Data control methods, data processing methods and related equipment

Abstract

This application relates to the field of data processing technology, and provides a data control method, a data processing method, an electronic device, and a storage medium. The data control method operates in an engine device, and the method includes: acquiring data to be processed; receiving operating status reports from multiple accelerator devices; determining multiple target accelerator devices from the multiple accelerator devices based on the data to be processed and the operating status; segmenting the data to be processed according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed; and sending each sub-data to be processed to the corresponding target accelerator device for processing. The method provided by this application improves the efficiency of data processing.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data control method, a data processing method, an electronic device, and a storage medium. Background Technology

[0002] In the field of artificial intelligence, especially in areas such as artificial neural networks, machine vision, and machine learning, accelerator devices are often used for acceleration. Accelerator devices are specialized hardware devices or computer systems designed to accelerate artificial intelligence applications, such as artificial intelligence (AI) accelerator devices. Currently, when using entry-level network topologies in AI applications, such as ResNet and MobileNet, accelerator devices perform well. However, when faced with more complex network topologies, accelerator devices often cannot meet the acceleration requirements, resulting in lower acceleration efficiency. Summary of the Invention

[0003] In view of this, the main objective of this application is to provide a data control method, a data processing method, an electronic device, and a storage medium, aiming to solve the technical problem of how to improve data processing efficiency.

[0004] The first aspect of this application provides a data control method, which operates in an engine device, the data control method comprising:

[0005] Obtain the data to be processed;

[0006] Receive operational status reports from multiple accelerator devices;

[0007] Based on the data to be processed and the operating status, multiple target accelerator devices are determined from among the multiple accelerator devices;

[0008] The data to be processed is segmented according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed.

[0009] Each piece of data to be processed is sent to the corresponding target accelerator device for processing.

[0010] According to an optional embodiment of this application, determining the plurality of target accelerator devices among the plurality of accelerator devices based on the data to be processed and the operating conditions includes:

[0011] Calculate the load on each accelerator device based on the described operating conditions;

[0012] Based on the load, a plurality of candidate accelerator devices are determined from the plurality of accelerator devices;

[0013] Based on the data to be processed, multiple target accelerator devices are determined from among the multiple candidate accelerator devices.

[0014] According to an optional embodiment of this application, determining a plurality of target accelerator devices from the plurality of candidate accelerator devices based on the data to be processed includes:

[0015] Based on the data to be processed, calculate the acceleration amount;

[0016] Based on the acceleration amount, a plurality of target accelerator devices are determined from the plurality of candidate accelerator devices.

[0017] According to an optional embodiment of this application, the step of segmenting the data to be processed based on the operating status of the plurality of target accelerator devices to obtain processed data includes:

[0018] Obtain the preset load threshold for each target accelerator device;

[0019] The target acceleration amount is calculated based on the operating status of each target accelerator device and the corresponding preset load threshold.

[0020] The data to be processed is divided into multiple sub-data based on the target acceleration amount.

[0021] According to an optional embodiment of this application, the step of segmenting the data to be processed according to the target acceleration amount to obtain multiple sub-data to be processed includes:

[0022] Obtain the data tags of the data to be processed;

[0023] The data to be processed is segmented according to the target acceleration amount and the data label to obtain multiple sub-data to be processed.

[0024] According to an optional embodiment of this application, the method further includes:

[0025] When multiple target accelerator devices cannot be determined among the multiple accelerator devices based on the data to be processed and the operating conditions, the data to be processed is processed in the engine device.

[0026] According to an optional embodiment of this application, after sending each sub-data to be processed to the corresponding target accelerator device for processing, the method further includes:

[0027] Obtain the processing results of the multiple target accelerator devices on the multiple sub-data to be processed;

[0028] A processing report for the data to be processed is generated based on the processing results.

[0029] A second aspect of this application provides a data processing method, operating in an accelerator device, the data control method comprising:

[0030] In response to the data processing command sent by the engine device, determine whether the calculation requirements corresponding to the data processing command are met;

[0031] When the computational requirements are not met, the operating status of the accelerator device is reported to the engine device.

