Data processing method and data processing device

Abstract

The invention discloses a data processing method and a related device, and the method comprises the steps: responding to the obtained task processing data, carrying out the at least one processing of the task processing data according to a target processing strategy, and obtaining the to-be-transmitted data, and the task processing data is the data obtained through the analysis of a target processing task by a first electronic device; and transmitting the to-be-transmitted data to target electronic equipment so as to execute a target processing task corresponding to the to-be-transmitted data by using hardware resources of the target electronic equipment.

Description

Technical Field

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

[0002] Due to their limited processing capabilities or their own processing resources being occupied, electronic devices such as desktops and laptops will request other devices with powerful processing capabilities, such as cloud servers, to process pending tasks. For example, the electronic device sends the pending task to the cloud server, the cloud server processes the pending task, and feeds back the processing result to the electronic device. This solution of processing the pending tasks of electronic devices by sharing the computing power of the cloud server can make up for the deficiency that the electronic device cannot complete the processing alone. In the above-mentioned computing power sharing scheme, data is transmitted in the form of a string, which takes a long time to transmit and has a high overhead. It can be seen that how to achieve efficient transmission of data in the computing power sharing scheme has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a data processing method and a data processing device to at least solve the above technical problems existing in the prior art.

[0004] According to a first aspect of the present application, a data processing method is provided, which is applied to a first electronic device, and the method includes:

[0005] In response to obtaining the task processing data, processing the task processing data at least once according to the target processing strategy to obtain data to be transmitted, wherein the task processing data is data obtained by parsing the target processing task by the first electronic device;

[0006] The data to be transmitted is transmitted to a target electronic device, so that the hardware resources of the target electronic device are used to execute a target processing task corresponding to the data to be transmitted.

[0007] In one possible implementation manner, the step of processing the task processing data at least once according to the target processing strategy to obtain the data to be transmitted includes:

[0008] Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set;

[0009] Based on the second reference data corresponding to the classification set, the classification set is packaged according to a target packaging strategy to obtain the data to be transmitted, and the first reference data and the second reference data are the same or different.

[0010] In one possible implementation manner, the step of processing the task processing data at least once according to the target processing strategy to obtain the data to be transmitted includes:

[0011] Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set;

[0012] Based on the third reference data corresponding to the classification set, configuring the classification set with a target configuration strategy to obtain a data set to be packaged;

[0013] Based on the second reference data corresponding to the classification set, the data set to be packaged is packaged according to a target packaging strategy to obtain the data to be transmitted.

[0014] In one possible implementation, the task processing data is processed using a target classification strategy to obtain a classification set, including at least one of the following:

[0015] The task processing data includes a processing function and corresponding parameters, and the processing function is classified and processed based on the first reference data of the processing function to obtain a first classification set, so that after the target electronic device receives the data to be transmitted, the processing function running on the hardware resources of the target electronic device is used to execute the target processing task;

[0016] The task processing data includes a processing function and a processing rule corresponding to the processing function, and the processing function is classified and processed based on the first reference data of the processing function to obtain a second classification set, so as to execute the processing rule after the target electronic device obtains a processing result of the processing function;

[0017] The first reference data includes at least one of the following:

[0018] The function type of the processing function;

[0019] The frequency with which the processing function is called during execution;

[0020] The data priority of the processing function during execution;

[0021] The association relationship between the processing functions.

[0022] In one possible implementation, the task processing data is processed using a target classification strategy to obtain at least one classification set, including:

[0023] Performing a first classification process on the task processing data based on first sub-reference data in the first reference data to obtain at least one first subset;

[0024] Performing a second classification process on the first subset based on second sub-reference data in the first reference data to obtain a classification set;

[0025] The first reference data includes at least one of the following:

[0026] The function type of the processing function;

[0027] The frequency with which the processing function is called during execution;

[0028] The data priority of the processing function during execution;

[0029] The association relationship between the processing functions;

[0030] the number of the target electronic devices;

[0031] The number of processing units of the second electronic device, where the second electronic device is the electronic device that receives the data to be transmitted.

[0032] In one possible implementation manner, the step of packaging the data set to be packaged using a target packaging strategy to obtain the data to be transmitted includes:

[0033] Based on the second reference data, the to-be-packed data set is packaged to generate packet header data;

[0034] Packaging the task processing data to generate package data;

[0035] Using the data composed of the packet header data and the packet body data as the data to be transmitted;

[0036] Wherein, the second reference data includes at least one of the following:

[0037] the number of the target electronic devices;

[0038] The device identification of the target electronic device;

[0039] The hash value of the data set to be packaged.

[0040] According to a second aspect of the present application, a data processing method is provided, which is applied to a second electronic device, and the method includes:

[0041] In response to obtaining the data to be transmitted, parsing the data to be transmitted to obtain task processing data;

[0042] The target processing task corresponding to the task processing data is executed using the hardware resources of the second electronic device to obtain task result data.

[0043] In one possible implementation manner, the parsing the data to be transmitted to obtain task processing data includes:

[0044] Based on the second reference data, the data to be transmitted is parsed using a target parsing strategy to obtain at least one classification set;

[0045] Based on the first reference data, the classification set is restored using a target restoration strategy to obtain the task processing data.

[0046] In one possible implementation manner, the using the hardware resources of the second electronic device to execute the target processing task corresponding to the task processing data includes:

[0047] In the case where there are multiple second electronic devices, based on the third reference data, the target processing task is executed using the hardware resources of the multiple second electronic devices; or,

[0048] In the case that the second electronic device is unique, the target processing task is executed by using multiple computing units of the second electronic device based on the third reference data.

[0049] According to a third aspect of the present application, there is provided a data processing device, comprising:

[0050] A processing unit, configured to, in response to obtaining task processing data, process the task processing data at least once according to a target processing strategy to obtain data to be transmitted, wherein the task processing data is data obtained by parsing the target processing task;

[0051] The transmission unit is used to transmit the data to be transmitted to a target electronic device, so as to utilize the hardware resources of the target electronic device to execute a target processing task corresponding to the data to be transmitted.

