Acceleration processing method, system, electronic device and medium based on progressive transmission

Abstract

The acceleration processing method, system, electronic device and medium based on progressive transmission provided by the present invention include: the client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compression instruction; the client transmits the compressed data of the preset data transmission ratio and the compression instruction to the operation end; the operation end determines the target acceleration algorithm according to the compression instruction; the operation end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client; if the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete progressive transmission. The present invention saves transmission time by transmitting compressed data of the preset data transmission ratio, further selects an acceleration algorithm for the data transmitted to the operation end, performs accelerated analysis processing, saves analysis time, and improves the efficiency of data transmission and analysis.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an accelerated processing method, system, electronic device and medium based on progressive transmission. Background Art

[0002] When conducting various data analysis tasks, if faced with huge computing needs, huge computing power is usually required to meet the needs of data processing. For example, the analysis work can be ultrasonic imaging, electrocardiogram, infrasound, emotion prediction, etc. When analyzing ultrasonic imaging, electrocardiogram, infrasound, or spectrograms for emotion prediction, a large amount of computing power is required, which usually comes from large computing servers, supercomputing centers, or computing servers independently equipped with hardware resources.

[0003] Currently, data analysis is mainly based on hardware acceleration, and the data transmission time cannot be optimized, resulting in long data transmission and analysis time and low efficiency.

[0004] Therefore, the prior art has defects and needs to be improved and developed. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an accelerated processing method, system, electronic device and medium based on progressive transmission in view of the above-mentioned defects of the prior art, aiming to solve the problem of long time consumption and low efficiency in data transmission and analysis in the prior art.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0007] A method for accelerating processing based on progressive transmission, comprising:

[0008] The client compresses the original data to be transmitted and the corresponding data analysis instructions respectively to obtain compressed data and compressed instructions;

[0009] The client transmits the compressed data of the preset data transmission ratio and the compression instruction to the computing end;

[0010] The computing end determines a target acceleration algorithm according to the compression instruction;

[0011] The computing end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client;

[0012] If the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission.

[0013] In one embodiment of the present application, the client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compressed instructions, including:

[0014] If the information to be transmitted of the client includes the original data to be transmitted and the corresponding data analysis instruction, the client determines the target data type of the original data to be transmitted;

[0015] The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type;

[0016] The client compresses the transmission original data according to the target compression mode to obtain compressed data, and compresses the data analysis instruction to obtain a compression instruction;

[0017] The original data to be transmitted is pictures, videos or audios.

[0018] In one embodiment of the present application, the computing end determines a target acceleration algorithm according to the compression instruction, including:

[0019] The computing end decompresses the compressed instruction to obtain a data analysis code, and performs dependency analysis on the data analysis code to obtain a dependency relationship;

[0020] The computing end determines a target acceleration algorithm according to the dependency relationship.

[0021] In one embodiment of the present application, the computing end determines a target acceleration algorithm according to the dependency relationship, including:

[0022] If the dependency is an instruction-level dependency, the operation end determines the target instruction-level dependency type of the data analysis code, searches a preset acceleration algorithm library, and obtains a target acceleration algorithm corresponding to the target instruction-level dependency type, wherein the acceleration algorithm library includes: a correspondence between the instruction-level dependency type and the acceleration algorithm;

[0023] If the dependency is a loop-level dependency, the operation end determines the execution mode in the data analysis code. If the execution mode in the data analysis code includes serial execution, the operation end adjusts the serially executed data analysis code to a parallel executed data analysis code to execute a parallel acceleration algorithm.

[0024] In an embodiment of the present application, if the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission, including:

[0025] If the analysis fails, the client adjusts the preset data transmission ratio according to the preset increase ratio to obtain an adjusted data transmission ratio;

[0026] The client transmits the compressed data with the adjusted data transmission ratio to the computing end, and the computing end performs acceleration analysis on the compressed data with the adjusted data transmission ratio according to the target acceleration algorithm until a successful acceleration analysis result is obtained, thereby completing the progressive transmission.

[0027] In one embodiment of the present application, the accelerated processing method based on progressive transmission further includes:

[0028] If the information to be transmitted of the client includes the original data to be transmitted, the corresponding data analysis instruction and the specified acceleration analysis algorithm, the client determines the target data type of the original data to be transmitted;

[0029] The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type;

[0030] The client compresses the transmitted original data according to the target compression method to obtain compressed data, and compresses the data analysis instruction and the specified acceleration analysis algorithm to obtain compression instruction parameters;

[0031] The client transmits the compressed data of the preset data transmission ratio and the compression instruction parameters to the computing end;

[0032] The computing end decompresses the compression instruction parameters to obtain a data analysis code and a target acceleration algorithm, and performs acceleration analysis on the compressed data of a preset data transmission ratio according to the target acceleration algorithm and the data analysis code, and feeds back the acceleration analysis results to the client.

