Data processing method, device, storage medium, electronic device and system

Abstract

The present disclosure relates to a method, device, storage medium, electronic device and system for data processing; by determining the data type of operation data of a preset network model; determining a target processing device corresponding to the data type from multiple processing devices; the multiple processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the operation data is processed by the target processing device; through the above technical solution, data processing can be performed together by processing devices with different performance parameters, thereby effectively reducing the operation cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operation programs for different processing devices, which greatly reduces the development cost and improves work efficiency.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular, to a method, device, storage medium, electronic device and system for data processing. Background Art

[0002] At present, computer big models have good development prospects in all walks of life. Big models refer to deep learning models with tens of millions or even hundreds of millions of parameters. Since the amount of data can reach GB or TB level, there are huge challenges in data operation and processing.

[0003] In related technologies, large model data processing can be performed by GPU (Graphic Process Unit) and CPU (Central Process Unit) to reduce operating costs. However, during data processing, GPU and CPU need to run different types of computer programs, resulting in poor flexibility in the data processing process, increased development costs, and reduced work efficiency. Summary of the invention

[0004] The purpose of the present disclosure is to provide a data processing method, device, storage medium, electronic device and system.

[0005] In order to achieve the above objectives, in a first aspect, the present disclosure provides a data processing method, the method comprising:

[0006] Determine the data type of the operation data of the preset network model;

[0007] Determine a target processing device corresponding to the data type from a plurality of processing devices; the plurality of processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; a performance parameter of the first processing device is greater than a performance parameter of the second processing device;

[0008] The operating data is processed by the target processing device.

[0009] Optionally, the determining of the data type of the operation data of the preset network model includes:

[0010] The data type of the operating data is determined by a preset data processing algorithm.

[0011] Optionally, determining, according to the data type, a target processing device corresponding to the data type from a plurality of processing devices comprises:

[0012] determining a candidate processing device from a plurality of processing devices according to the data type;

[0013] The target processing device is determined from the candidate processing devices.

[0014] Optionally, the data type includes hotspot data or non-hotspot data; and determining a candidate processing device from a plurality of processing devices according to the data type includes:

[0015] In the case where it is determined that the data type is the hot data, the first processing device among the multiple processing devices is used as the candidate processing device; or,

[0016] When it is determined that the data type is the non-hotspot data, the second processing device among the multiple processing devices is used as the candidate processing device.

[0017] Optionally, determining the target processing device from the candidate processing devices includes:

[0018] Determining the storage capacity of each of the candidate processing devices according to a preset processing order;

[0019] The candidate processing device whose storage capacity is greater than or equal to a preset capacity threshold is used as the target processing device.

[0020] In a second aspect, the present disclosure provides a data processing device, the device comprising:

[0021] A first determination module, used to determine the data type of the operation data of the preset network model;

[0022] a second determining module, configured to determine a target processing device corresponding to the data type from a plurality of processing devices; the plurality of processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; a performance parameter of the first processing device is greater than a performance parameter of the second processing device;

[0023] A processing module is used to process the operation data through the target processing device.

[0024] Optionally, the first determination module is used to determine the data type of the operating data by using a preset data processing algorithm.

[0025] Optionally, the second determining module is used to determine a candidate processing device from a plurality of processing devices according to the data type; and determine the target processing device from the candidate processing devices.

[0026] Optionally, the data type includes hotspot data or non-hotspot data; the second determination module is used to, when it is determined that the data type is the hotspot data, use the first processing device among the multiple processing devices as the candidate processing device; or, when it is determined that the data type is the non-hotspot data, use the second processing device among the multiple processing devices as the candidate processing device.

[0027] Optionally, the second determination module is used to determine the storage capacity of each of the candidate processing devices according to a preset processing order; and to use the candidate processing device whose storage capacity is greater than or equal to a preset capacity threshold as the target processing device.

[0028] In a third aspect, the present disclosure provides a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data processing method described in the first aspect above.

[0029] In a fourth aspect, the present disclosure provides an electronic device, including:

[0030] a memory having a computer program stored thereon;

[0031] A processor is used to execute the computer program in the memory to implement the data processing method described in the first aspect above.