[0032] When the computational requirements are met, the data corresponding to the data processing instruction is accelerated, and the acceleration result is sent to the engine device, which is used to execute the data control method described above.

[0033] A third aspect of this application provides an electronic device, the electronic device comprising:

[0034] Memory for storing at least one instruction;

[0035] A processor, configured to implement the data control method or the data processing method as described above when executing the at least one instruction.

[0036] A fourth aspect of this application provides a computer-readable storage medium storing at least one instruction that, when executed by a processor, implements the data control method or the data processing method described above.

[0037] As can be seen from the above technical solution, this application obtains the data to be processed and the operating status of multiple accelerator devices, and determines multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status; then, it segments the data to be processed according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed; finally, it sends each sub-data to be processed to the corresponding target accelerator device for processing. By segmenting the data to be processed according to the operating status of multiple target accelerator devices, it is determined that each target accelerator device can process the corresponding sub-data to be processed, thereby improving the acceleration efficiency. Attached Figure Description

[0038] Figure 1 This is a schematic flowchart illustrating a data control method according to an embodiment of this application;

[0039] Figure 2 A schematic diagram illustrating a scenario in which the data to be processed is divided into multiple sub-data to be processed in an embodiment of this application;

[0040] Figure 3 This is a schematic flowchart illustrating a data processing method according to an embodiment of this application;

[0041] Figure 4 This is a schematic block diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0043] The schematic flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be executed in the described order. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0044] This application provides a data control method, a data processing method, an electronic device, and a computer-readable storage medium.

[0045] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0046] Please see Figure 1 , Figure 1 This is a schematic flowchart of a data control method provided in an embodiment of this application.

[0047] like Figure 1 As shown, the data control method specifically includes steps S11 to S15. Depending on different requirements, the order of the steps in this schematic flowchart can be changed, and some steps can be omitted. This data control method runs in an engine device, which includes an inference engine.

[0048] Step S11: Obtain the data to be processed.

[0049] Acquire data to be processed, which may include image data and / or document data.

[0050] In some embodiments, the engine device can establish communication with a target device to receive data sent by the target device and integrate the data to obtain data to be processed. The engine device can establish communication with one or more target devices.

[0051] For example, integrating the data to obtain the data to be processed includes: integrating the data according to its importance to obtain the data to be processed. For example, the importance of the data can be determined according to its data tags, or it can be determined according to its execution time; for example, the earlier the execution time, the higher the corresponding importance.

[0052] The data can be sorted according to its importance, and then integrated according to the sorting order to obtain the data to be processed. The data with higher importance is presented first, followed by the data with lower importance, thereby accelerating the processing speed.

[0053] Step S12: Receive the operating status reported by multiple accelerator devices.

[0054] For example, the engine device obtains the operational status of multiple accelerator devices that have undergone device authentication, further ensuring the security of acceleration. The device authentication includes the accelerator devices registering with the engine device.

[0055] Step S13: Determine multiple target accelerator devices from among the multiple accelerator devices based on the data to be processed and the operating status.

[0056] Based on the amount of data to be processed and the operating status of the multiple accelerator devices, multiple target accelerator devices are determined from among the multiple accelerator devices.

[0057] In some implementations, determining multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status includes:

[0058] Calculate the load on each accelerator device based on the described operating conditions;

[0059] Based on the load, a plurality of candidate accelerator devices are determined from the plurality of accelerator devices;

[0060] Based on the data to be processed, multiple target accelerator devices are determined from among the multiple candidate accelerator devices.

[0061] For example, the load of each accelerator device is calculated based on the operating conditions; it is determined whether the load is less than a preset load threshold, and the accelerator devices that are less than the preset load threshold are identified as candidate accelerator devices.