[0052] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, wherein:

[0054] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0055] Figure 1 The schematic diagram shows the implementation process of the data processing method applied to the first electronic device in the embodiment of the present application. Figure 1 ;

[0056] Figure 2 The schematic diagram shows the implementation process of the data processing method applied to the first electronic device in the embodiment of the present application. Figure 2 ;

[0057] Figure 3 The schematic diagram shows the implementation process of the data processing method applied to the first electronic device in the embodiment of the present application. Figure 3 ;

[0058] Figure 4 A schematic diagram of the implementation flow of the data processing method applied to the second electronic device in an embodiment of the present application is shown;

[0059] Figure 5 The implementation block diagram of the data processing method in the embodiment of the present application is shown. Figure 1 ;

[0060] Figure 6 The implementation block diagram of the data processing method in the embodiment of the present application is shown. Figure 2 ;

[0061] Figure 7 A schematic diagram of the composition structure of the first data processing device in an embodiment of the present application is shown;

[0062] Figure 8 A schematic diagram of the structure of the second data processing device in an embodiment of the present application is shown;

[0063] Fig. 9 A schematic diagram of the structure of an electronic device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0064] In order to make the purpose, features, and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0065] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0066] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0067] In the following description, the terms "first\second" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0069] It should be understood that in the various embodiments of the present application, the size of the serial number of each implementation process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0070] The present application relates to two electronic devices: a first electronic device and a target electronic device. Among them, the first electronic device can be regarded as a device with limited processing resources or occupied processing resources in a computing power sharing scheme. The target electronic device can be regarded as a device with powerful processing capabilities or resources in a computing power sharing scheme. Among them, the first electronic device can be devices such as servers, mobile phones, tablet computers, laptops, desktop computers, etc. The target electronic device can be a server device with large computing power such as a cloud server, an ordinary server, a server cluster or a distributed system, or it can be a device of the same type as the first electronic device and has a communication connection with the first electronic device. The first electronic device and the target electronic device can be directly or indirectly connected via wired or wireless communication.

[0071] Figure 1 The implementation process of the data processing method in the embodiment of the present application is shown in FIG. Figure 1 The data processing method is applied to the first electronic device and can be implemented through the following steps.

[0072] S101: In response to obtaining task processing data, the task processing data is processed at least once according to a target processing strategy to obtain data to be transmitted, wherein the task processing data is data obtained by the first electronic device parsing the target processing task.

[0073] In this step, the target processing task is a task that needs to be implemented through a computing power sharing solution parsed by the first electronic device, and is a task to be processed. For example, an intelligent question-and-answer task based on a large model, a graph-generating task. The task processing data may be the unit data constituting the target processing task. In the computer field, processing tasks are all completed by numerical calculation methods, and the numerical calculation method can be implemented in the form of functions. For example, in a task in which data is first added and then multiplied and finally image processing is performed, the addition function add is first used to implement the addition operation of two or more numerical values, and the multiplication function multiply is used to implement the multiplication operation of two or more numerical values. Finally, the image processing function process Image can be used to implement image processing, such as time-frequency domain transformation, inverse transformation, fitting, etc. It can be seen that the task to be processed in this application can be considered as a task implemented by calculating multiple or a large number of functions. Functions can be regarded as unit data constituting the target processing task. That is, the task processing data may be the functions constituting the target processing task. During implementation, the target processing task is parsed using the first electronic device to obtain the task processing data constituting the target processing task.

[0074] In the present application, when task processing data is obtained, the task processing data can be responded to, and the task processing data can be processed once or multiple times according to the target processing strategy to obtain data to be transmitted that can be transmitted. Different from the scheme in the related art where the data to be transmitted is the task to be processed, the data to be transmitted in the present application is the data to be transmitted, and the data to be transmitted is the data obtained by processing the task processing data once or multiple times using the target processing strategy. That is, the data to be transmitted in the present application is the data obtained by performing certain processing on the task processing data.

[0075] In the related art, data is transmitted in the form of a string. When data is transmitted in the form of a string, the amount of data transmitted is large and there may be a blocking phenomenon during transmission, which results in a large transmission overhead. In the present application, the target processing strategy is a strategy that can transform task processing data into a strategy with short transmission time and low transmission overhead. The data to be transmitted obtained by performing certain processing on the task processing data according to the target processing strategy is naturally transmission data with short transmission time and low transmission overhead. When transmitting data with such characteristics, the transmission time of the data can be effectively reduced and the efficiency of data transmission can be improved.

[0076] S102: Transmitting the data to be transmitted to the target electronic device, so as to utilize the hardware resources of the target electronic device to execute the target processing task corresponding to the data to be transmitted.

[0077] In this step, it is considered that the task processing data B is the data obtained by parsing a task to be processed A, and the data to be transmitted C is the data obtained by performing certain processing on the task processing data B according to the target processing strategy. Therefore, it can be considered that the target processing task corresponding to the data to be transmitted C is the task to be processed A.

[0078] The target electronic device may be a device that shares computing power with the first electronic device, and the number may be one or more. The hardware resources of the target electronic device may be hardware with certain or greater processing capabilities in the target electronic device, such as a graphics processing unit (GPU), a central processing unit (CPU), an embedded neural network processor (NPU), a digital signal processor (DSP), etc. This type of hardware in the target electronic device may replace the first electronic device to implement the processing of the target processing task of the first electronic device.

[0079] During implementation, the data to be transmitted obtained by processing once or multiple times according to the target processing strategy is transmitted to one or more second electronic devices different from the first electronic device, so as to utilize the hardware resources of the one or more second electronic devices to execute the target processing tasks corresponding to the data to be transmitted.

[0080] In S101-S102, the target processing task is parsed to obtain task processing data, and in response to the task processing data, the task processing data is processed once or multiple times according to the target processing strategy to obtain data to be transmitted with short transmission time and low transmission overhead. Such data to be transmitted is transmitted to the target electronic device to implement the computing power sharing solution, which can reduce the transmission overhead and duration of the data and ensure the high efficiency of data transmission in the computing power sharing solution.

[0081] In some embodiments, the technical solution of processing the task processing data at least once according to the target processing strategy to obtain the data to be transmitted can be achieved by Figure 2 The method shown in one or by Figure 3 The second method shown is implemented. One of the two solutions is implemented:

[0082] like Figure 2 As shown, the first method can be:

[0083] S201: Based on first reference data corresponding to the task processing data, the task processing data is processed using a target classification strategy to obtain at least one classification set.

[0084] S202: Based on the second reference data corresponding to the classification set, the classification set is packaged according to the target packaging strategy to obtain data to be transmitted, and the first reference data and the second reference data are the same or different.

[0085] In the first method, the target processing strategy includes a target classification strategy and a target packaging strategy. The task processing data is processed according to the target processing strategy, including two steps of classification processing of the task processing data using the target classification strategy and packaging processing of the classification set using the target packaging strategy, which is referred to as a two-step process including classification and packaging. In the two-step process, each step uses corresponding reference data and uses respective strategies to achieve classification and packaging of the task processing data.