[0032] In a fifth aspect, the present disclosure provides a data processing system, the system comprising a controller and a plurality of processing devices; the controller is connected to the plurality of processing devices respectively; the plurality of processing devices comprises a first processing device and a second processing device; the first processing device and the second processing device are the same device; a performance parameter of the first processing device is greater than a performance parameter of the second processing device;

[0033] The controller is used to execute the data processing method described in the first aspect above.

[0034] Optionally, the second processing device includes a video memory module and a memory module; the controller is connected to the video memory module and the memory module respectively; the video memory module is connected to the memory module; the operation data includes processing data and data to be processed; the data volume of the processing data is the same as the storage capacity of the video memory module; the data to be processed is the data in the operation data except the processing data;

[0035] The display memory module is used to receive the processing data sent by the controller and process the processing data;

[0036] The memory module is used to receive the data to be processed sent by the controller; and store the data to be processed.

[0037] Through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces development costs and improves work efficiency.

[0038] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings are used to provide further understanding of the present disclosure and constitute a part of the specification. They are used to explain the present disclosure together with the following detailed description, but do not constitute a limitation of the present disclosure.

[0040] Figure 1 The present invention is a flowchart of a data processing method according to an exemplary embodiment.

[0041] Figure 2 is based on Figure 1 An exemplary embodiment is shown as a flow chart of a method of data processing.

[0042] Figure 3 The present invention is a flowchart of another data processing method according to an exemplary embodiment.

[0043] Figure 4 The invention is a block diagram of a data processing device according to an exemplary embodiment.

[0044] Figure 5 The invention is a block diagram of a data processing system according to an exemplary embodiment.

[0045] Figure 6 is based on Figure 5 A block diagram of a data processing system is shown in accordance with an exemplary embodiment.

[0046] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment.

[0047] Figure 8 is a block diagram of another electronic device according to an exemplary embodiment.

[0048] Fig. 9 It is a block diagram of another data processing system according to an exemplary embodiment. DETAILED DESCRIPTION

[0049] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0050] First, the application scenario of the present disclosure is introduced. The present disclosure is applied in the scenario of running a large model through a processor. A large model refers to a deep learning model with tens of millions or even hundreds of millions of parameters. Since its data volume can reach GB or TB level, there are huge challenges in data operation and processing.

[0051] In the related art, GPU can be used to process data of large models, but the cost of GPU is relatively high; in order to reduce the cost, GPU and CPU can also be used to process data of large models together. However, during the data processing process, GPU and CPU need to run different types of computer programs, which leads to poor flexibility in the data processing process, increases development costs, and reduces work efficiency.

[0052] In order to solve the above problems, the present disclosure provides a method, device, storage medium, electronic device and system for data processing; by determining the data type of operation data of a preset network model; determining a target processing device corresponding to the data type from multiple processing devices; the multiple processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the operation data is processed by the target processing device; through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operation cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operation programs for different processing devices, which greatly reduces the development cost and improves work efficiency.

[0053] Figure 1 FIG. 1 is a flow chart of a method for data processing according to an exemplary embodiment. Figure 1 As shown, the method can be applied to a controller, and the method can include the following steps.

[0054] S101. Determine the data type of the operation data of the preset network model.

[0055] For example, the preset network model may be a large language model, a computer vision model, an audio model, or a large multimodal model, etc., which is not limited here; the data type may include hotspot data or non-hotspot data.

[0056] In some embodiments, the above step S101 may include: determining the data type of the operating data through a preset data processing algorithm. For example, the preset data processing algorithm may be an algorithm that accelerates the reasoning speed of a preset network model, such as a PowerInfer algorithm. For example, the operating data may be divided into hotspot data and non-hotspot data through a preset data processing algorithm; the hotspot data refers to data that is used more frequently in the preset network model, and the non-hotspot data refers to data that is used less frequently in the preset network model.

[0057] It should be noted that since the preset data processing algorithm can be combined with the preset network model, the data type can also be hotspot neurons and non-hotspot neurons; the hotspot neurons refer to the data with higher frequency of use in the preset network model and the operations corresponding to the data with higher frequency of use; the non-hotspot neurons refer to the data with lower frequency of use in the preset network model and the operations corresponding to the data with lower frequency of use.