[0033] In one embodiment of the present application, the accelerated processing method based on progressive transmission includes:

[0034] If the information to be transmitted of the client only contains original data to be transmitted, the client determines the target data type of the original data to be transmitted;

[0035] The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type;

[0036] The client compresses the transmission original data according to the target compression method to obtain compressed data, and transmits the compressed data to the computing end.

[0037] The present application also provides an accelerated processing system based on progressive transmission, which includes:

[0038] The client is used to compress the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compressed instructions, and transmit the compressed data with a preset data transmission ratio and the compressed instructions to the computing end;

[0039] A computing end, for determining a target acceleration algorithm according to the compression instruction, performing acceleration analysis on the compressed data of a preset data transmission ratio according to the target acceleration algorithm, and feeding back the acceleration analysis result to the client;

[0040] The client is also used to adjust the preset data transmission ratio according to the accelerated analysis result to complete progressive transmission if the analysis fails.

[0041] The present application also provides an electronic device, which includes a data transmission processing module, and the data includes but is not limited to one or more of ultrasonic imaging, electrocardiogram, infrasound, and spectrogram for emotion prediction. The electronic device implements the accelerated processing method based on progressive transmission as described above through the data transmission processing module.

[0042] The present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program can be executed to implement the steps of the accelerated processing method based on progressive transmission as described above.

[0043] The present invention provides an acceleration processing method, system, electronic device and medium based on progressive transmission, the method comprising: the client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compression instruction; the client transmits the compressed data of the preset data transmission ratio and the compression instruction to the operation end; the operation end determines the target acceleration algorithm according to the compression instruction; the operation end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client; if the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete progressive transmission. The present invention saves transmission time by transmitting compressed data of the preset data transmission ratio, further selects an acceleration algorithm for the data transmitted to the operation end, and then performs accelerated analysis processing, which saves analysis time and improves the efficiency of data transmission and analysis; and the client can dynamically adjust the preset increase ratio or maximum transmission ratio according to the success rate of historical transmission and analysis to optimize transmission efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a preferred embodiment of the accelerated processing method based on progressive transmission in the present invention.

[0045] Figure 2It is a specific flow chart of step S100 in a preferred embodiment of the accelerated processing method based on progressive transmission in the present invention.

[0046] Figure 3 It is a specific flow chart of step S300 in a preferred embodiment of the accelerated processing method based on progressive transmission in the present invention.

[0047] Figure 4 It is a comparison chart between the average operation time used by two algorithmically different analysis methods in a preferred embodiment of the accelerated processing method based on progressive transmission in the present invention.

[0048] Figure 5 It is a functional principle block diagram of a preferred embodiment of the accelerated processing system based on progressive transmission in the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] Since the current data acceleration algorithm is mainly based on hardware acceleration, and is targeted at a single type of hardware, and is mostly used in a single terminal scenario, it is impossible to optimize data transmission time and optimize the collaboration mode between terminals. Therefore, in addition to utilizing the computing power of high-computing platforms, how to effectively shorten data transmission time and optimize the collaboration mode between terminals is also a key technical challenge at present.

[0051] The following describes the acceleration processing method, system, electronic device and medium based on progressive transmission of the embodiment of the present application with reference to the accompanying drawings. In view of the problem that the data is analyzed mainly by hardware acceleration in the related technology mentioned in the above background technology, and the data transmission time cannot be optimized, resulting in a long time consumption and low efficiency in data transmission and analysis, the present application provides an acceleration processing method based on progressive transmission, in which the client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compression instructions; the client transmits the compressed data of the preset data transmission ratio and the compression instruction to the operation end; the operation end determines the target acceleration algorithm according to the compression instruction; the operation end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client. The present invention saves transmission time by transmitting compressed data of the preset data transmission ratio, further selects an acceleration algorithm for the data transmitted to the operation end, and then performs accelerated analysis processing, which saves analysis time and improves the efficiency of data transmission and analysis.

[0052] Specifically, see Figure 1 , Figure 1 FIG. 1 is a flow chart of the progressive transmission acceleration method of the present invention. Figure 1 As shown, the accelerated processing method based on progressive transmission described in the embodiment of the present invention includes:

[0053] Step S100: The client compresses the original data to be transmitted and the corresponding data analysis instructions respectively to obtain compressed data and compression instructions.