[0086] In the present application, task processing data includes processing functions and corresponding parameters. The first reference data includes at least one of the following: the function type of the processing function; the calling frequency of the processing function during the execution process; the data priority of the processing function during the execution process; the association relationship between the processing functions. Among them, the function type of the processing function includes: mathematical operation function, logical operation function, image processing function, etc. The calling frequency of the processing function during the execution process includes: the calling frequency of the processing function during the execution process is high, low or general. The data priority of the processing function during the execution process includes: the priority of the function with high importance in the same task is higher than the priority of the function with low importance, the priority of the function appearing in the front in the same task is higher or lower than the priority of the function appearing later, the priority of the function with more parameters in the same task is higher or lower than the priority of the function with fewer parameters, and the priority of the function with a large amount of data in the same task is higher or lower than the priority of the function with a small amount of data, which can be selected and configured according to actual needs. The association relationship between the processing functions includes: two or more processing functions belong to the same task, and the calculation of some processing functions depends on the calculation results of other processing functions.

[0087] For example, assume that the functions obtained by parsing a target processing task include an addition function add, a multiplication function multiply, and an image processing function process Image. Among them, the function types of add and multiply are mathematical operation functions, with high calling frequency and small data volume. Process Image is an image processing function with low calling frequency and large data volume. For these three functions, if they are classified from the perspective of function type, that is, classified according to the target classification strategy of function type, they can be classified into two classification sets including data operation functions and image processing functions, classification set 1 and classification set 2. Among them, classification set 1 includes add and multiply. Classification set 2 includes process Image. For these three functions, if they are classified from the perspective of data volume, that is, classified according to the target classification strategy of data priority, they can be classified into two classification sets with small data volume and large data volume, classification set 1 and classification set 2. Among them, classification set 1 includes add and multiply. Classification set 2 includes process Image.

[0088] Each classification set can be regarded as a classification block, and each classification block corresponds to a classification block number and classification information. The classification block number is used to uniquely indicate each classification set, and the classification information is used to indicate which reference data of the first reference data is used for classification and the reference data value under the classification. For example,

[0089] Classification block 1 (classification block number of classification set 1):

[0090] Function type: mathematical operation (classification information of classification set 1);

[0091] Function: add(a,b); multiply(x,y).

[0092] Classification block 2 (classification block number of classification set 2):

[0093] Function type: Image processing (classification information of classification set 2);

[0094] Function: process Image(imgData).

[0095] In the packing step, the second reference data includes the classification block number and / or the classification information of the classification block. If the second reference data is the classification block number, the target packing strategy may be a strategy related to the classification block number. For example, according to the target packing strategy, a classification set with an odd-numbered classification block number may be packed into one data packet, and a classification set with an even-numbered classification block number may be packed into one data packet. If the second reference data is the classification information of the classification block, the target packing strategy may be a strategy related to the classification information. For example, according to the target packing strategy, a classification set with the same functional type may be packed into one data packet, a classification set with a higher priority may be packed into one data packet, and a classification set with a lower priority may be packed into one data packet. Each packaged data packet may be used as data to be transmitted.

[0096] The scheme shown in S201~S202 is applicable to the case where the target electronic device is one electronic device. In the case where the target electronic device is one electronic device, the first electronic device only needs to perform two steps of classification and packaging to obtain the data to be transmitted that needs to be transmitted to the same electronic device. Among them, based on the first reference data, the scheme of classifying the task processing data with a target classification strategy can facilitate the subsequent implementation of the packaging step and improve the packaging efficiency. Based on the second reference data, the classification set is packaged using the target packaging strategy to obtain the data to be transmitted in the form of a data packet. Compared with the scheme of transmitting data in the form of a character string in the related art, the transmission in the form of a data packet in the present application can reduce the data transmission time and overhead and improve the data transmission efficiency.

[0097] like Figure 3 As shown, the second method can be:

[0098] S301: Based on first reference data corresponding to the task processing data, the task processing data is processed using a target classification strategy to obtain at least one classification set.

[0099] S302: Based on the third reference data corresponding to the classification set, the classification set is configured with a target configuration strategy to obtain a data set to be packaged.

[0100] S303: Based on the second reference data corresponding to the classification set, the data set to be packaged is packaged according to the target packaging strategy to obtain data to be transmitted.

[0101] In the second method, the target processing strategy includes a target classification strategy, a target configuration strategy and a target packaging strategy. The task processing data is processed according to the target processing strategy, including three steps of classification processing of the task processing data using the target classification strategy, configuration processing of the classification set using the target configuration strategy, and packaging processing of the data set to be packaged using the target packaging strategy, which is referred to as the three-step processing including classification, configuration and packaging. In the three-step processing, each step uses corresponding reference data and uses respective strategies to achieve classification, configuration and packaging of the task processing data.

[0102] In the case where the task processing data includes processing functions and corresponding parameters, the specific implementation of the classification step can be found in Figure 2 The relevant instructions are not repeated here. In actual applications, the data to be transmitted can be transmitted to one electronic device, or to two or more electronic devices (when there are two or more target electronic devices). Method 2 in the present application is suitable for the situation where the data to be transmitted is transmitted to two or more electronic devices. The third reference data may be the address information of the target electronic device and / or the quantity information of the target electronic device. In the case where at least one classification set is obtained through the classification processing of S301, the classification set is configured with a target configuration strategy, such as configuring the address information of the electronic device to which the classification set needs to be transmitted for each classification set, and the classification set is configured to obtain a set of data to be packaged. For example,

[0103] Data set 1 to be packaged (the number of data set 1 to be packaged):

[0104] Function type: mathematical operation;

[0105] Transmission address: IP address of device 1;

[0106] Function: add(a,b); multiply(x,y).

[0107] Data set 2 to be packaged (the number of data set 2 to be packaged):

[0108] Function type: Image processing;

[0109] Transmission address: IP address of device 2;

[0110] Function: process Image(imgData).

[0111] In the above examples, the functions in each data set to be packaged appear in the form of functions and function parameters. In addition, the functions in each data set to be packaged may also appear in the form of a function list. The function list may include information such as functions, function parameters, function identifiers (IDs), and function names. In the data sets to be packaged, the numbers of the data sets to be packaged are used to indicate each packaged data set. Information such as function type, transmission address, function identifier, and function name may be used as classification information items.

[0112] In S303, if the second reference data is a classification block number, according to the target packaging strategy, every 10 data sets to be packaged can be packaged into a data packet, such as packaging number 1 (the number of data set 1 to be packaged)-number 10 (the number of data set 10 to be packaged) into a data packet. If the second reference data is classification information such as transmission address information, then according to the target packaging strategy, the data sets to be packaged with the same transmission address can be packaged into a data packet, and the data sets to be packaged with different transmission addresses can be packaged into different data packets. Each packaged data packet is transmitted from the first electronic device to each electronic device having the transmission address of the data packet. For example, packaged data packet 1 is transmitted to cloud server 1, and packaged data packet 2 is transmitted to cloud server 2.

[0113] In the present application, the data to be transmitted can be transmitted to different (hardware) computing units of the same electronic device. Different (hardware) computing units are: GPU, CPU, DSP, etc. In this case, the transmission addresses configured for different sets of data to be packaged can be the identifiers of different computing units to realize the transmission of data packets to different computing units of the same electronic device.