[0058] S102: Determine a target processing device corresponding to the data type from a plurality of processing devices.

[0059] The plurality of processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; and a performance parameter of the first processing device is greater than a performance parameter of the second processing device.

[0060] For example, the processing device may be a processor, such as a GPU; the first processing device may be a high-performance GPU, the second processing device may be a low-performance GPU, and the working performance of the first processing device is higher than the working performance of the second processing device. The performance parameters may include operating speed, data computing volume, or power consumption, etc., which are not limited here. For example, the operating speed of the first processing device may be higher than the operating speed of the second processing device; the data computing volume of the first processing device may be higher than the data computing volume of the second processing device; the power consumption of the first processing device may be lower than the power consumption of the second processing device, etc.

[0061] S103: Process the operation data through the target processing device.

[0062] Through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces development costs and improves work efficiency.

[0063] Figure 2 is based on Figure 1 A flowchart of a method for data processing is shown in an exemplary embodiment. Figure 2 As shown, the above step S102 may include the following steps.

[0064] S1021. Determine a candidate processing device from a plurality of processing devices according to the data type.

[0065] In some embodiments, the data type includes hotspot data or non-hotspot data; the above S1021 may include: when it is determined that the data type is the hotspot data, the first processing device among the multiple processing devices is used as the candidate processing device. Alternatively, when it is determined that the data type is the non-hotspot data, the second processing device among the multiple processing devices is used as the candidate processing device.

[0066] For example, each processing device includes an identifier, and the identifier can be used to determine whether the processing device is the first processing device or the second processing device. In this way, different processing devices can be determined to perform data processing based on the type of operating data, so as to improve the efficiency of data processing.

[0067] S1022: Determine the target processing device from the candidate processing devices.

[0068] In some embodiments, the above S1022 may include: determining the storage capacity of each candidate processing device according to a preset processing order; and using the candidate processing device whose storage capacity is greater than or equal to a preset capacity threshold as the target processing device.

[0069] For example, the preset processing order can be set by the user according to the actual operation situation; the preset capacity threshold can be set by the user according to the actual operation situation, which is not limited here. For example, the first storage capacity of the first candidate processing device can be determined, and when the first storage capacity is greater than or equal to the preset capacity threshold, the first candidate processing device is used as the candidate processing device; when the storage capacity of the first candidate processing device is less than the preset capacity threshold, the second storage capacity of the second candidate processing device is determined, and when the second storage capacity is greater than or equal to the preset capacity threshold, the second processing device is used as the candidate processing device; when the second storage capacity is less than the preset capacity threshold, the above steps of determining the storage capacity of the processing device are repeated until a candidate processing device is determined.

[0070] The first candidate processing device may be a processing device in any order in the preset processing sequence, and the second candidate processing device may be a next processing device in the preset processing sequence that is immediately adjacent to the first candidate processing device. In this way, the operation data can be sequentially stored in the processing devices for processing in the preset sequence, so that each processing device can process data at full capacity, avoiding omissions and improving work efficiency.

[0071] Figure 3 FIG. 1 is a flow chart of a method for data processing according to an exemplary embodiment. Figure 3 As shown, the method may include the following steps.

[0072] S301. Determine the data type of the operating data by using a preset data processing algorithm.

[0073] The data type may include hotspot data or non-hotspot data.

[0074] S302: Determine a candidate processing device from a plurality of processing devices according to the data type.

[0075] S303: Determine the storage capacity of each candidate processing device according to a preset processing order.

[0076] S304: The candidate processing device whose storage capacity is greater than or equal to a preset capacity threshold is used as the target processing device.

[0077] S305: Process the operation data through the target processing device.

[0078] Through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces development costs and improves work efficiency.

[0079] Figure 4 FIG. 1 is a block diagram of a data processing device according to an exemplary embodiment. Figure 4 As shown, the apparatus 400 may include a first determination module 410, a second determination module 420 and a processing module 430;

[0080] The first determination module 410 is used to determine the data type of the operation data of the preset network model;

[0081] The second determination module 420 is used to determine a target processing device corresponding to the data type from a plurality of processing devices; the plurality of processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device;

[0082] The processing module 430 is used to process the operation data through the target processing device.

