application divides the logical storage body into multiple physical storage bodies and multiple independent slices. The routing complexity of a single slice is only 1 / N (where N is the total number of slices), which greatly reduces the wiring complexity of local interconnections. The present application groups banks in a hierarchical aggregation manner, first aggregating the data within the bank group, and then aggregating the data between bank groups, cutting long lines into short lines, and greatly reducing the difficulty of timing convergence. The present application solves the congestion problem, can further compress the chip area, and improve the area utilization of the chip.

[0121] In an AI processor, the data processing system is a key component. The physically wired-friendly high-bandwidth on-chip storage system design proposed in this application reduces the overall layout and wiring complexity, can run at a relatively high main frequency in terms of timing, and consumes a small chip area, effectively improving the competitiveness of the product.

[0122] The overall architecture of the present application supports multiple master device request sources (corresponding to the aforementioned first master device or second master device), such as Fig.11As shown, each master request (corresponding to the aforementioned first data processing request or the second data processing request) enters the master req split unit, and converts the original read and write request into a request for a different BGU (corresponding to the aforementioned storage block grouping). The original master request includes two parameters: start_addr (corresponding to the aforementioned starting access address) and req_len (corresponding to the aforementioned first data length). Start_addr represents the starting address of memory access, and req_len represents the request length, that is, the request length. All accesses here are continuous data requests. The above parameters can be used to determine the BGU (corresponding to the aforementioned first storage block grouping) that is currently being accessed. The master req split module outputs requests to all BGUs. For the BGU that needs to be accessed, its input sub-request is valid, and for the BGU that does not need to be accessed, its input sub-request is invalid.

[0123] In an exemplary embodiment, BGU is a collection of banks (corresponding to the aforementioned first logical storage block), and here, a 512KByte on-chip cache is used as an example for explanation, which is logically divided into 64 banks, and the capacity of each bank is 512KB / 64=8KB. The data width of each bank is defined as 512bit, and the depth of each bank is 128. Here, every 4 banks are divided into a group, and the specific number of banks and the bank group grouping method can be adapted according to different scenarios and can be arbitrarily expanded. Each bank is 512 bits wide, or 64 bytes, and the address unit is Byte, so addresses 0 to 63 are in the first row of BGU[0].bank0, addresses 64 to 127 are in the first row of BGU[1].bank0, and so on, addresses 960 to 1023 are in the first row of BGU

[15] .bank0; addresses 1024 to 1087 are in the second row of BGU[0].bank0...; the req brcast (request broadcast) unit in BGU completes the request broadcast. Since each request input may access any bank in BGU, the request needs to be broadcast to the four banks. Each bank corresponds to a req sel (request select) unit to select the request.

[0124] It should be noted that the above request selection can be based on but not limited to the following two aspects: first, judging whether the request currently broadcast to the bank is valid, and the judgment basis is whether the request address falls within the address range of the bank; second, if there are multiple requests that are valid, select one from them for execution, and the selection basis can be selected according to the priority of the request. The specific selection strategy can be determined according to the needs of the application scenario.

[0125] First, after selecting a request, the request address addr is latched into the register. If it is a write operation, the write data also needs to be latched. Then the request is sent to the bank. Since the bank access requires a clock cycle, the field segment identifying the bank id in the address needs to be tapped for subsequent inter-bank data selection.

[0126] Secondly, Fig.11 Also shown are multiple muxes (corresponding to the aforementioned multiplexers) for selecting read access data from different request sources from the bank. The specific number of muxes can be determined based on the maximum number of parallel read requests in the system. The data selected by the mux is stored in registers d0 to 3, corresponding to the four request sources master0 to 3 respectively.

[0127] It should be noted that BGU outputs read data to the rotate unit for cyclic shift (corresponding to Fig.11 The function of the process is to select the final read data. Since 16 BGUs are set, the maximum parallel bit width of each channel is 16*512bit=8Kbit. For each master request, the read data needs to be restored according to its start address. There are two cases: with BGU wrapping and without BGU wrapping.

[0128] Among them, for the case without BGU wrapping, assuming that the starting address of an access falls on BGU[9], its req_len=256Byte, that is, its access spans 4 BGUs, so the BGUs it accesses are 9~12, and the rotate unit shifts the BGU[9] read data corresponding to its first address to the lowest bit.

