Task processing method and device, computer system, computer equipment and readable storage medium

Abstract

The invention relates to a task processing method and device, a computer system, computer equipment and a readable storage medium, and is applied to a central processor, the method comprises the following steps: receiving a to-be-processed task aiming at an intelligent application service, and obtaining task features of the to-be-processed task under a plurality of preset dimensions; determining a task type of the to-be-processed task based on the task features of each dimension; and distributing the to-be-processed task to a corresponding task processor according to the task type. By adopting the method, the performance and energy efficiency of intelligent application can be improved, the energy consumption is reduced, and the stability and reliability of the whole system are improved. By quickly identifying the task type and reasonably distributing the processors, the method reduces the response time of the task and meets the application requirements of real-time requirements.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a task processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] In industrial computers, various AI applications are becoming more and more widespread, such as machine vision, speech recognition, natural language processing, etc. These application services usually involve many tasks that need to be processed, and these tasks vary significantly in nature, computational complexity, and real-time requirements. Traditional task processing methods often rely on fixed processing strategies, allocate and process tasks in a fixed manner, and fail to effectively utilize the system's hardware resources. Summary of the invention

[0003] Based on this, it is necessary to provide a task processing method, apparatus, computer equipment, computer-readable storage medium and computer program product to address the above technical issues.

[0004] In a first aspect, the present application provides a task processing method. The method comprises:

[0005] Receiving a task to be processed for an intelligent application service, and obtaining task features of the task to be processed in a plurality of preset dimensions;

[0006] Determine the task type of the task to be processed based on the task characteristics of each dimension;

[0007] According to the task type, the to-be-processed task is assigned to a corresponding task processor.

[0008] In one embodiment, the obtaining of task features of the to-be-processed task in multiple preset dimensions includes:

[0009] Obtaining the task content and performance requirements of the task to be processed;

[0010] The task content and the performance requirement are used as a plurality of task features of the task to be processed.

[0011] In one embodiment, determining the task type of the task to be processed based on the task features in each dimension includes:

[0012] When the task content is graphic data and the performance requirement meets a first preset condition, determining that the task type is a first task type requiring high-performance display;

[0013] When the task content is computing data and the performance requirement meets the second preset condition, the task type is determined to be a second task type requiring high computing power.

[0014] In one embodiment, allocating the to-be-processed task to a corresponding task processor according to the task type includes:

[0015] If the task type is the first task type, allocating the task to be processed to a graphics processor;

[0016] If the task type is the second task type, the task to be processed is allocated to a neural network processor.

[0017] In one embodiment, after allocating the to-be-processed tasks to corresponding task processors according to the task types, the method further includes:

[0018] Outputting and displaying the processing result through the display interface of the graphics processor, or receiving the processing result returned by the graphics processor;

[0019] Receive the processing result returned by the neural network processor.

[0020] In a second aspect, the present application also provides a task processing device. The device comprises:

[0021] An acquisition module, used to receive a task to be processed for an intelligent application service, and acquire task features of the task to be processed in a plurality of preset dimensions;

[0022] A determination module, used to determine the task type of the task to be processed based on the task characteristics of each dimension;

[0023] The allocation module is used to allocate the to-be-processed tasks to corresponding task processors according to the task types.

[0024] In a third aspect, the present application also provides a computer system. The computer system includes a central processing unit, a graphics processing unit, and a neural network processor;

[0025] The central processor is used to receive a task to be processed for an intelligent application service, obtain task features of the task to be processed in multiple preset dimensions; determine the task type of the task to be processed based on the task features of each dimension; and assign the task to be processed to the graphics processor or the neural network processor according to the task type;

[0026] The graphics processor and the neural network processor are used to process the tasks to be processed assigned by the central processing unit.

[0027] In a fourth aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0028] Receiving a task to be processed for an intelligent application service, and obtaining task features of the task to be processed in a plurality of preset dimensions;

[0029] Determine the task type of the task to be processed based on the task characteristics of each dimension;

[0030] According to the task type, the to-be-processed task is assigned to a corresponding task processor.

[0031] In a fifth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0032] Receiving a task to be processed for an intelligent application service, and obtaining task features of the task to be processed in a plurality of preset dimensions;

[0033] Determine the task type of the task to be processed based on the task characteristics of each dimension;

[0034] According to the task type, the to-be-processed task is assigned to a corresponding task processor.