[0083] Through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces development costs and improves work efficiency.

[0084] Optionally, the first determination module 410 is used to determine the data type of the operating data by using a preset data processing algorithm.

[0085] Optionally, the second determination module 420 is configured to determine a candidate processing device from a plurality of processing devices according to the data type; and determine the target processing device from the candidate processing devices.

[0086] Optionally, the data type includes hotspot data or non-hotspot data; the second determination module 420 is used to, when it is determined that the data type is the hotspot data, use the first processing device among the multiple processing devices as the candidate processing device; or, when it is determined that the data type is the non-hotspot data, use the second processing device among the multiple processing devices as the candidate processing device.

[0087] Optionally, the second determination module 420 is used to determine the storage capacity of each candidate processing device according to a preset processing order; and to use the candidate processing device whose storage capacity is greater than or equal to a preset capacity threshold as the target processing device.

[0088] 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.

[0089] Figure 5 FIG. 1 is a block diagram of a data processing system according to an exemplary embodiment. Figure 5 As shown, the system 500 includes a controller 510 and a plurality of processing devices; the controller 510 is connected to the plurality of processing devices respectively; the plurality of processing devices include a first processing device 520 and a second processing device 530; the first processing device 520 and the second processing device 530 are the same device; the performance parameter of the first processing device 520 is greater than the performance parameter of the second processing device 530;

[0090] The controller 510 is used to execute the data processing method of the first aspect mentioned above.

[0091] It should be noted that the connection sequence between the controller 510 and the first processing device 520 and the second processing device 530 can be multiple, and the number of the first processing device 520 and the second processing device 530 can be multiple, which is not limited here. Figure 5The connection sequence shown is only one of many possible connection sequences. Other sequences and quantities are not shown. Figure 5 Shown in.

[0092] Through the above technical solution, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces development costs and improves work efficiency.

[0093] Figure 6 is based on Figure 5 FIG. 1 is a block diagram of a data processing system shown in FIG. Figure 6 As shown, the second processing device 530 includes a video memory module 531 and a memory module 532; the controller 510 is connected to the video memory module 531 and the memory module 532 respectively; the video memory module 531 is connected to the memory module 532; the operation data includes processing data and data to be processed; the data volume of the processing data is the same as the storage capacity of the video memory module 531; the data to be processed is the data in the operation data except the processing data;

[0094] The display memory module 531 is used to receive the processing data sent by the controller 510 and process the processing data;

[0095] The memory module 532 is used to receive the data to be processed sent by the controller 510 and store the data to be processed.

[0096] For example, the graphics memory module may be a GPU, which is used to process data; the memory module may be a memory of a server, which is used to store data, thereby increasing the storage capacity of the processing device.

[0097] In some embodiments, the controller 510 can be used to determine the processed data and the data to be processed according to the storage capacity of the video memory module 531 ; and send the processed data to the video memory module 531 , and send the data to be processed to the memory module 532 .

[0098] In other embodiments, the controller 510 may be used to use the operating data of the same amount as the storage capacity as the processing data according to a preset data sequence, and use the data in the operating data other than the processing data as the data to be processed. The preset data sequence may be the sequence in which the controller obtains the operating data, or may be set by the user, which is not limited here. In this way, the video memory module may be operated in a full-load state to realize data processing and improve the processing efficiency of the video memory module.

[0099] In some embodiments, the controller 510 can be used to obtain the data to be processed stored in the memory module 532 through the video memory module 531 when it is determined that the first storage capacity of the video memory module 531 is greater than or equal to the first preset capacity threshold; the video memory module 531 is used to process the data to be processed. In this way, after the video memory module completes processing of part of the data, the data to be processed in the memory module can be stored in the video memory module in time, and data processing can be continued, thereby avoiding interruption of the processing process and improving processing efficiency.

[0100] In other embodiments, the controller 510 can be used to determine the remaining storage capacity of the video memory module 531 when it is determined that the first storage capacity of the video memory module 531 is greater than or equal to a first preset capacity threshold, and obtain the first to-be-processed data stored in the memory module 532 through the video memory module 531; the data volume of the first to-be-processed data is the same as the remaining storage capacity.