[0114] The classification scheme for task processing data shown in S301 to S303 can facilitate the subsequent implementation of the packaging steps and improve the packaging efficiency. The configuration scheme for the classification set provides a guarantee for the transmission of different data packets to different electronic devices. For the packaging scheme of the data set to be packaged, the data to be transmitted is obtained in the form of a data packet. Compared with the scheme of transmitting data in the form of a string in the related art, the data packet transmission of the present application can reduce the data transmission time and overhead and improve the data transmission efficiency. It provides a technical support for the accurate and efficient transmission to two or more electronic devices in the computing power sharing scheme.

[0115] In the above-mentioned scheme, the address of the target electronic device appearing in the data set to be packaged is used as an example for explanation. In addition, considering that the function result calculated by the target electronic device in the computing power sharing scheme also needs to be fed back to the first electronic device, in order to achieve accurate feedback, the address of the first electronic device can also be used as a classification item information and appear in the data set to be packaged or the classification set.

[0116] exist Figure 2 and Figure 3 In the illustrated scheme, the task processing data is processed with a target classification strategy to obtain a classification set, including at least one of the following:

[0117] Solution A: If the task processing data includes a processing function and corresponding parameters, the processing function is classified based on the first reference data of the processing function to obtain a first classification set, so that after the target electronic device receives the data to be transmitted, the processing function running on the hardware resources of the target electronic device is used to execute the target processing task.

[0118] Solution B: If the task processing data includes a processing function and a processing rule corresponding to the processing function, the processing function is classified based on the first reference data of the processing function to obtain a second classification set, so as to execute the processing rule after the target electronic device obtains the processing result of the processing function.

[0119] In Scheme A and Scheme B, please refer to the above-mentioned related instructions for the scheme of classifying the processing functions to obtain the classification set, which will not be repeated here. Compared with Scheme A, Scheme B can not only transmit the data to be transmitted based on the task processing data to the target electronic device, but also transmit the processing rules for the processing function to the target electronic device. The processing rules for the target processing task can be: after calculating the processing function using the hardware resources of the target electronic device, the calculation results of each processing function are integrated using the integration rules, and the integrated results are fed back to the first electronic device.

[0120] For example, in a scenario where the target processing task is to generate a graph from a graph, such as generating a graph from a graph A, the processing function is a function generated to generate a graph A, and the processing function is calculated using the hardware resources of the target electronic device in order to generate a graph A. In the case of generating a graph A, the integration rule C transmitted by the first electronic device to the target electronic device is used as a processing rule for the processing function, and a graph B is generated according to a graph A to achieve completion of the task of generating a graph from a graph.

[0121] Figure 2 and Figure 3The classification scheme in the scheme is a classification scheme for the processing function based on the function type, calling frequency, data priority, association relationship, etc. of the processing function. This classification scheme can be regarded as a classification scheme implemented by one-step classification. In addition, the present application also includes another classification scheme. Another classification scheme for processing the task processing data with a target classification strategy to obtain a classification set can be: performing a first classification process on the task processing data based on the first sub-reference data in the first reference data to obtain at least one first subset; performing a second classification process on the first subset based on the second sub-reference data in the first reference data to obtain a classification set; wherein the first reference data includes at least one of the following: the function type of the processing function; the calling frequency of the processing function during execution; the data priority of the processing function during execution; the association relationship between processing functions; the number of target electronic devices;

[0122] The number of processing units of the second electronic device.

[0123] The first sub-reference data and the second sub-reference data may be different reference data in the aforementioned first reference data, so as to classify the processing function through different reference angles or dimensions.

[0124] For example, assume that the functions obtained by parsing a target processing task include add, multiply, image processing functions process Image 1 and process Image 2. If classified from the perspective of function type (used as the first sub-reference data), that is, classified according to the target classification strategy of function type, it can be classified into two classification subsets including data operation functions and image processing functions, classification subset 1 and classification subset 2. Among them, classification subset 1 includes add and multiply, and classification subset 2 includes process Image 1 and process Image 2. Assuming that add, multiply and process Image 1 are all functions with high priority, and process Image 2 is a function with low priority, if it is further classified from the perspective of priority (used as the second sub-reference data), that is, the first subset is classified according to the perspective of priority, 3 classification sets can be obtained. Among them, classification set 1 includes add and multiply, classification set 2 includes process Image 1, and classification set 3 includes process Image 2.

[0125] The aforementioned scheme is a classification scheme for task processing data through two-step classification, which not only improves the classification accuracy, but also facilitates the implementation of subsequent packaging schemes or configuration and packaging schemes, providing a certain basis for efficient transmission in computing power sharing schemes.

[0126] exist Figure 3 In the illustrated scheme, the first reference data includes the number of target electronic devices. When the target electronic device is the second electronic device, target classification processing is performed on the processing function based on the first reference data to obtain at least one classification set. When the target electronic device is multiple electronic devices, the task processing data is first classified based on the first sub-reference data in the first reference data to obtain a first subset; the first subset is second classified based on the second sub-reference data in the first reference data to obtain a classification set, and the first sub-reference data and the second sub-reference data in the first reference data are different.

[0127] Exemplarily, in the case where the target electronic device is a plurality of electronic devices, the several functions obtained by parsing the target processing task are classified according to the number of second electronic devices (used as the first sub-reference data). The number of second electronic devices is 2, and the several functions are classified into two classification subsets, classification subset 1 and classification subset 2. Each classification subset includes mathematical operation functions and image processing functions, and each classification subset is classified according to the function type (used as the second sub-reference data), to obtain classification set 1 to classification set 4.

[0128] In layman's terms, in the present application, in the classification, configuration and packaging scheme of the processing function, if the target electronic device is one electronic device, a one-step classification scheme can be used to classify the processing function. If the target electronic device is multiple electronic devices, a two-step classification scheme can be used to classify the processing function, thereby improving the classification accuracy of the processing function. This lays a certain foundation for better implementation of packaging or configuration and packaging schemes.

[0129] exist Figure 2 In the scheme shown, the target packaging strategy is used to package the data set to be packaged, and the scheme for obtaining the data to be transmitted may include: packaging the classification set to generate header data; packaging the task processing data to generate body data; and using the data composed of header data and body data as the data to be transmitted.

[0130] In the data to be transmitted obtained by classifying and packaging the processing functions, for the packaging scheme therein, a data packet can be generated based on the classification set. The data packet consists of header data and body data. During implementation, the header data can be generated based on the classification information of the classification block in the classification set, and the body data can be generated based on the processing function in the classification set. Usually, the number of data packets generated based on the classification set can be one or more, preferably more than one. Among them, the header data includes the number of one or more classification blocks and at least one of the multiple classification information of the classification block. Take the following data packet generated based on a classification set as an example. Among them, the header data includes the classification block number of the classification set, the function ID in the classification set, and other classification information items such as the mathematical operation function type function. The body data includes the function name in the classification set and the parameter value of the function.