[0062] In some implementations, determining a plurality of target accelerator devices from the plurality of candidate accelerator devices based on the data to be processed includes:

[0063] Based on the data to be processed, calculate the acceleration amount;

[0064] Based on the acceleration amount, a plurality of target accelerator devices are determined from the plurality of candidate accelerator devices.

[0065] For example, based on the amount of acceleration, an accelerator device capable of processing the data to be processed is determined from among the multiple candidate accelerator devices, that is, multiple target accelerator devices are determined.

[0066] Step S14: Divide the data to be processed according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed.

[0067] For example, the acceleration amount corresponding to each target accelerator device is determined based on the operating status of each target accelerator device, and the data to be processed is segmented according to the acceleration amount to obtain multiple sub-data to be processed.

[0068] For example, such as Figure 2 As shown, the data to be processed is divided into five sub-data to be processed based on the operating status of the multiple target accelerator devices, namely task1, task2, task3, task4 and task5.

[0069] In some implementations, segmenting the data to be processed based on the operating status of the plurality of target accelerator devices to obtain processed data includes:

[0070] Obtain the preset load threshold for each target accelerator device;

[0071] The target acceleration amount is calculated based on the operating status of each target accelerator device and the corresponding preset load threshold.

[0072] The data to be processed is divided into multiple sub-data based on the target acceleration amount.

[0073] For example, a load threshold can be preset for each target accelerator device, under which the target accelerator device can accelerate normally. The target acceleration amount for each target accelerator device is calculated based on its operating status and the corresponding preset load threshold. The data to be processed is then segmented according to the magnitude of the multiple target acceleration amounts to obtain multiple sub-data sets.

[0074] In some implementations, the step of segmenting the data to be processed according to the target acceleration amount to obtain multiple sub-data to be processed includes:

[0075] Obtain the data tags of the data to be processed;

[0076] The data to be processed is segmented according to the target acceleration amount and the data label to obtain multiple sub-data to be processed.

[0077] For example, the data to be processed may include multiple data tags, and different data tags can be used to identify different data content. The data to be processed is segmented according to the target acceleration amount and the data tags to obtain multiple sub-data sets, such that data with the same data tag is contained within one sub-data set.

[0078] Step S15: Send each sub-data to be processed to the corresponding target accelerator device for processing.

[0079] For example, after determining the target accelerator device corresponding to the sub-data to be processed based on the acceleration amount of the sub-data to be processed, the sub-data to be processed is sent to the corresponding target accelerator device for processing.

[0080] In some embodiments, after sending each sub-data to be processed to the corresponding target accelerator device for processing, the method further includes:

[0081] Obtain the processing results of the multiple target accelerator devices on the multiple sub-data to be processed;

[0082] A processing report for the data to be processed is generated based on the processing results.

[0083] For example, the processing results of each target accelerator device on the corresponding sub-data to be processed are obtained, resulting in multiple processing results. A processing report of the data to be processed is generated based on the multiple processing results according to preset rules.

[0084] In some embodiments, the method further includes:

[0085] When multiple target accelerator devices cannot be determined among the multiple accelerator devices based on the data to be processed and the operating conditions, the data to be processed is processed in the engine device.

[0086] When multiple target accelerator devices cannot be identified among the multiple accelerator devices, the data to be processed is processed directly in the engine device, thereby ensuring that the data to be processed is accelerated normally and improving acceleration efficiency. For example, when multiple target accelerator devices cannot be identified among the multiple accelerator devices, a prompt message can be generated according to preset rules.

[0087] The data control method provided in the above embodiments acquires the data to be processed and the operating status of multiple accelerator devices, and determines multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status; then, the data to be processed is segmented according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed; finally, each sub-data to be processed is sent to the corresponding target accelerator device for processing. By segmenting the data to be processed according to the operating status of multiple target accelerator devices, it is determined that each target accelerator device can complete the processing of the corresponding sub-data to be processed, thereby improving the acceleration efficiency.