[0101] For example, the first preset capacity threshold may be the maximum storage capacity of the video memory module, which may be set by the user and is not limited here. The remaining storage capacity may be determined based on the first preset capacity threshold and the first storage capacity. For example, the difference between the first preset capacity threshold and the first storage capacity may be used as the remaining storage capacity. In this way, data to be processed having the same amount of data as the remaining storage capacity of the video memory module may be obtained, thereby preventing the amount of data from exceeding the storage capacity of the video memory module.

[0102] In some embodiments, the controller 510 can control the video memory module 531 to obtain the data to be processed stored in the memory module 532 through a preset processing algorithm. For example, the preset control algorithm can be a Managed Memory algorithm, which has been disclosed in relevant technical documents and will not be described in detail here. In this way, the mobilization between the processed data and the data to be processed can be achieved, thereby improving work efficiency.

[0103] In summary, the present disclosure provides a method, device, storage medium, electronic device and system for data processing; by determining the data type of the operating data of a preset network model; determining a target processing device corresponding to the data type from multiple processing devices; the multiple processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the operating data is processed by the target processing device; through the above technical scheme, data processing can be performed jointly by processing devices with different performance parameters, thereby effectively reducing the operating cost of data processing and improving the flexibility of data processing; and, there is no need to develop different operating programs for different processing devices, which greatly reduces the development cost and improves work efficiency.

[0104] Figure 7 FIG. 7 is a block diagram of an electronic device 700 according to an exemplary embodiment. Figure 7 As shown, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output interface 704, and a communication component 705.

[0105] The processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above data processing method. The memory 702 is used to store various types of data to support the operation of the electronic device 700. For example, these data may include instructions for any application or method used to operate on the electronic device 700, and application-related data, such as contact data, messages sent and received, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 702 or sent through the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The input / output interface 704 provides an interface between the processor 701 and other interface modules, and the other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 705 may include: Wi-Fi module, Bluetooth module, NFC module, etc.

[0106] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned data processing method.

[0107] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned data processing method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 702 including program instructions, and the above-mentioned program instructions can be executed by the processor 701 of the electronic device 700 to complete the above-mentioned data processing method.

[0108] Figure 8 8 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be provided as a server. Figure 8 The electronic device 800 includes a processor 822, which may be one or more, and a memory 832 for storing a computer program executable by the processor 822. The computer program stored in the memory 832 may include one or more modules, each corresponding to a set of instructions. In addition, the processor 822 may be configured to execute the computer program to perform the above-mentioned data processing method.

[0109] In addition, the electronic device 800 may further include a power supply component 826 and a communication component 850, wherein the power supply component 826 may be configured to perform power management of the electronic device 800, and the communication component 850 may be configured to implement communication, for example, wired or wireless communication, of the electronic device 800. In addition, the electronic device 800 may further include an input / output interface 858. The electronic device 800 may operate based on an operating system stored in the memory 832.

[0110] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by the processor, the steps of the above-mentioned data processing method are implemented. For example, the non-transitory computer-readable storage medium can be the above-mentioned memory 832 including program instructions, and the above-mentioned program instructions can be executed by the processor 822 of the electronic device 800 to complete the above-mentioned data processing method.

[0111] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above data processing method when executed by the programmable device.

[0112] Fig. 9 FIG. 1 is a block diagram of another data processing system according to an exemplary embodiment. Fig. 9 As shown, the system 900 may include an electronic device 700 .