[0131] Baotou:

[0132] Classification block number: 1

[0133] Function ID: 1

[0134] Category information: Mathematical operations

[0135] Body:

[0136] Function name: add

[0137] Parameters: (a=1, b=2)

[0138] It can be understood that the packet body data in the data packet is payload data-processing function. In the classification set obtained by classifying the first reference data with the target classification strategy, other information except the processing function, such as the classification block number and at least one of the classification information items, can be used as packet header data. By using the payload data as the packet body and other data except the payload data as the packet header, a data packet that is guaranteed to be transmitted smoothly and efficiently can be obtained.

[0139] Usually, payload data refers to pure information data transmitted in network communication except for the data header or the data header and the data tail. Because, in the computing power sharing scheme of the present application, the first electronic device mainly requests the second electronic device to calculate the processing function and the corresponding parameters through the transmission of data, so the processing function and the corresponding parameters can be pure information data in the computing power sharing scheme. The processing function and the corresponding parameters appear in the packet body. Classification item information such as classification block number, function ID, function type of processing function, etc. appears in the packet header.

[0140] because Figure 2The scheme shown is applicable to the case where the target electronic device is one electronic device, that is, one or more data packets formed by packaging are transmitted to the same electronic device, so the transmission address of the same electronic device may not appear in the header data of each data packet. In addition, the transmission address of the same electronic device may also appear as a header data in each data packet that can be transmitted to the same electronic device.

[0141] In the above scheme, one classification set can be packaged into one or more data packets. Multiple classification sets can be packaged into one data packet. The specific packaging of each classification set obtained by classification, such as which classification sets can be packaged into the same data packet, can be determined according to the above target packaging strategy. The process can be referred to the above related understanding and will not be repeated.

[0142] exist Figure 3 In the scheme shown, the scheme for packaging the data set to be packaged with the target packaging strategy to obtain the data to be transmitted may include: packaging the data set to be packaged based on the second reference data to generate header data; packaging the task processing data to generate body data; using the data composed of the header data and the body data as the data to be transmitted; wherein the second reference data includes at least one of the following: the number of target electronic devices; the device identification of the target electronic device; the hash value of the data set to be packaged.

[0143] exist Figure 3 In the scheme shown, the data set to be packaged includes the transmission address item of the target electronic device to which each data set to be packaged can be sent. In this way, if the number of target electronic devices is multiple, if the second reference data is the number of target electronic devices, the number of target electronic devices may appear in the header data, and the data set to be packaged can be packaged into the same number of data packets as the number of target electronic devices. The body data of each data packet includes the payload data in one or more data sets to be packaged, and the header data is the classification information item other than the payload data in one or more data sets to be packaged. If the second reference data is the device identification of the target electronic device, and the header data includes the device identification item, the data sets to be packaged with different device identifications can be packaged into one data packet, or the data sets to be packaged with the same device identification can be packaged into one data packet. If the second reference data is the hash value of the data set to be packaged, the header data can include a hash value, so that when the packaged data packet is transmitted to the target electronic device, the target electronic device verifies whether the received data packet is correct or complete according to the hash value. If the hash data obtained by the target electronic device after verification is consistent with the received hash value, it is considered that the data packet transmission is correct and complete, and there is no packet loss or sticking.

[0144] It can be understood that for the classification block number information and / or function ID included in the packet header data, the target electronic device can determine whether there is a packet loss or packet sticking phenomenon based on the information. The sequentiality between the classification block number information and the sequentiality between the function IDs can ensure that the order is not disordered when sending, reflecting the transmission consistency. For a data set to be packaged, if it contains many processing functions and a large amount of data, it can be packaged into two or more data packets. After the target electronic device receives all the data packets split for the same data set to be packaged, it uses hardware resources to execute the target processing task to ensure data integrity.

[0145] When transmitting the data to be transmitted, the first electronic device and the second electronic device can use an agreed communication protocol such as the Internet Protocol (TCP) and the File Transfer Protocol (FTP) to send and receive data. In this way, the length of each data packet can be specified, such as fixing each data packet to 200 bytes. If one or more classification sets or one or more data sets to be packaged are packaged to form a data packet, the length of which does not reach a certain number of bytes, such as 200 bytes, then spaces can be used to fill in the idle bytes in the data packet, so as to achieve the sequential segmentation or identification of the data packet by the second electronic device in a fixed length manner, and avoid the sticking of data packets.

[0146] It is also possible to utilize the FTP protocol to add separators between data packets to achieve smooth and efficient segmentation of different data packets by the second electronic device, thereby avoiding the problem of being unable to segment or identify each packet due to data packet sticking.

[0147] exist Figure 3 In the scheme shown, the data consisting of header data and body data is used as the data to be transmitted, and may also include: the classification set also includes a second subset, and when the amount of data in the second subset is less than the parameter threshold, the second subset is packaged based on the third reference data to generate second header data containing task processing data, and the second header data is used as the data to be transmitted.

[0148] In this solution, the second subset of the classification set can be the payload data-processing function in the classification set. If the data volume of the payload data in the classification set is less than the parameter threshold, such as the number of processing functions is small or the function parameters of the processing function are small, then in addition to the aforementioned classification block number, function ID, transmission address and other classification item information, the header data can also include the payload data as an information item in the header data. Such a data packet has no body data, only header data, and the header data includes the payload data. In layman's terms, if the amount of function data in the classification set is small, the function and function parameters can be placed in the header data, and transmitted to the target electronic device as a data packet together with other items in the header data. In this way, the transmission efficiency can be increased.

[0149] The above scheme is a data processing scheme applied to the first electronic device. In addition, the present application also provides a data processing method applied to a target electronic device such as a second electronic device. Figure 4 As shown, the method includes:

[0150] S401: In response to obtaining the data to be transmitted, parsing the data to be transmitted to obtain task processing data.

[0151] In this step, a data packet is received from the first electronic device, and in response to the received data packet, the data packet is parsed to obtain a function and function parameters that the first electronic device expects the second electronic device to process.

[0152] S402: Utilize the hardware resources of the second electronic device to execute the target processing task corresponding to the task processing data, and obtain task result data.

[0153] In this step, the same hardware resources of the same electronic device, different hardware resources of the same electronic device, or different electronic devices are used to calculate each function, and the calculation results of each function are used as task result data. Alternatively, after obtaining the calculation results of each function, the results are integrated to a certain extent, and the integrated results are used as task result data to complete the tasks that need to be completed by the target electronic device in the computing power sharing solution.