[0129] Among them, for the case of BGU wrapping, assuming that the starting address of an access falls on BGU

[14] , its req_len = 256Byte, that is, its access spans 4 BGUs, so the BGUs it accesses are 14~15 and 0~1. The rotate unit shifts the BGU

[14] read data corresponding to its first address to the lowest bit. After rotation, the read data of BGU[0] is arranged behind BGU

[15] .

[0130] In an exemplary embodiment, for a bank with a logical width of 512 bits (total_width), its width is too large and the local routing congestion is high. This application introduces a strategy of splitting multiple slices (corresponding to the aforementioned multiple physical storage blocks). Each slice is completely isomorphic and each slice also includes 64 banks, but the width of each bank is 512 bits / N (N is the total number of slices). Here, N=2 is used as an example for explanation. The logical bank with a width of 512 bits is divided into two physical banks with a width of 256 bits. The upper 256 bits are distributed in slice[1] (corresponding to the aforementioned first physical storage block), and the lower 256 bits are distributed in slice[0] (corresponding to the aforementioned second physical storage block). In this way, the local wiring complexity is reduced by half, and except for the output data that needs to be interleaved between the slices, there is no interaction between the remaining logics, which is conducive to layout and routing.

[0131] Since slice[0] stores the low-order data in the original logical bank and slice[1] stores the high-order data in the original logical bank, these two parts of data need to be interleaved during final output. Assuming that the read data width of the request source is 2Kbit, slice[0] outputs 1Kbit of read data and slice[1] outputs 1Kbit of read data. They are interleaved in units of 256 bits.

[0132] It should be noted that the interleaving process does not require any processing logic during physical implementation, but only requires the signal lines to be rearranged, which consumes very little area.

[0133] Through this embodiment, by utilizing the division of multiple slices, high-bandwidth read and write data routing is dispersed into each slice, there is no excessive interaction between each slice, and the routing complexity within the slice is greatly reduced. A two-level convergence strategy is adopted to divide 64 banks into 16 bank groups. First, routing is converged in the bank group, and then rotated between bank groups to finally obtain read data. This two-level convergence method shortens the routing transmission path, is timing-friendly, and can run at a higher main frequency. The routing complexity of this application is low. Under a read-write interface with a width of 2048bit, the wiring tool can successfully complete the wiring work. Moreover, after physical implementation of this application, the timing can converge to but not limited to a main frequency of 1.5GHz, and the performance is high.

[0134] It is understandable that in the specific implementation of this application, related data such as user information is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0135] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0136] According to another aspect of the embodiment of the present application, a data processing system for implementing the above data processing method is also provided. Fig.13 As shown, the system includes:

[0137] A first host device 1302 is connected to a data storage device and is used to send a first data processing request, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in the data processing system.

[0138] The data storage device 1304 includes a plurality of storage block groups, each of which is configured to be pre-wired by a plurality of logic storage blocks, and the plurality of storage block groups includes a first storage block group;

[0139] The first storage block group 1306 is used to obtain first data and second data from a first physical storage block and a second physical storage block respectively in response to the first storage block group detecting that the first data processing request is valid, wherein the first storage block group includes a first logical storage block, and the first logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, and the multiple physical storage blocks include the first physical storage block and the second physical storage block, and the first data and the second data are used for interleaving processing to generate first target data, and the first target data represents the data that the first data processing request expects to process.

[0140] As an optional solution, the system also includes: a reading unit, connected to the data storage device, for determining a first access address and a first data length according to a first data processing request sent by a first master device after receiving the first data processing request, wherein the first access address is used to indicate a starting access address of the first target data, and the first data length represents a data length of the first target data; obtaining a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of the multiple storage block groups, the number of storage blocks in each of the storage block groups, and the data bandwidth of each storage block; a computing unit, connected to the reading unit, for determining the first address information according to the first access address and the first data length, and determining that the first data processing request is valid when the first address information falls within the first storage address range.