[0113] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0114] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0115] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for data processing, characterized in that: The method is applied to a data processing system, the system comprising a controller and a plurality of processing devices; the controller is connected to the plurality of processing devices respectively; the plurality of processing devices comprises a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the second processing device comprises a video memory module and a memory module; The controller is connected to the video memory module and the memory module respectively; The video memory module is connected to the memory module; the method comprises: Determine, by the controller, a data type of operation data of a preset network model; determine a target processing device corresponding to the data type from a plurality of processing devices; the plurality of processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; a performance parameter of the first processing device is greater than a performance parameter of the second processing device; The determining, by the controller according to the data type, from a plurality of processing devices, a target processing device corresponding to the data type comprises: Determining, by the controller, a candidate processing device from a plurality of processing devices according to the data type; and determining the target processing device from the candidate processing devices; The data type includes hotspot data or non-hotspot data; and determining, by the controller according to the data type, a candidate processing device from a plurality of processing devices includes: By means of the controller, when it is determined that the data type is the hot data, the first processing device among the multiple processing devices is used as the candidate processing device; or, when it is determined that the data type is the non-hot data, the second processing device among the multiple processing devices is used as the candidate processing device; Processing the operation data by the target processing device; The operating data includes processed data and data to be processed; the amount of the processed data is the same as the storage capacity of the video memory module; the data to be processed is the data in the operating data except the processed data; In the case where the target processing device is the second processing device, the processing data sent by the controller is received through the display memory module; and the processing data is processed; the data to be processed sent by the controller is received through the memory module; and the data to be processed is stored; The method further comprises: When the target processing device is the second processing device and it is determined that the first storage capacity of the video memory module is greater than or equal to a first preset capacity threshold, the data to be processed stored in the memory module is obtained through the video memory module; and the data to be processed is processed through the video memory module.

2. The method according to claim 1, characterized in that: The data type of the operation data of the preset network model determined by the controller includes: The data type of the operating data is determined by the controller based on a preset data processing algorithm.

3. The method according to claim 1, characterized in that The determining, by the controller, the target processing device from the candidate processing devices comprises: The storage capacity of each of the candidate processing devices is determined by the controller according to a preset processing order; and the candidate processing devices whose storage capacity is greater than or equal to a preset capacity threshold are used as the target processing devices.

4. A data processing device, characterized in that: The device is applied to a data processing system, the system comprising a controller and a plurality of processing devices; the controller is connected to the plurality of processing devices respectively; the plurality of processing devices comprises a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the second processing device comprises a video memory module and a memory module; The controller is connected to the video memory module and the memory module respectively; The video memory module is connected to the memory module; The data processing device comprises: A first determination module is used to determine the data type of the operation data of the preset network model through the controller; determine the target processing device corresponding to the data type from multiple processing devices; the multiple processing devices include a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the controller determines the target processing device corresponding to the data type from multiple processing devices according to the data type, including: determining a candidate processing device from multiple processing devices according to the data type through the controller; determining the target processing device from the candidate processing devices; the data type includes hotspot data or non-hotspot data; the controller determines the candidate processing device from multiple processing devices according to the data type, including: through the controller, when determining that the data type is the hotspot data, the first processing device among the multiple processing devices is used as the candidate processing device; or, when determining that the data type is the non-hotspot data, the second processing device among the multiple processing devices is used as the candidate processing device; a processing module, configured to process the operating data through the target processing device; the operating data includes processing data and data to be processed; the amount of the processing data is the same as the storage capacity of the video memory module; the data to be processed is data in the operating data other than the processing data; when the target processing device is the second processing device, the processing data sent by the controller is received through the video memory module; and the processing data is processed; the data to be processed sent by the controller is received through the memory module; and the data to be processed is stored; A target processing module is used to obtain the data to be processed stored in the memory module through the video memory module when the target processing device is the second processing device and it is determined that the first storage capacity of the video memory module is greater than or equal to a first preset capacity threshold; and process the data to be processed through the video memory module.

5. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 3 are implemented.

6. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 3.

7. A data processing system, characterized in that: The system comprises a controller and a plurality of processing devices; the controller is connected to the plurality of processing devices respectively; the plurality of processing devices comprises a first processing device and a second processing device; the first processing device and the second processing device are the same device; the performance parameter of the first processing device is greater than the performance parameter of the second processing device; the second processing device comprises a video memory module and a memory module; the controller is connected to the video memory module and the memory module respectively; the video memory module is connected to the memory module; the operation data comprises processing data and data to be processed; the data volume of the processing data is the same as the storage capacity of the video memory module; the data to be processed is the data in the operation data except the processing data; The controller is used to execute the data processing method according to any one of claims 1 to 3; The display memory module is used to receive the processing data sent by the controller and process the processing data; The memory module is used to receive the data to be processed sent by the controller; and store the data to be processed; The video memory module is further used to obtain the data to be processed stored in the memory module when it is determined that the first storage capacity is greater than or equal to a first preset capacity threshold; and process the data to be processed.

Images