[0088] Please see Figure 3 , Figure 3 This is a schematic flowchart illustrating a data processing method provided in an embodiment of this application. The data processing method specifically includes steps S31 to S33. Depending on different requirements, the order of the steps in this schematic flowchart can be changed, and some steps can be omitted. This data processing method is applied to accelerator devices, including AI accelerator devices.

[0089] Step S31: In response to the data processing command sent by the engine device, determine whether the calculation requirements corresponding to the data processing command are met.

[0090] For example, in response to a data processing instruction sent by the engine device, data corresponding to the data processing instruction is obtained, and it is determined whether the calculation requirements corresponding to the data processing instruction are met based on the data. For instance, it is determined whether the calculation requirements corresponding to the data processing instruction are met based on the data and the load status of the accelerator device.

[0091] Step S32: When the calculation requirements are not met, report the operating status of the accelerator device to the engine device.

[0092] For example, when the plan requirements are not met, the operating status of the accelerator device is reported to the engine device so that the engine device can re-determine the accelerator device to execute the data.

[0093] Step S33: When the calculation requirements are met, the data corresponding to the data processing instruction is accelerated, and the acceleration result is sent to the engine device.

[0094] The engine device is used to execute the data control method described above.

[0095] When the computational requirements are met, the data corresponding to the data processing instruction is accelerated, and the acceleration result is sent to the engine device.

[0096] The data processing method provided in the above embodiments determines whether the calculation requirements corresponding to the data processing instructions are met in response to data processing instructions sent by the engine device; when the calculation requirements are not met, the operating status of the accelerator device is reported to the engine device; when the calculation requirements are met, the data corresponding to the data processing instructions is accelerated, and the acceleration result is sent to the engine device, ensuring that the data corresponding to the data processing instructions can be accelerated normally and improving acceleration efficiency.

[0097] Please see Figure 4 , Figure 4 This is a schematic block diagram of an electronic device provided in an embodiment of this application. The electronic device 40 can be an engine device or an accelerator device.

[0098] The engine device can be an inference engine. The accelerator device can be an AI accelerator or other electronic device.

[0099] The networks in which electronic device 40 is located include, but are not limited to, the Internet, wide area network, metropolitan area network, local area network, and virtual private network (VPN).

[0100] like Figure 4 As shown, the electronic device 40 includes a communication interface 401, a memory 402, a processor 403, an input / output (I / O) interface 404, and a bus 405. The processor 403 is coupled to the communication interface 401, the memory 402, and the I / O interface 404 via the bus 405.

[0101] Communication interface 401 is used for communication. Communication interface 401 can be an existing interface on electronic device 40 or a newly created interface on electronic device 40. Communication interface 401 can be a network interface, such as a Wireless Local Area Network (WLAN) interface, a cellular network communication interface, or a combination thereof.

[0102] The memory 402 can be used to store operating systems and computer programs. For example, the memory 402 stores the program corresponding to the data control method described above, or the memory 402 stores the program corresponding to the data processing method described above.

[0103] It should be understood that the memory 402 may include a program storage area and a data storage area. The program storage area may be used to store the operating system, application programs required by at least one method (such as data control methods, data processing methods, etc.), etc.; the data storage area may store data created based on the use of the electronic device 40 (such as sub-data to be processed, etc.). Furthermore, the memory 402 may include volatile memory, and may also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other storage devices.

[0104] Processor 403 provides computing and control capabilities to support the operation of the entire computer device. For example, processor 403 is used to execute computer programs stored in memory 402 to implement the steps in the data control method or the data processing method described above.

[0105] It should be understood that processor 403 is a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0106] I / O interface 404 is used to provide a channel for user input or output. For example, I / O interface 404 can be used to connect various input and output devices (mouse, keyboard or 3D touch device, etc.) and displays, so that users can enter information or visualize information.

[0107] Bus 405 is used at least to provide a channel for communication between communication interface 401, memory 402, processor 403 and I / O interface 404 in electronic device 40.