[0141] As an optional scheme, the system also includes: a starting storage block grouping determination unit, connected to the reading unit, and used to determine the number of storage block groups that need to be accessed by the first data processing request according to the first data length, wherein the number of storage block groups is determined by the ratio of the first data length to a data bandwidth pre-set for the storage block; determining the starting storage block group according to the first access address, wherein the first storage block group includes the starting storage block group; a data processing unit, connected to the starting storage block grouping determination unit, and used to obtain the first data from the first physical storage block according to the starting storage block group and the number of storage block groups.

[0142] As an optional scheme, the data processing unit includes: a termination storage block group calculation subunit, used to determine the termination storage block group according to the starting storage block group and the number of storage block groups; a group sequence number comparison subunit, used to compare the starting group sequence number of the starting storage block group and the termination group sequence number of the termination storage block group; a circular shift subunit, used to determine that the multiple storage block groups are wrapped when the termination group sequence number is less than the starting group sequence number, and perform a circular shift operation on the multiple storage block groups to move the data corresponding to the first storage block group to the low bit of the corresponding register, wherein the starting storage block group is arranged after the termination storage block group; a data acquisition subunit, used to obtain the first data from the first physical storage block corresponding to the first storage block group when the data corresponding to the first storage block group is moved to the low bit of the corresponding register.

[0143] As an optional solution, the system also includes: a second master device, connected to the data storage device, and used to send a second data processing request, wherein the second data processing request includes second address information, the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device; a second storage block group, used to respond to the second storage block group to detect that the second data processing request is valid, and obtain third data and fourth data from a third physical storage block and a fourth physical storage block respectively, wherein the multiple storage block groups include a second storage block group, the second storage block group includes a second logical storage block, the second logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, the multiple physical storage blocks include the third physical storage block and the fourth physical storage block, the third data and the fourth data are used for interleaving processing to generate second target data, and the second target data represents the data that the second data processing request expects to process.

[0144] As an optional scheme, the system also includes: a priority information acquisition unit, used to acquire first priority information and second priority information when the first data processing request and the second data processing request are received, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; a detection unit, used to detect whether the first data processing request is valid when the first priority information is higher than the second priority information; detect whether the second data processing request is valid when the first priority information is lower than the second priority information; and randomly detect whether the first data processing request or the second data processing request is valid when the first priority information is the same as the second priority information.

[0145] As an optional scheme, the system also includes: a first multiplexer, used to respond to the first storage block group detecting that the first data processing request is valid, obtain the first data and the second data from the first logical storage block, and obtain the fifth data and the sixth data from the third logical storage block, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request; a second multiplexer, used to respond to the second storage block group detecting that the second data processing request is valid, obtain the third data and the fourth data from the second logical storage block, and obtain the seventh data and the eighth data from the fourth logical storage block, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

[0146] As an optional scheme, the first logical storage block includes: N physical storage blocks, wherein the first logical storage block includes the N physical storage blocks, and when the preset data width of the first logical storage block is M, the data width of each of the N physical storage blocks is M / N, and N and M are both positive integers.

[0147] According to another aspect of the embodiment of the present application, a data processing device for implementing the above data processing method is also provided. Fig.14 As shown, the device comprises:

[0148] The receiving module 1402 is configured to receive a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, wherein the data processing system includes a plurality of storage block groups, and each storage block group is configured to be pre-routing converged by a plurality of logical storage blocks;

[0149] The acquisition module 1404 is configured to acquire the first data and the second data from the first physical storage block and the second physical storage block respectively in response to the first storage block group detecting that the first data processing request is valid, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a first physical storage block and a second physical storage block;

[0150] The processing module 1406 is configured to perform interleaving processing on the first data and the second data to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

[0151] As an optional scheme, the above-mentioned device is also used to: after receiving a first data processing request sent by a first master device, determine a first access address and a first data length according to the first data processing request, wherein the first access address is used to indicate the starting access address of the first target data, and the first data length represents the data length of the first target data; obtain a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of multiple storage block groups, the number of storage blocks in each storage block group, and the data bandwidth of each storage block; determine first address information according to the first access address and the first data length, and determine that the first data processing request is valid when the first address information falls within the first storage address range.