[0035] In a sixth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0036] Receiving a task to be processed for an intelligent application service, and obtaining task features of the task to be processed in a plurality of preset dimensions;

[0037] Determine the task type of the task to be processed based on the task characteristics of each dimension;

[0038] According to the task type, the to-be-processed task is assigned to a corresponding task processor.

[0039] The above-mentioned task processing method, device, computer system, computer equipment, computer-readable storage medium and computer program product obtain the task characteristics of the task to be processed in multiple preset dimensions through the central processing unit; determine the task type of the task to be processed based on the task characteristics of each dimension; and assign the task to be processed to the corresponding task processor according to the task type. By optimizing the processing flow of AI application tasks, the performance and energy efficiency of these applications can be significantly improved, energy consumption can be reduced, and the stability and reliability of the overall system can be improved. By quickly identifying the task type and reasonably allocating the processor, the method reduces the response time of the task and meets the application requirements with real-time requirements. For example, an industrial robot can process sensor data in real time through a neural network processor to achieve more accurate control and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] FIG1 is an application environment diagram of a task processing method in one embodiment;

[0041] FIG2 is a schematic diagram of a task processing method in one embodiment;

[0042] FIG3 is a schematic diagram of a flow chart of a task processing method in another embodiment;

[0043] FIG4 is a schematic diagram of the structure of a computer system in one embodiment;

[0044] FIG5 is a schematic diagram of an integrated AI computer system in one embodiment;

[0045] FIG6 is a block diagram of a task processing device in one embodiment;

[0046] FIG. 7 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] 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 numbers used in this way can be interchanged 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.

[0049] The task processing method provided in the embodiment of the present application can be applied to the application environment shown in (a) or (b) of Figure 1, including a central processing unit 102 and a task processor 104. The task processor 104 includes a graphics processor and a neural network processor. The neural network processor is built into the central processing unit 102. The graphics processor can be built into the central processing unit 102, or it can be independently set up and communicated with the central processing unit 102. In specific applications, the central processing unit 102 receives tasks to be processed for intelligent application services, obtains task characteristics of the tasks to be processed in multiple preset dimensions; determines the task type of the tasks to be processed based on the task characteristics of each dimension; and assigns the tasks to be processed to the corresponding task processor 104 according to the task type. This achieves the optimization of the computing process of various AI applications, which can significantly improve the performance and energy efficiency of these applications, reduce energy consumption, and improve the stability and reliability of the overall system.

[0050] In one embodiment, as shown in FIG. 2 , a task processing method is provided, which is described by taking the method applied to the central processor 102 in FIG. 1 as an example, and includes the following steps:

[0051] Step S210: receiving a task to be processed for the intelligent application service, and obtaining task features of the task to be processed in multiple preset dimensions.

[0052] Among them, intelligent application services refer to AI application services such as machine vision, speech recognition, and natural language processing.

[0053] Among them, the preset multiple dimensions may include task content, performance requirements, and user requirements.

[0054] In a specific implementation, after receiving the task to be processed, the central processor extracts the task features under multiple preset dimensions (such as task content, performance requirements, user requirements, etc.), and different feature extraction methods can be used according to different dimensions.

[0055] Step S220: determining the task type of the task to be processed based on the task characteristics of each dimension.

[0056] The task type includes a first task type and a second task type. The first task type represents a task that requires high-performance display, and the second task type represents a task that requires high computing power.

[0057] In the specific implementation, the task type is determined by applying preset rules or machine learning models based on the extracted task features. For example, if the task content is image data and the delay requirement is within a certain threshold, the task type is determined to be a "graphics processing task", which belongs to the first task type. If the task is computational data and has high complexity, the task type is determined to be a "computation-intensive task", which belongs to the second task type.

[0058] Step S230: assigning the tasks to be processed to corresponding task processors according to the task types.

[0059] In a specific implementation, after determining the task type, the task can be allocated according to the task type. Specifically, if the task is a graphics processing task, it is allocated to the graphics processor. If the task is a computationally intensive task, it is allocated to the neural network processor.

[0060] In the above-mentioned task processing method, the task characteristics of the task to be processed in multiple preset dimensions are obtained through the central processing unit; based on the task characteristics of each dimension, the task type of the task to be processed is determined; according to the task type, the task to be processed is assigned to the corresponding task processor. By optimizing the processing flow of AI application tasks, the performance and energy efficiency of these applications can be significantly improved, energy consumption can be reduced, and the stability and reliability of the overall system can be improved. By quickly identifying the task type and reasonably allocating the processor, this method reduces the response time of the task and meets the application requirements with real-time requirements. For example, industrial robots can process sensor data in real time through a neural network processor to achieve more accurate control and decision-making.