[0108] Those skilled in the art will understand that Figure 4The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0109] In one embodiment, when the processor 403 executes a computer program stored in the memory 402 to implement the data control method, it performs the following steps:

[0110] Obtain the data to be processed;

[0111] Receive operational status reports from multiple accelerator devices;

[0112] Based on the data to be processed and the operating status, multiple target accelerator devices are determined from among the multiple accelerator devices;

[0113] The data to be processed is segmented according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed.

[0114] Each piece of data to be processed is sent to the corresponding target accelerator device for processing.

[0115] In some implementations, when processor 403 performs the step of determining multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status, it is specifically used to implement:

[0116] Calculate the load on each accelerator device based on the described operating conditions;

[0117] Based on the load, a plurality of candidate accelerator devices are determined from the plurality of accelerator devices;

[0118] Based on the data to be processed, multiple target accelerator devices are determined from among the multiple candidate accelerator devices.

[0119] In some implementations, when processor 403 performs the step of determining multiple target accelerator devices from the multiple candidate accelerator devices based on the data to be processed, it is specifically used to implement:

[0120] Based on the data to be processed, calculate the acceleration amount;

[0121] Based on the acceleration amount, a plurality of target accelerator devices are determined from the plurality of candidate accelerator devices.

[0122] In some implementations, when processor 403 performs the step of segmenting the data to be processed based on the operating status of the plurality of target accelerator devices to obtain processed data, it specifically implements the following:

[0123] Obtain the preset load threshold for each target accelerator device;

[0124] The target acceleration amount is calculated based on the operating status of each target accelerator device and the corresponding preset load threshold.

[0125] The data to be processed is divided into multiple sub-data based on the target acceleration amount.

[0126] In some implementations, when processor 403 performs the step of segmenting the data to be processed according to the target acceleration amount to obtain multiple sub-data to be processed, it is specifically used to implement:

[0127] Obtain the data tags of the data to be processed;

[0128] The data to be processed is segmented according to the target acceleration amount and the data label to obtain multiple sub-data to be processed.

[0129] In some implementations, after the processor 403 executes the method of sending each sub-data to be processed to the corresponding target accelerator device for processing, the method is specifically used to implement:

[0130] Obtain the processing results of the multiple target accelerator devices on the multiple sub-data to be processed;

[0131] A processing report for the data to be processed is generated based on the processing results.

[0132] In some implementations, when the processor 403 executes a computer program stored in memory to implement the data control method, it is also used to implement:

[0133] When multiple target accelerator devices cannot be determined among the multiple accelerator devices based on the data to be processed and the operating conditions, the data to be processed is processed in the engine device.

[0134] Specifically, the processor 403's specific implementation method for the above instructions can be found in the description of the relevant steps in the aforementioned data control method embodiment, and will not be repeated here.

[0135] In one embodiment, when the processor 403 executes a computer program stored in the memory 402 to implement a data processing method, it performs the following steps:

[0136] In response to the data processing command sent by the engine device, determine whether the calculation requirements corresponding to the data processing command are met;

[0137] When the computational requirements are not met, the operating status of the accelerator device is reported to the engine device.

[0138] When the computational requirements are met, the data corresponding to the data processing instruction is accelerated, and the acceleration result is sent to the engine device.

[0139] The engine device is used to execute the data control method described above.

[0140] Specifically, the processor 403's specific implementation method for the above instructions can be found in the description of the relevant steps in the aforementioned data processing method embodiment, and will not be repeated here.

[0141] This application also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, and the method implemented when the program instructions are executed can refer to various embodiments of the data control method or the data processing method of this application.

[0142] The computer-readable storage medium can be an internal storage unit of the electronic device described in the foregoing embodiments, such as a hard disk or memory of the electronic device. Alternatively, the computer-readable storage medium can be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device.

[0143] Furthermore, the computer-readable storage medium may primarily include a program storage area and a data storage area, wherein the program storage area may store the operating system, at least one application program required for a function, etc.; and the data storage area may store data created based on the use of the electronic device, etc.