[0154] Exemplarily, in the target processing task of graph generation, each function parsed by the target electronic device is a function generated to generate graph A. The target electronic device can use the calculation results of each function as task result data. The target electronic device can generate graph B according to graph A based on the graph generation rule (integration rule), and use graph B as task result data.

[0155] In S401-S402, by analyzing the received data to be transmitted, the functions that need to be calculated by the second electronic device are obtained, so as to use the hardware resources to complete the corresponding tasks and ensure the accurate completion of the tasks. Among them, the data to be transmitted is to process the task processing data once and multiple times according to the target processing strategy to obtain data with short transmission time and low transmission overhead. The second electronic device can realize smooth and efficient reception of the data to be transmitted, which provides a guarantee for the accurate completion of the task.

[0156] exist Figure 4 In the data processing scheme shown, parsing the data to be transmitted to obtain the task processing data can be accomplished by the following scheme: based on the second reference data, the data to be transmitted is parsed with a target parsing strategy to obtain at least one classification set. Based on the first reference data, the classification set is restored with a target restoration strategy to obtain the task processing data.

[0157] It can be understood that on the first electronic device side, the data packet is obtained through two steps of classification and packaging, or through three steps of classification, configuration and packaging. Corresponding to the classification and packaging processing at the sending end (the first electronic device end), for the data packet received by the second electronic device, the data packet can be unpacked first to obtain a classification set, and then the classification set can be declassified to restore the processing function.

[0158] Among them, the target parsing strategy is an unpacking strategy, which corresponds to the target packing strategy of the sending end. Exemplarily, if the target packing strategy is to form a data packet for every 10 classification blocks in the order of classification blocks, it can be known that the current data packet is a data packet including 10 classification blocks, then the target parsing strategy can be to split the current data packet into 10 classification sets according to the header data. Read the header data of the current data packet, and for a classification block number obtained from the header data and a function ID contained in the classification block, obtain the function and function parameters with the same identifier as the function ID from the body data, and the function and function parameters are the function and function parameters in the classification block. When the classification block number, function ID, function and function parameters and other classification item information such as the function type obtained from the header data are known, it is easy to obtain the classification set. The example of the classification set is similar to the above and will not be repeated.

[0159] The target restoration strategy is a declassification strategy, which corresponds to the target classification strategy of the sending end. Exemplarily, if the target restoration strategy is a classification based on function type, the target restoration strategy can be to restore the processing functions in each classification block according to the function type. For example, classification set 1 is a classification set whose function type is mathematical operation, and all functions included in the classification set, such as function 1 and function 2, can be used as task processing data restored from classification set 1. For example, classification set 2 is a classification set whose function type is image processing, and all functions included in the classification set, such as function 3 and function 4, can be used as task processing data restored from classification set 2. The task processing data of all classification sets are restored, so that the functions that the first electronic device expects the second electronic device to calculate can be obtained.

[0160] In response to the classification and packaging scheme of the sending end, the receiving end (the second electronic device end) provides a corresponding splitting and declassification scheme, thereby achieving the acquisition of the content that needs to be processed or calculated by the second electronic device. Among them, the processing or calculation content is efficiently transmitted using data packets with a certain form (such as header + body form), so that the second electronic device avoids the problem of slow splitting and declassification due to high transmission overhead and long time, ensures the efficiency of splitting and declassification, and improves the efficiency of completing the target processing task.

[0161] In the aforementioned scheme, utilizing the hardware resources of the second electronic device to execute the target processing task corresponding to the task processing data may include: when there are multiple second electronic devices, based on third reference data, utilizing the hardware resources of multiple second electronic devices to execute the target processing task; or, when there is only one second electronic device, based on the third reference data, utilizing multiple computing units of the second electronic device to execute the target processing task.

[0162] Taking the third reference data as the device identification of each second electronic device as an example, because the address information of the electronic device to which each data packet needs to be transmitted will be displayed in the header part of the data packet, the functions contained in the data packet can also be known, so it is easy to know which functions need to be assigned to each electronic device for processing, and assign the parsed functions in the same target processing task to the electronic device to which each function should be assigned, so that each electronic device can calculate its own function.

[0163] At the sending end, if multiple computing units of the same electronic device are needed to process the target processing task, the transmission address in the packet header data can be the identification of each computing unit of the electronic device. At the receiving end, according to the transmission address in the packet header data, the parsed functions for the same target processing task can be assigned to the computing unit to which each function should be assigned, so that each computing unit of the electronic device can calculate the function it is responsible for.

[0164] For example, assume that there are 10 functions parsed from a target processing task, of which the first 3 functions are processed by device 1, the last 3 functions are processed by device 3, and the middle 4 functions are processed by device 2. Through the respective processing of these three devices and the cooperation with each other, the calculation of all functions of the target processing task is realized. Compared with using one electronic device to calculate all functions of a target processing task, using multiple electronic devices or multiple computing units of the same electronic device to calculate all functions of the target processing task has higher processing efficiency and completes the target processing task faster.

[0165] In the present application, when the second electronic device calculates the result of each function, the result of each function can be fed back to the first electronic device. During implementation, on the second electronic device side, for each function and each function calculation result, the function can be first classified based on the first reference data corresponding to the function to obtain a third classification set. Based on the second reference data corresponding to the classification set, the third classification set is packaged with a target packaging strategy to obtain data to be fed back. That is, on the second electronic device side, for feedback on the function calculation result, a scheme similar to the transmission function on the transmitting end (first electronic device) side can be adopted to classify and package the functions, and then feedback is given to the first electronic device in the form of packaged data packets.

[0166] Among them, the classification scheme for functions at the second electronic device end can be the same as the classification scheme for functions at the first electronic device end. For example, both ends use functional types to classify functions. For a certain function, its function ID and classification block number at the receiving end (the second electronic device) can be the same as the function ID and classification block number at the sending end. The packaging scheme for the classification set at the second electronic device end can be the same as the packaging scheme for the classification set at the first electronic device end. At the receiving end, the packaged data packet may include header data and body data. Among them, the header data is the same as the header data of the data to be transmitted. Unlike the body data in the data to be transmitted, which is the function name and the function parameter value, in the data packet packaged at the receiving end, the body data can be the function name and the function calculation result, or the function calculation result. For example, an example of a data packet packaged by the receiving end:

[0167] Baotou:

[0168] Classification block number: 1

[0169] Function ID: 1

[0170] Category information: Mathematical operations

[0171] Body:

[0172] Calculation result: (result = 3)

[0173] In the present application, the feedback of the calculated function result by the second electronic device to the first electronic device is also feedback in the form of a data packet after function classification and packaging. It can reduce the data transmission time and improve the data transmission efficiency. If multiple second electronic devices calculate the function at the receiving end, each electronic device can use the classification and packaging scheme of the receiving end after calculating the function result, and feedback the function calculation result to the first electronic device in the form of a data packet.