[0061] In an exemplary embodiment, the above step S210 obtains task features of the task to be processed in multiple preset dimensions, including: obtaining the task content and performance requirements of the task to be processed; and using the task content and performance requirements as multiple task features of the task to be processed.

[0062] The task content may be graphics, audio, text or video data, etc.

[0063] Among them, performance requirements may include latency requirements, processing power, resource consumption (such as memory and computing time), etc.

[0064] Specifically, different feature extraction methods can be used according to different dimensions. For example, for the task content dimension: natural language processing technology can be used to analyze the description of the task to extract whether the task content is image processing, video analysis, or data computing. For the performance requirement dimension: the latency requirement, throughput, energy consumption, etc. of the task can be evaluated to determine the performance requirement of the task.

[0065] In some embodiments, when there are a large number of tasks to be processed and multiple task processors are required, the preset dimensions may also include task priority. The task type is determined based on the task content, performance requirements and task priority. Tasks with higher priorities can obtain faster processing resources.

[0066] In this embodiment, through multi-dimensional analysis of task features, the task type can be identified more intelligently, thereby optimizing the task processing efficiency.

[0067] In an exemplary embodiment, the above-mentioned step S220 determines the task type of the task to be processed based on the task characteristics of each dimension, including: when the task content is graphic data and the performance requirement meets the first preset condition, determining the task type as a first task type requiring high-performance display; when the task content is computing data and the performance requirement meets the second preset condition, determining the task type as a second task type requiring high computing power.

[0068] In the specific implementation, if the task content is graphic data, when the delay requirement of the performance requirement is less than a certain threshold (such as 10 milliseconds), the task type is determined as the first task type that requires high-performance display. If the task content is computing data, when the computational complexity of the performance requirement is greater than a certain set value, and the requirement for processing time is relatively loose (the delay requirement is greater than 30 milliseconds), the task type is determined as the second task type that requires high computing power.

[0069] For example, tasks such as graphics rendering and video processing are of the first task type, and tasks such as deep learning reasoning and model training are of the second task type.

[0070] In this embodiment, tasks are dynamically allocated to corresponding processors according to their characteristics, thus avoiding unreasonable resource allocation, improving the overall system performance, and making effective use of computing resources.

[0071] In an exemplary embodiment, the above step S230 assigns the task to be processed to the corresponding task processor according to the task type, including: if the task type is the first task type, assigning the task to be processed to the graphics processor; if the task type is the second task type, assigning the task to be processed to the neural network processor.

[0072] Specifically, the graphics processing unit (GPU) is used to process graphics and visual rendering tasks and can effectively handle parallel computing. The neural network processing unit (NPU) is designed to accelerate neural network computing and can efficiently perform matrix operations and vector operations in deep learning models. If it is determined that the task type belongs to the first task type that requires high-performance display, the task to be processed is assigned to the graphics processing unit; if the task type belongs to the second task type that requires high computing power, the task to be processed is assigned to the neural network processor.

[0073] In some embodiments, a fault tolerance and recovery mechanism can also be set. During the task execution process, the processing status and performance of the task processor can be continuously monitored. When any processor fails or does not respond for a long time, the task is transferred to other available task processors for processing to ensure the stability and reliability of the system. In one embodiment, the task allocation strategy can also be adjusted according to the monitoring situation to optimize resource usage and improve processing efficiency.

[0074] In some embodiments, task processing by a neural network processor mainly includes the following steps:

[0075] Data input: Input data is first sent to the input buffer of the neural network processor.

[0076] Computational scheduling: The control unit of the neural network processor schedules computing resources according to the requirements of the neural network model and prepares to execute specific computing tasks.

[0077] Parallel computing: The computing units of the neural network processor perform various operations in the neural network in parallel, such as matrix multiplication, vector addition, etc.

[0078] Result output: After the calculation is completed, the calculation result is output to the output buffer for subsequent processing or storage.

[0079] Feedback Adjustment: When training a neural network, the neural network processor also participates in the back-propagation process to calculate gradients and update the network weights.

[0080] In this embodiment, by assigning tasks to the most suitable processor, high-performance display tasks and high-computing power tasks are optimized and processed separately, so that the system can utilize computing resources with the highest efficiency. Since the overall system response time and execution efficiency are improved, it can meet the application requirements of higher real-time performance.