[0144] The electronic device and computer-readable storage medium provided in the foregoing embodiments can acquire data to be processed and the operating status of multiple accelerator devices, and determine multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status; then, the data to be processed is segmented according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed; finally, each sub-data to be processed is sent to the corresponding target accelerator device for processing. By segmenting the data to be processed according to the operating status of multiple target accelerator devices, it is determined that each target accelerator device can process the corresponding sub-data to be processed, thereby improving the acceleration efficiency.

[0145] The electronic device and computer-readable storage medium provided in the foregoing embodiments can also determine whether the computational requirements corresponding to the data processing instructions are met in response to the data processing instructions sent by the engine device; when the computational requirements are not met, the operating status of the accelerator device is reported to the engine device; when the computational requirements are met, the data corresponding to the data processing instructions is accelerated, and the acceleration result is sent to the engine device, ensuring that the data corresponding to the data processing instructions can be accelerated normally and improving acceleration efficiency.

[0146] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0147] It should also be understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. It should be noted that, herein, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0148] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above descriptions are merely specific implementations of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A data control method, operating in an engine device, characterized in that, The data control method includes: Obtain the data to be processed; Receive operational status reports from multiple accelerator devices; Based on the data to be processed and the operating status, multiple target accelerator devices are determined from among the multiple accelerator devices; The data to be processed is segmented according to the operating status of the multiple target accelerator devices to obtain multiple sub-data to be processed, including: obtaining a preset load threshold for each target accelerator device; calculating a target acceleration amount based on the operating status of each target accelerator device and the corresponding preset load threshold; and segmenting the data to be processed according to the target acceleration amount to obtain the multiple sub-data to be processed, wherein the target accelerator device accelerates normally at the preset load threshold. The step of segmenting the data to be processed according to the target acceleration amount to obtain multiple sub-data to be processed includes: obtaining the data label of the data to be processed, and segmenting the data to be processed according to the target acceleration amount and the data label to obtain multiple sub-data to be processed; Each piece of data to be processed is sent to the corresponding target accelerator device for processing.

2. The data control method according to claim 1, characterized in that, The step of determining multiple target accelerator devices among the multiple accelerator devices based on the data to be processed and the operating status includes: Calculate the load on each accelerator device based on the described operating conditions; Based on the load, a plurality of candidate accelerator devices are determined from the plurality of accelerator devices; Based on the data to be processed, multiple target accelerator devices are determined from the multiple candidate accelerator devices.

3. The data control method according to claim 2, characterized in that, The step of determining multiple target accelerator devices from the multiple candidate accelerator devices based on the data to be processed includes: Calculate the acceleration amount based on the data to be processed; Based on the acceleration amount, a plurality of target accelerator devices are determined from the plurality of candidate accelerator devices.

4. The data control method according to claim 1, characterized in that, The method further includes: When multiple target accelerator devices cannot be determined among the multiple accelerator devices based on the data to be processed and the operating conditions, the data to be processed is processed in the engine device.

5. The data control method according to any one of claims 1 to 4, characterized in that, After sending each sub-data to be processed to the corresponding target accelerator device for processing, the method further includes: Obtain the processing result of each target accelerator device on the corresponding sub-data to be processed; A processing report of the data to be processed is generated based on multiple processing results.

6. A data processing method, operating in an accelerator device, characterized in that, The data control method includes: In response to the data processing command sent by the engine device, determine whether the calculation requirements corresponding to the data processing command are met; When it is determined that the calculation requirements are not met, the operating status of the accelerator device is reported to the engine device. When the computational requirements are met, the data corresponding to the data processing instruction is accelerated, and the acceleration result is sent to the engine device, so that the engine device executes the data control method as described in any one of claims 1 to 5.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor; The memory is used to store at least one instruction; The processor is used to implement the data control method as described in any one of claims 1 to 5 or the data processing method as described in claim 6 when executing the at least one instruction.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the data control method as described in any one of claims 1 to 5 or the data processing method as described in claim 6.

Images