[0054] The client is an interface or device that the user interacts with directly. It is responsible for initiating data transmission requests, processing user input, and displaying the results returned by the computing end. It can be a mobile device, personal computer, embedded device, Web browser, etc. Mobile devices such as smart phones and tablets have network connection capabilities and can act as clients to send data to the remote computing end and receive processing results; personal computers, including desktops and notebooks, have strong computing power and large storage space, and can act as clients to run complex applications and send large amounts of data or requests to the computing end through the network; embedded devices such as smart home devices and car systems usually have specific functions and limitations, but they can also connect to the computing end through the network to achieve remote data processing and analysis; in Web applications, the user's Web browser can act as a client to communicate with the server (i.e., the computing end) through protocols such as HTTP to achieve data transmission and processing.

[0055] In the embodiments of this application, please refer to Figure 2 , the step S100 specifically includes:

[0056] Step S110: If the information to be transmitted of the client includes the original data to be transmitted and the corresponding data analysis instruction, the client determines the target data type of the original data to be transmitted;

[0057] Step S120: The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type;

[0058] Step S130: The client compresses the original data to be transmitted according to the target compression method to obtain compressed data, and compresses the data analysis instruction to obtain a compression instruction; wherein the original data to be transmitted is a picture, video or audio.

[0059] Specifically, the embodiments of the present application perform different processing based on the needs of the user. If the information to be transmitted of the client includes the original data to be transmitted and the corresponding data analysis instructions, for example, the original data to be transmitted is a picture, and the corresponding data analysis instructions can be to identify whether there is a person in the picture. Data compression can be performed based on the extension of the original data to be transmitted, the large model multimodal mechanism, and the feature value extraction mechanism. For example, the picture is compressed and features are extracted by the image resampling method; the video file is compressed in the H.264 format, and the audio file is compressed using, for example, the modified discrete cosine (MDCT) method.

[0060] The embodiment of the present application selects a suitable data compression and feature extraction method according to the original data to be transmitted, so as to optimize the overall transmission time.

[0061] like Figure 1 As shown, the accelerated processing method based on progressive transmission described in this embodiment also includes:

[0062] Step S200: The client transmits the compressed data of the preset data transmission ratio and the compression instruction to the computing end.

[0063] Specifically, before starting the transmission, the client will set an initial data transmission ratio, which can be the result of comprehensive consideration of multiple factors such as the total amount of data, network bandwidth, computing end processing power, etc. For example, if the total amount of data is large and the network bandwidth is limited, a smaller initial ratio can be set for trial transmission.

[0064] The embodiment of the present application pre-sets an initial data transmission ratio, for example, 50%, and the client first transmits 50% of the compressed data to the computing end. For example, if the compressed data is a picture, 50% of the feature points in the picture are first transmitted to the computing end for analysis. If the compression instruction is to identify whether there is a person in the picture, then 50% of the feature points are sufficient to analyze the result.

[0065] like Figure 1 As shown, the accelerated processing method based on progressive transmission described in this embodiment also includes:

[0066] Step S300: The computing end determines a target acceleration algorithm according to the compression instruction.

[0067] The computing end is a device or system responsible for processing the data sent by the client, executing data analysis instructions, and returning the processing results. It usually has powerful computing and data processing capabilities. The computing end can be a cloud server, data center, edge computing device, high-performance computing cluster, etc. The cloud server is located in a remote data center and has a high-performance processor, large-capacity storage device and high-speed network connection. The client can transmit data to the cloud server through the Internet for processing and receive the processing results. Multiple servers and storage devices are deployed in the data center. These servers can serve as computing ends to provide data processing services for users. In some scenarios, in order to reduce network latency and improve processing efficiency, the computing end may be deployed at the edge of the network (such as base stations, routers, etc.). These devices can process data from nearby clients in real time and quickly return processing results. For tasks that require large-scale parallel computing, the computing end can be a cluster consisting of multiple high-performance computers, which are connected together through a high-speed network to work together to complete complex computing tasks.

[0068] When performing acceleration analysis, the embodiment of the present application will intelligently determine the required acceleration algorithm to facilitate algorithm acceleration.

[0069] In the embodiments of this application, please refer to Figure 3 , the step S300 specifically includes:

[0070] Step S310: the computing end decompresses the compressed instruction to obtain a data analysis code, and performs dependency analysis on the data analysis code to obtain a dependency relationship;

[0071] Step S320: The computing end determines a target acceleration algorithm according to the dependency relationship.

[0072] Specifically, not all data analysis requires an accelerated algorithm, but requires judgment. Dependency analysis can analyze such dependencies in the code to better understand the code behavior.

[0073] The embodiment of the present application performs dependency analysis on the data analysis code and then determines and obtains the target acceleration algorithm.