[0081] In an exemplary embodiment, after the above step S230 assigns the task to be processed to the corresponding task processor according to the task type, it also includes: outputting and displaying the processing result through the display interface of the graphics processor, or receiving the processing result returned by the graphics processor; receiving the processing result returned by the neural network processor.

[0082] In the specific implementation, if the task is assigned to the graphics processor, after the graphics processor completes the processing, the processing result can be directly output through the display interface. For example, for image processing tasks, the generated image data can be displayed on the screen. For tasks assigned to the neural network processor, the calculation results returned by it need to be received, which can be predicted values, classification outputs or other relevant data. Usually this data will be used for further applications or decision-making.

[0083] In this embodiment, by directly displaying the processing results on the user interface, users can quickly get feedback, which improves the interactive experience. For example, in a graphics processing task, the processed image or video effect is presented in a timely manner. The result processing of the neural network processor focuses on numerical and data output, which can be used by other subsequent processing links or user analysis.

[0084] Referring to FIG3 , which is a flowchart of a task allocation method provided in another embodiment, in this embodiment, the following steps are included:

[0085] Step S310, receiving a task to be processed for the intelligent application service, and obtaining the task content and performance requirements of the task to be processed;

[0086] Step S320, when the task content is graphic data and the performance requirement meets the first preset condition, determining the task type as a first task type requiring high-performance display;

[0087] Step S330, assigning the task to be processed to the graphics processor;

[0088] Step S340, outputting and displaying the processing result through the display interface of the graphics processor, or receiving the processing result returned by the graphics processor;

[0089] Step S350, when the task content is computing data and the performance requirement meets the second preset condition, determining the task type as a second task type requiring high computing power;

[0090] Step S360, assigning the task to be processed to the neural network processor;

[0091] Step S370, receiving the processing result returned by the neural network processor.

[0092] In this embodiment, by optimizing the processing flow of AI application tasks, the performance and energy efficiency of these applications can be significantly improved, energy consumption can be reduced, and the stability and reliability of the overall system can be improved. By quickly identifying the task type and reasonably allocating processors, this method reduces the response time of the task and meets the application requirements of real-time requirements. For example, industrial robots can process sensor data in real time through neural network processors to achieve more accurate control and decision-making.

[0093] In one embodiment, as shown in FIG4 , the present application further provides a computer system, including a central processing unit, a graphics processing unit, and a neural network processor;

[0094] The central processing unit is used to receive tasks to be processed for intelligent application services, obtain task characteristics of the tasks to be processed in multiple preset dimensions; determine the task type of the tasks to be processed based on the task characteristics of each dimension; and assign the tasks to be processed to the graphics processing unit or the neural network processor according to the task type;

[0095] Graphics processors and neural network processors are used to handle pending tasks assigned by the central processing unit.

[0096] In some embodiments, the neural network processor is built into the central processing unit; the graphics processor is built into the central processing unit, or is independently configured and communicatively connected to the central processing unit.

[0097] The present application provides an AI industrial computer based on a neural network processing unit, specifically, a neural network processor in the computer works in coordination with a central processing unit and a graphics processing unit. The central processing unit allocates tasks, tasks that require high-performance display such as image processing are allocated to the graphics processor, and tasks that require high computing power such as AI computing are allocated to the neural network processor, and together they complete complex computing tasks. Among them, the neural network processor is a processor specially designed for artificial intelligence applications such as deep learning and neural networks. The neural network processor significantly improves processing speed and energy efficiency by optimizing the computing process of the neural network.

[0098] Referring to FIG5 , a schematic diagram of an integrated AI computer system provided by another embodiment of the present application is provided, which integrates a central processing unit, a graphics processing unit, and a neural network processor, and supports DP, HDMI, EDP, VGA, and Type C display functions to meet the display requirements of different applications, and is configured with 6 COM interfaces, 2 2.5G high-speed Ethernet interfaces, 1 WIFI interface, 1 4G / 5G interface, 8 USB interfaces, and 2 sets of audio input and output interfaces to meet various data access requirements. Among them, the WIFI interface and the 4G / 5G interface can be used as both an input interface and an output interface.

[0099] The integrated AI computer system provided by the present application has the following beneficial effects: (1) High efficiency: Computers supporting neural network processors are more efficient than traditional computers when performing deep learning tasks. (2) Low power consumption: Due to the optimized design of the neural network processor, industrial computers consume less energy than other computers when performing deep learning tasks. (3) Fast response: Through the collaboration of the neural network processor, the central processing unit, and the graphics processing unit, the computing needs of the neural network can be quickly processed.