[0174] The first electronic device receives the data to be fed back from the second electronic device, and after unpacking and declassifying the data to be fed back, the function calculation result calculated by the second electronic device can be obtained. For understanding of the scheme for unpacking and declassifying the data to be fed back, you can refer to the scheme for splitting and declassifying the transmitted data by the second electronic device. The unpacking strategy and declassification strategy used in the unpacking and declassification scheme must correspond to the classification strategy and packing strategy when the second electronic device packs the data to be fed back. For details, please refer to the aforementioned related content for understanding, and the similarities will not be repeated.

[0175] Figure 5 and Figure 6 These are two implementation diagrams of the data processing method. Figure 5 and Figure 6 In the example, the local end can be used as the first electronic device, and the server end can be used as the second electronic device. Since the processing resources of the local end's CPU are already occupied, it is expected that the GPU of the server end will share computing power and request the GPU of the server end to realize the calculation of the target processing task. Figure 5 The situation where the local end sends a data packet to request the server's GPU computing target to process the task. Figure 6 This is the situation where the server feeds back the calculation results to the local end.

[0176] exist Figure 5In the process, the local end parses a target processing task to be processed to obtain the functions and function parameters constituting the target processing task. The functions are classified by a classifier to obtain classification sets. The classification sets are packaged by a packager to form multiple data packets. The data packets are transmitted from the local end to the server end through communication protocols such as wireless local area network (WiFi), Ethernet, and Internet Protocol (TCP / IP). On the server end, the data packets are unpacked by a depacketizer, and the classification sets obtained by unpacking are restored by a declassifier. The calculation of each function of the target processing task is realized by using the GPU of the server end.

[0177] exist Figure 6 In the example, when the GPU of the server calculates the results of each function, the classifier is used to classify each function to obtain each classification set. The classification set including the function calculation results is packaged by the packager to form multiple data packets. The data packet is transmitted from the server to the local end through communication protocols such as WiFi, Ethernet, TCP / IP, etc., so as to feedback the function calculation results to the local end. The local end uses the depacketizer to depacketize the data packet, and uses the declassifier to restore the function and the function calculation results of the depacketized classification set.

[0178] From the above content, it can be seen that the data transmission from the local end to the server end and the data feedback from the server end to the local end are both transmitted in the form of a data packet consisting of a packet header and a packet body. Compared with the solution in the related art that uses a string form to transmit the target processing task, in this application, the target processing task is transmitted in a certain data form such as a data packet form, which can reduce transmission overhead and duration, and ensure the efficiency of data transmission in the computing power sharing solution.

[0179] exist Figure 5 and Figure 6 The solution classifier can also be a restorer, which is used to restore functions and function parameters, functions and function calculation results.

[0180] From the perspective of both ends (local end and service end), the technical solution of this application is a computing power sharing solution realized through classification, packaging, transmission, unpacking, declassification and other links. In the computing power sharing solution of this application, each link can be executed independently without interfering with each other, and an efficient computing power sharing solution is realized through the cooperation of each link. It provides a technical support for the computing power sharing solution.

[0181] The present application provides a first data processing device, which may be the aforementioned first electronic device or a device located in the first electronic device. Figure 7 As shown, the device comprises:

[0182] A processing unit 701 is used for processing the task processing data at least once according to a target processing strategy in response to obtaining the task processing data to obtain data to be transmitted, wherein the task processing data is data obtained by parsing the target processing task;

[0183] The transmission unit 702 is used to transmit the data to be transmitted to the target electronic device, so as to use the hardware resources of the target electronic device to perform the target processing task corresponding to the data to be transmitted.

[0184] In some embodiments, the processing unit 701 is configured to:

[0185] Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set;

[0186] Based on the second reference data corresponding to the classification set, the classification set is packaged according to a target packaging strategy to obtain the data to be transmitted, and the first reference data and the second reference data are the same or different.

[0187] In some embodiments, the processing unit 701 is configured to:

[0188] Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set;

[0189] Based on the third reference data corresponding to the classification set, configuring the classification set with a target configuration strategy to obtain a data set to be packaged;

[0190] Based on the second reference data corresponding to the classification set, the data set to be packaged is packaged according to a target packaging strategy to obtain the data to be transmitted.

[0191] In some embodiments, the processing unit 701 is configured to:

[0192] The task processing data includes a processing function and corresponding parameters, and the processing function is classified and processed based on the first reference data of the processing function to obtain a first classification set, so that after the target electronic device receives the data to be transmitted, the processing function running on the hardware resources of the target electronic device is used to execute the target processing task;

[0193] The task processing data includes a processing function and a processing rule corresponding to the processing function. The processing function is classified based on the first reference data of the processing function to obtain a second classification set, so as to execute the processing rule after the target electronic device obtains the processing result of the processing function; wherein the first reference data includes at least one of the following: the functional type of the processing function; the calling frequency of the processing function during the execution process; the data priority of the processing function during the execution process; and the association relationship between the processing functions.

[0194] In some embodiments, the processing unit 701 is configured to:

[0195] Performing a first classification process on the task processing data based on first sub-reference data in the first reference data to obtain a first subset;

[0196] Performing a second classification process on the first subset based on second sub-reference data in the first reference data to obtain a classification set;

[0197] The first reference data includes at least one of the following:

[0198] The function type of the processing function;

[0199] The frequency with which the processing function is called during execution;

[0200] The data priority of the processing function during execution;

[0201] The association relationship between the processing functions;

[0202] the number of the target electronic devices;

[0203] The number of processing units of the second electronic device, where the second electronic device is the electronic device that receives the data to be transmitted.

[0204] In some embodiments, the processing unit 701 is configured to:

[0205] The third reference data includes the number of the target electronic devices;

[0206] In a case where the target electronic device is a second electronic device, performing target classification processing on the processing function based on the first reference data to obtain at least one classification set;

[0207] In the case where the target electronic device is multiple electronic devices, a first classification process is performed on the task processing data based on the first sub-reference data in the first reference data to obtain at least one first subset; a second classification process is performed on the first subset based on the second sub-reference data in the first reference data to obtain a classification set, and the first sub-reference data and the second sub-reference data are different.

[0208] In some embodiments, the processing unit 701 is configured to:

[0209] Based on the third reference data, the to-be-packed data set is packaged to generate packet header data;

[0210] Packaging the task processing data to generate package data;

[0211] Using the data composed of the packet header data and the packet body data as the data to be transmitted;

[0212] The third reference data includes at least one of the following:

[0213] the number of the target electronic devices;

[0214] The device identification of the target electronic device;

[0215] The hash value of the data set to be packaged.