[0152] As an optional solution, the above-mentioned device is used to obtain the first data from the first physical storage block in response to the first storage block group detecting that the first data processing request is valid in the following manner: determining the number of storage block groups that need to be accessed for the first data processing request based on the first data length, wherein the number of storage block groups is determined by the ratio of the first data length to the data bandwidth pre-set for the storage block; determining the starting storage block group based on the first access address, wherein the first storage block group includes the starting storage block group; and obtaining the first data from the first physical storage block based on the starting storage block group and the number of storage block groups.

[0153] As an optional scheme, the above-mentioned device is used to obtain the first data from the first physical storage block according to the starting storage block grouping and the number of storage block groups in the following manner: determine the ending storage block group according to the starting storage block grouping and the number of storage block groups; compare the starting group sequence number of the starting storage block group and the ending group sequence number of the ending storage block group; when the ending group sequence number is smaller than the starting group sequence number, determine that multiple storage block groups are wrapped, perform a circular shift operation on the multiple storage block groups to move the data corresponding to the first storage block group to the low bit of the corresponding register, wherein the starting storage block group is arranged after the ending storage block group; and obtain the first data from the first physical storage block corresponding to the first storage block group.

[0154] As an optional solution, the above-mentioned device is also used to: receive a second data processing request sent by a second master device, wherein the second data processing request includes second address information, the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device; in response to the second storage block group detecting that the second data processing request is valid, obtain third data and fourth data from a third physical storage block and a fourth physical storage block respectively, wherein the second storage block group includes a second logical storage block, and the second logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, and the multiple physical storage blocks include a third physical storage block and a fourth physical storage block; interleave the third data and the fourth data to generate second target data, wherein the second target data represents the data that the second data processing request expects to process.

[0155] As an optional scheme, the above-mentioned device is also used to: when receiving the first data processing request and the second data processing request, obtain the first priority information and the second priority information, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; when the first priority information is higher than the second priority information, detect whether the first data processing request is valid; when the first priority information is lower than the second priority information, detect whether the second data processing request is valid; when the first priority information is the same as the second priority information, randomly detect whether the first data processing request or the second data processing request is valid.

[0156] As an optional scheme, the above-mentioned device is also used to: in response to the first storage block group detecting that the first data processing request is valid, use the first multiplexer to obtain the first data and the second data from the first logical storage block, and obtain the fifth data and the sixth data from the third logical storage block, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request; in response to the second storage block group detecting that the second data processing request is valid, use the second multiplexer to obtain the third data and the fourth data from the second logical storage block, and obtain the seventh data and the eighth data from the fourth logical storage block, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

[0157] As an optional scheme, the above-mentioned device is also used to: in response to the first storage block grouping detecting that the first data processing request is valid, before obtaining the first data and the second data from the first physical storage block and the second physical storage block respectively, when the preset data width of the first logical storage block is M, divide the first logical storage block into N physical storage blocks, wherein the data width of each physical storage block is M / N, and N and M are both positive integers.

[0158] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0159] According to one aspect of the present application, a computer program product is provided, the computer program product comprising a computer program / instruction, the computer program / instruction comprising a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 1509, and / or installed from a removable medium 1511. When the computer program is executed by the central processor 1501, various functions provided in the embodiments of the present application are executed.

[0160] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0161] Fig.15 The structure block diagram of a computer system for implementing an electronic device according to an embodiment of the present application is schematically shown.

[0162] It should be noted that Fig.15 The computer system 1500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0163] like Fig.15 As shown, the computer system 1500 includes a central processing unit 1501 (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory 1502 (ROM) or the program loaded from the storage part 1508 to the random access memory 1503 (RAM). Various programs and data required for system operation are also stored in the random access memory 1503. The central processing unit 1501, the read-only memory 1502 and the random access memory 1503 are connected to each other through a bus 1504. The input / output interface 1505 (Input / Output interface, i.e., I / O interface) is also connected to the bus 1504.

[0164] The following components are connected to the input / output interface 1505: an input section 1506 including a keyboard, a mouse, etc.; an output section 1507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1508 including a hard disk, etc.; and a communication section 1509 including a network interface card such as a LAN card, a modem, etc. The communication section 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to the input / output interface 1505 as needed. A removable medium 1511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1510 as needed so that a computer program read therefrom is installed into the storage section 1508 as needed.