[0100] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0101] Based on the same inventive concept, the embodiment of the present application also provides a task processing device for implementing the task processing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more task processing device embodiments provided below can refer to the limitations on the task processing method above, and will not be repeated here.

[0102] In one embodiment, as shown in FIG6 , a task processing device is provided, including: an acquisition module 610, a determination module 620 and an allocation module 630, wherein:

[0103] The acquisition module 610 is used to receive the to-be-processed tasks for the intelligent application service, and obtain the task features of the to-be-processed tasks in multiple preset dimensions;

[0104] A determination module 620, configured to determine the task type of the task to be processed based on the task characteristics of each dimension;

[0105] The allocation module 630 is used to allocate the tasks to be processed to corresponding task processors according to the task types.

[0106] In one embodiment, the acquisition module 610 is further used to acquire the task content and performance requirements of the task to be processed; and use the task content and performance requirements as multiple task features of the task to be processed.

[0107] In one embodiment, the determination module 620 is also used to determine that the task type is a first task type requiring high-performance display when the task content is graphic data and the performance requirement meets the first preset condition; and to determine that the task type is a second task type requiring high computing power when the task content is computing data and the performance requirement meets the second preset condition.

[0108] In one embodiment, the allocation module 630 is further used to allocate the task to be processed to the graphics processor if the task type is the first task type; if the task type is the second task type, allocate the task to be processed to the neural network processor.

[0109] In one of the embodiments, the device also includes a result processing module, which is used to output and display the processing results through the display interface of the graphics processor, or receive the processing results returned by the graphics processor; receive the processing results returned by the neural network processor.

[0110] Each module in the above task processing device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0111] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG7 . The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data during task processing. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a task processing method is implemented.

[0112] Those skilled in the art will understand that the structure shown in FIG. 7 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0113] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0114] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0115] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0116] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0117] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0118] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A task processing method, characterized in that: Applied to a central processing unit, the method comprises: Receiving a task to be processed for an intelligent application service, and obtaining task features of the task to be processed in a plurality of preset dimensions; Determine the task type of the task to be processed based on the task characteristics of each dimension; According to the task type, the to-be-processed task is assigned to a corresponding task processor.

2. The method according to claim 1, characterized in that The obtaining of task features of the to-be-processed task in multiple preset dimensions includes: Obtaining the task content and performance requirements of the task to be processed; The task content and the performance requirement are used as a plurality of task features of the task to be processed.

3. The method according to claim 2, characterized in that The determining of the task type of the task to be processed based on the task characteristics of each dimension includes: When the task content is graphic data and the performance requirement meets a first preset condition, determining that the task type is a first task type requiring high-performance display; When the task content is computing data and the performance requirement meets the second preset condition, the task type is determined to be a second task type requiring high computing power.

4. The method according to claim 1, characterized in that: The step of allocating the to-be-processed tasks to corresponding task processors according to the task types includes: If the task type is the first task type, allocating the task to be processed to a graphics processor; If the task type is the second task type, the task to be processed is allocated to a neural network processor.

5. The method according to claim 4, characterized in that After allocating the to-be-processed tasks to corresponding task processors according to the task types, the method further includes: Outputting and displaying the processing result through the display interface of the graphics processor, or receiving the processing result returned by the graphics processor; Receive the processing result returned by the neural network processor.

6. A task processing device, characterized in that: Set in a central processing unit, the device includes: An acquisition module, used to receive a task to be processed for an intelligent application service, and acquire task features of the task to be processed in a plurality of preset dimensions; A determination module, used to determine the task type of the task to be processed based on the task characteristics of each dimension; The allocation module is used to allocate the to-be-processed tasks to corresponding task processors according to the task types.

7. A computer system, characterized in that: The computer system includes a central processing unit, a graphics processing unit and a neural network processor; The central processor is used to receive tasks to be processed for intelligent application services, and obtain task features of the tasks to be processed under multiple preset dimensions; Determine the task type of the task to be processed based on the task characteristics of each dimension; assign the task to be processed to the graphics processor or the neural network processor according to the task type; The graphics processor and the neural network processor are used to process the tasks to be processed assigned by the central processing unit.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the task processing method according to any one of claims 1 to 5 are implemented.

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

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the task processing method according to any one of claims 1 to 5 are implemented.

Images