[0216] In some embodiments, the processing unit 701 is configured to:

[0217] The classification set also includes a second subset. When the data volume of the second subset is less than the parameter threshold, the second subset is packaged based on the third reference data to generate second packet header data containing task processing data, and the second packet header data is used as the data to be transmitted.

[0218] The present application provides a second data processing device, which may be the aforementioned second electronic device or a device located in the second electronic device. Figure 8 As shown, the device comprises:

[0219] A parsing unit 901 is used for parsing the data to be transmitted to obtain task processing data in response to obtaining the data to be transmitted;

[0220] The obtaining unit 902 is used to use hardware resources to execute the target processing task corresponding to the task processing data to obtain task result data.

[0221] In some embodiments, the parsing unit 901 is used to:

[0222] Based on the second reference data corresponding to the classification set, the data to be transmitted is parsed with a target parsing strategy to obtain at least one classification set;

[0223] Based on the first reference data corresponding to the classification set, the classification set is restored using a target restoration strategy to obtain the task processing data.

[0224] In some embodiments, the obtaining unit 902 is configured to:

[0225] In the case where there are multiple second electronic devices, based on the third reference data, the target processing task is executed using hardware resources of the multiple second electronic devices; or,

[0226] In the case that the second electronic device is unique, the target processing task is executed by using multiple computing units of the second electronic device based on the third reference data.

[0227] It should be noted that the two data processing devices in the embodiments of the present application solve the problems according to the principles similar to those of the aforementioned data processing method. Therefore, the implementation process and implementation principles of the two data processing devices can be referred to the description of the implementation process and implementation principles of the aforementioned method, and the repeated parts will not be repeated.

[0228] According to an embodiment of the present application, the present application also provides an electronic device and a readable storage medium.

[0229] The electronic device includes at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the data processing method described in the present application. The computer instructions are used to enable the computer to perform the data processing method described in the present application.

[0230] The present application also provides a computer program product, including a computer program / instruction, which implements the data processing method of the present application when executed by a processor.

[0231] Fig. 9A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0232] like Fig. 9 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0233] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0234] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as data processing methods. For example, in some embodiments, the data processing method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the data processing method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the data processing method in any other appropriate manner (e.g., by means of firmware).

[0235] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0236] The program code for implementing the method of the present application can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, implements the functions / operations specified in the flow chart and / or block diagram. The program code can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0237] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0238] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0239] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0240] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0241] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A data processing method, applied to a first electronic device, the method comprising: In response to obtaining the task processing data, processing the task processing data at least once according to the target processing strategy to obtain data to be transmitted, wherein the task processing data is data obtained by parsing the target processing task by the first electronic device; The data to be transmitted is transmitted to a target electronic device, so that the hardware resources of the target electronic device are used to execute a target processing task corresponding to the data to be transmitted.

2. According to the method of claim 1, the step of processing the task processing data at least once according to the target processing strategy to obtain the data to be transmitted comprises: Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set; Based on the second reference data corresponding to the classification set, the classification set is packaged according to a target packaging strategy to obtain the data to be transmitted, and the first reference data and the second reference data are the same or different.

3. The method according to claim 1, wherein the step of processing the task processing data at least once according to the target processing strategy to obtain the data to be transmitted comprises: Based on the first reference data corresponding to the task processing data, the task processing data is processed with a target classification strategy to obtain at least one classification set; Based on the third reference data corresponding to the classification set, configuring the classification set with a target configuration strategy to obtain a data set to be packaged; Based on the second reference data corresponding to the classification set, the data set to be packaged is packaged according to a target packaging strategy to obtain the data to be transmitted.

4. The method according to claim 2 or 3, wherein the task processing data is processed using a target classification strategy to obtain a classification set, including at least one of the following: The task processing data includes a processing function and corresponding parameters, and the processing function is classified and processed based on the first reference data of the processing function to obtain a first classification set, so that after the target electronic device receives the data to be transmitted, the processing function running on the hardware resources of the target electronic device is used to execute the target processing task; The task processing data includes a processing function and a processing rule corresponding to the processing function, and the processing function is classified and processed based on the first reference data of the processing function to obtain a second classification set, so as to execute the processing rule after the target electronic device obtains a processing result of the processing function; in, The first reference data includes at least one of the following: The function type of the processing function; The frequency with which the processing function is called during execution; The data priority of the processing function during execution; The association relationship between the processing functions.

5. The method according to claim 2 or 3, wherein the task processing data is processed using a target classification strategy to obtain at least one classification set, comprising: Performing a first classification process on the task processing data based on first sub-reference data in the first reference data to obtain at least one first subset; Performing a second classification process on the first subset based on second sub-reference data in the first reference data to obtain a classification set; The first reference data includes at least one of the following: The function type of the processing function; The frequency with which the processing function is called during execution; The data priority of the processing function during execution; The association relationship between the processing functions; the number of the target electronic devices; The number of processing units of the second electronic device, where the second electronic device is an electronic device that receives the data to be transmitted.

6. The method according to claim 3, wherein the step of packaging the data set to be packaged using a target packaging strategy to obtain the data to be transmitted comprises: Based on the second reference data, the to-be-packed data set is packaged to generate packet header data; Packaging the task processing data to generate package data; Using the data composed of the packet header data and the packet body data as the data to be transmitted; Wherein, the second reference data includes at least one of the following: the number of the target electronic devices; The device identification of the target electronic device; The hash value of the data set to be packaged.

7. A data processing method, applied to a second electronic device, the method comprising: In response to obtaining the data to be transmitted, parsing the data to be transmitted to obtain task processing data; The target processing task corresponding to the task processing data is executed using the hardware resources of the second electronic device to obtain task result data.

8. The method according to claim 7, wherein the step of parsing the data to be transmitted to obtain task processing data comprises: Based on the second reference data, the data to be transmitted is parsed using a target parsing strategy to obtain at least one classification set; Based on the first reference data, the classification set is restored using a target restoration strategy to obtain the task processing data.

9. The method according to claim 8, wherein the step of using the hardware resources of the second electronic device to execute the target processing task corresponding to the task processing data comprises: In the case where there are multiple second electronic devices, based on the third reference data, using hardware resources of the multiple second electronic devices to execute the target processing task; or, In the case that the second electronic device is unique, the target processing task is executed by using multiple computing units of the second electronic device based on the third reference data.

10. A data processing device, comprising: A processing unit, configured to, in response to obtaining task processing data, process the task processing data at least once according to a target processing strategy to obtain data to be transmitted, wherein the task processing data is data obtained by parsing the target processing task; The transmission unit is used to transmit the data to be transmitted to a target electronic device, so as to utilize the hardware resources of the target electronic device to execute a target processing task corresponding to the data to be transmitted.

Images