[0165] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1509, and / or installed from the removable medium 1511. When the computer program is executed by the central processor 1501, various functions defined in the system of the present application are executed.

[0166] According to another aspect of the embodiment of the present application, an electronic device for implementing the above data processing method is also provided. The electronic device may be Figure 1 The terminal device or server shown in the figure. This embodiment is described by taking the electronic device as a terminal device as an example. Fig.16 As shown, the electronic device includes a memory 1602 and a processor 1604. The memory 1602 stores a computer program, and the processor 1604 is configured to execute the steps in any of the above method embodiments through the computer program.

[0167] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0168] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0169] S1, receiving a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, the data processing system includes a plurality of storage block groups, and each storage block group is configured to be pre-routing converged by a plurality of logical storage blocks;

[0170] S2, in response to the first storage block group detecting that the first data processing request is valid, obtaining the first data and the second data from the first physical storage block and the second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a first physical storage block and a second physical storage block;

[0171] S3, interleaving the first data and the second data to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

[0172] Alternatively, a person skilled in the art may understand that: Fig.16The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (MID), a PAD, and other terminal devices. Fig.16 The electronic device and the electronic equipment described above are not limited in structure. Fig.16 More or fewer components (such as network interfaces, etc.) as shown in, or with Fig.16 Different configurations are shown.

[0173] Among them, the memory 1602 can be used to store software programs and modules, such as program instructions / modules corresponding to the data processing method and device in the embodiments of the present application. The processor 1604 executes various functional applications and data processing by running the software programs and modules stored in the memory 1602, that is, realizing the above-mentioned data processing method. The memory 1602 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1602 may further include a memory remotely located relative to the processor 1604, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 1602 can be specifically, but not limited to, used to store information such as data. As an example, such as Fig.16 As shown, the memory 1602 may include but is not limited to the first host device 1302, the data processing system 1304 and the first storage block group 1306 in the data processing system. In addition, other module units in the data processing system may also be included but are not limited to, which will not be repeated in this example.

[0174] Optionally, the transmission device 1606 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wired network and a wireless network. In one example, the transmission device 1606 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 1606 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0175] In addition, the electronic device further includes: a display 1608 for displaying the data request to be processed; and a connection bus 1610 for connecting various module components in the electronic device.

[0176] In other embodiments, the terminal device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. Among them, the nodes may form a peer-to-peer (P2P, Peer To Peer) network, and any form of computing device, such as a server, terminal and other electronic devices, may become a node in the blockchain system by joining the peer-to-peer network.

[0177] According to one aspect of the present application, a computer-readable storage medium is provided, and a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the data processing method provided in various optional implementations of the above-mentioned data processing aspects.

[0178] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0179] S1, receiving a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, the data processing system includes a plurality of storage block groups, and each storage block group is configured to be pre-routing converged by a plurality of logical storage blocks;

[0180] S2, in response to the first storage block group detecting that the first data processing request is valid, obtaining the first data and the second data from the first physical storage block and the second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include a first physical storage block and a second physical storage block;

[0181] S3, interleaving the first data and the second data to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

[0182] Optionally, in this embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing hardware related to the terminal device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0183] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0184] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0185] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0186] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0187] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0188] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0189] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A data processing method, characterized in that: include: Receive a first data processing request sent by a first master device, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in a data processing system, the data processing system includes a plurality of storage block groups, and each of the storage block groups is configured to be pre-routing converged by a plurality of logical storage blocks; In response to the first storage block group detecting that the first data processing request is valid, obtaining first data and second data from a first physical storage block and a second physical storage block respectively, wherein the first storage block group includes a first logical storage block, the first logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include the first physical storage block and the second physical storage block; The first data and the second data are interleaved to generate first target data, wherein the first target data represents data that the first data processing request expects to process.

2. The method according to claim 1, characterized in that After receiving the first data processing request sent by the first master device, the method further includes: Determine a first access address and a first data length according to the first data processing request, wherein the first access address is used to indicate a starting access address of the first target data, and the first data length indicates a data length of the first target data; Acquire a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of the plurality of storage block groups, the number of storage blocks in each of the storage block groups, and the data bandwidth of each storage block; The first address information is determined according to the first access address and the first data length, and when the first address information falls within the first storage address range, it is determined that the first data processing request is valid.

3. The method according to claim 2, characterized in that The step of acquiring the first data from the first physical storage block in response to the first storage block group detecting that the first data processing request is valid comprises: Determining the number of storage block groups that the first data processing request needs to access according to the first data length, wherein the number of storage block groups is determined by a ratio of the first data length to a data bandwidth preset for a storage block; Determining a starting storage block group according to the first access address, wherein the first storage block group includes the starting storage block group; The first data is acquired from the first physical storage block according to the starting storage block group and the number of storage block groups.

4. The method according to claim 3, characterized in that The acquiring the first data from the first physical storage block according to the starting storage block group and the number of storage block groups comprises: Determine an ending storage block grouping according to the starting storage block grouping and the number of storage block groups; Comparing the starting group sequence number of the starting storage block group and the ending group sequence number of the ending storage block group; In the case where the end group sequence number is less than the start group sequence number, determining that the multiple storage block groups are wrapped, performing a circular shift operation on the multiple storage block groups to move data corresponding to the first storage block group to a low bit of a corresponding register, wherein the start storage block group is arranged after the end storage block group; The first data is obtained from the first physical storage block corresponding to the first storage block group.

5. The method according to claim 1, characterized in that The method further comprises: receiving a second data processing request sent by a second master device, wherein the second data processing request includes second address information, the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device; In response to the second storage block group detecting that the second data processing request is valid, obtaining third data and fourth data from a third physical storage block and a fourth physical storage block respectively, wherein the second storage block group includes a second logical storage block, and the second logical storage block is evenly divided into a plurality of physical storage blocks according to a preset data bandwidth to process data in parallel, and the plurality of physical storage blocks include the third physical storage block and the fourth physical storage block; The third data and the fourth data are interleaved to generate second target data, wherein the second target data represents data that the second data processing request expects to process.

6. The method according to claim 5, characterized in that The method further comprises: Upon receiving the first data processing request and the second data processing request, acquiring first priority information and second priority information, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; In a case where the first priority information is higher than the second priority information, detecting whether the first data processing request is valid; In a case where the first priority information is lower than the second priority information, detecting whether the second data processing request is valid; In a case where the first priority information is the same as the second priority information, whether the first data processing request or the second data processing request is valid is randomly detected.

7. The method according to claim 5, characterized in that The method further comprises: In response to the first storage block group detecting that the first data processing request is valid, using a first multiplexer to obtain the first data and the second data from the first logical storage block, and to obtain fifth data and sixth data from a third logical storage block, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request; In response to the second storage block group detecting that the second data processing request is valid, a second multiplexer is used to obtain the third data and the fourth data from the second logical storage block, and to obtain the seventh data and the eighth data from the fourth logical storage block, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

8. The method according to claim 1, characterized in that Before the first data processing request is detected to be valid in response to the first storage block group, and the first data and the second data are respectively acquired from the first physical storage block and the second physical storage block, the method further comprises: When the preset data width of the first logical storage block is M, the first logical storage block is divided into N physical storage blocks, wherein the data width of each physical storage block is M / N, and N and M are both positive integers.

9. A data processing system, characterized in that: include: A first master device, connected to the data storage device, and configured to send a first data processing request, wherein the first data processing request includes first address information, and the first address information is used to specify at least one logical storage block in the data processing system; The data storage device comprises a plurality of storage block groups, each of the storage block groups is configured to be pre-wired by a plurality of logic storage blocks, and the plurality of storage block groups comprises a first storage block group; The first storage block group is used to obtain first data and second data from a first physical storage block and a second physical storage block respectively in response to detecting that the first data processing request is valid, wherein the first storage block group includes a first logical storage block, and the first logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, and the multiple physical storage blocks include the first physical storage block and the second physical storage block, and the first data and the second data are used for interleaving processing to generate first target data, and the first target data represents the data that the first data processing request expects to process.

10. The system according to claim 9, characterized in that The system further comprises: a reading unit connected to the data storage device, and configured to determine a first access address and a first data length according to a first data processing request sent by a first master device after receiving the first data processing request, wherein the first access address is used to indicate a starting access address of the first target data, and the first data length indicates a data length of the first target data; and obtain a first storage address range corresponding to the first logical storage block, wherein the first storage address range is determined by the number of the plurality of storage block groups, the number of storage blocks in each of the storage block groups, and the data bandwidth of each storage block; A calculation unit is connected to the reading unit, and is used to determine the first address information according to the first access address and the first data length, and determine that the first data processing request is valid when the first address information falls within the first storage address range.

11. The system according to claim 10, characterized in that The system further comprises: a starting storage block group determining unit, connected to the reading unit, and configured to determine the number of storage block groups that need to be accessed by the first data processing request according to the first data length, wherein the number of storage block groups is determined by a ratio of the first data length to a data bandwidth preset for a storage block; and determine a starting storage block group according to the first access address, wherein the first storage block group includes the starting storage block group; A data processing unit is connected to the starting storage block grouping determination unit and is used to obtain the first data from the first physical storage block according to the starting storage block grouping and the number of storage block groups.

12. The system according to claim 11, characterized in that The data processing unit comprises: a termination storage block grouping calculation subunit, configured to determine a termination storage block grouping according to the start storage block grouping and the number of storage block groups; A group number comparison subunit, used for comparing the start group number of the start storage block group with the end group number of the end storage block group; a cyclic shift subunit, configured to determine, when the sequence number of the terminating group is less than the sequence number of the starting group, that the multiple storage block groups are wrapped, and to perform a cyclic shift operation on the multiple storage block groups to move data corresponding to the first storage block group to a low bit of a corresponding register, wherein the starting storage block group is arranged after the terminating storage block group; The data acquisition subunit is used to acquire the first data from the first physical storage block corresponding to the first storage block group when the data corresponding to the first storage block group is moved to the lower bit of the corresponding register.

13. The system according to claim 9, characterized in that The system further comprises: a second master device connected to the data storage device, and configured to send a second data processing request, wherein the second data processing request includes second address information, and the second address information is used to specify at least one logical storage block in the data processing system, and the second master device is different from the first master device; A second storage block grouping is used to obtain third data and fourth data from a third physical storage block and a fourth physical storage block respectively in response to the second storage block grouping detecting that the second data processing request is valid, wherein the multiple storage block groups include a second storage block group, the second storage block group includes a second logical storage block, the second logical storage block is evenly divided into multiple physical storage blocks according to a preset data bandwidth to process data in parallel, the multiple physical storage blocks include the third physical storage block and the fourth physical storage block, the third data and the fourth data are used for interleaving processing to generate second target data, and the second target data represents the data that the second data processing request expects to process.

14. The system according to claim 13, characterized in that The system further comprises: a priority information acquisition unit, configured to acquire first priority information and second priority information upon receiving the first data processing request and the second data processing request, wherein the first priority information is used to indicate the priority of the first data processing request, and the second priority information is used to indicate the priority of the second data processing request; A detection unit is used to detect whether the first data processing request is valid when the first priority information is higher than the second priority information; to detect whether the second data processing request is valid when the first priority information is lower than the second priority information; and to randomly detect whether the first data processing request or the second data processing request is valid when the first priority information is the same as the second priority information.

15. The system according to claim 13, characterized in that The system further comprises: a first multiplexer, configured to obtain the first data and the second data from the first logical storage block, and obtain the fifth data and the sixth data from the third logical storage block in response to the first storage block group detecting that the first data processing request is valid, wherein the first storage block group includes the third logical storage block, and the first multiplexer corresponds to the first data processing request; A second multiplexer is used for acquiring the third data and the fourth data from the second logical storage block and acquiring the seventh data and the eighth data from the fourth logical storage block in response to the second storage block group detecting that the second data processing request is valid, wherein the second storage block group includes the fourth logical storage block, and the second multiplexer corresponds to the second data processing request.

16. The system according to claim 9, characterized in that The first logic storage block includes: N physical storage blocks, wherein the first logical storage block includes the N physical storage blocks, and when the preset data width of the first logical storage block is M, the data width of each of the N physical storage blocks is M / N, and N and M are both positive integers.

17. A data processing chip, characterized in that: The data processing chip comprises the data processing system described in any one of claims 9 to 16.

Citation Information

Cited By

  • Data processing method, flash memory device and data processing system

    CN120469936A