[0074] In one embodiment of the present application, step S320 includes: if the dependency is an instruction-level dependency, the operation end determines the target instruction-level dependency type of the data analysis code, searches a preset acceleration algorithm library, and obtains a target acceleration algorithm corresponding to the target instruction-level dependency type, wherein the acceleration algorithm library includes: the correspondence between the instruction-level dependency type and the acceleration algorithm.

[0075] Specifically, a dependency analysis mechanism is added when intelligently determining the algorithm, and the dependency analysis mechanism is used to detect whether the algorithm can be executed in parallel, or to optimize the code that cannot be executed in parallel so that it can be executed in parallel.

[0076] Dependency means that program codes must be executed in a certain order. If they are not executed in this order, different running results will be produced. In this case, there is a dependency between these codes. Dependency analysis refers to analyzing such dependencies in the code to better understand the code behavior. Dependency analysis can find out which parts of the program code must be executed serially and which parts can be executed in parallel.

[0077] Dependencies can be divided into two levels: instruction-level dependencies and loop-level dependencies. Instruction-level dependencies include structural dependencies, data dependencies, and control dependencies. Loop-level dependencies include intra-loop dependencies and inter-loop dependencies. Among them, instruction-level dependency types include structural dependencies, data dependencies, and control dependencies. Different instruction-level dependency types correspond to different acceleration algorithms.

[0078] The algorithm library is constructed based on hardware and software. For example, a dedicated algorithm library is constructed by determining matching algorithms based on hardware resources of OpenCL (Open Computing Language) or CUDA (Compute Unified Device Architecture). The hardware resources may be GPGPU (General-purpose computing on graphics processing units), GPU (Graphics Processing Unit) or CPU (Central Processing Unit). Both OpenCL and CUDA are frameworks for parallel programming, allowing developers to use GPUs and other types of processors for high-performance computing.

[0079] The algorithm library may be a clFFT library, which is an FFT (Fast Fourier Transform) library implemented using the OpenCL framework and can perform efficient FFT calculations using parallel processing units such as GPUs.

[0080] When determining that the dependency is an instruction-level dependency, the embodiment of the present application obtains a suitable acceleration algorithm by searching an acceleration algorithm library, thereby achieving algorithm acceleration.

[0081] In one embodiment of the present application, step S320 also includes: if the dependency is a loop-level dependency, the operation end determines the execution method in the data analysis code; if the execution method in the data analysis code includes serial execution, the operation end adjusts the serially executed data analysis code to a parallel executed data analysis code to execute a parallel acceleration algorithm.

[0082] Specifically, for loop-level dependencies, first determine whether there is serially executed code in the data analysis code. If so, adjust the serially executed data analysis code to parallel executed data analysis code. If there is no serially executed code, no acceleration algorithm is needed.

[0083] In addition to invoking the hardware resources of the device itself, the embodiment of the present application can also accelerate the execution mode from serial to parallel, further improving the efficiency of data analysis.

[0084] like Figure 1 As shown, the accelerated processing method based on progressive transmission described in this embodiment also includes:

[0085] Step S400: The computing end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client.

[0086] Specifically, if the analysis is successful, the client saves the accelerated analysis result and completes the analysis of the original data to be transmitted.

[0087] like Figure 1 As shown, the accelerated processing method based on progressive transmission described in this embodiment also includes:

[0088] Step S500: If the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission.

[0089] In an embodiment of the present application, step S500 specifically includes: if the analysis fails, the client adjusts the preset data transmission ratio according to the preset increase ratio to obtain the adjusted data transmission ratio; the client transmits the compressed data of the adjusted data transmission ratio to the computing end, and the computing end performs accelerated analysis on the compressed data of the adjusted data transmission ratio according to the target acceleration algorithm until an accelerated analysis result indicating a successful analysis is obtained, thereby completing the progressive transmission.

[0090] Specifically, the value range of the preset increase ratio is 5% to 20%. Since part of the data is transmitted, the result obtained may be successful or failed. For example, 50% of the feature points in the picture are transmitted to the operation end for analysis. If the data analysis instruction is to identify whether there is a person in the picture, then 50% of the feature points are sufficient to analyze the result, and the accelerated analysis result is a successful analysis. However, if the data analysis instruction is to identify who is in the picture, then 50% of the feature points are not enough to analyze the result, so the accelerated analysis result at this time is an analysis failure. When the client receives feedback that the analysis failed, the preset data transmission ratio will be adjusted according to the preset increase ratio. For example, if the preset increase ratio is 10%, then the data transmission ratio of the second transmission is 60%, and the accelerated analysis is continued according to 60% of the data. That is to say, when the data is transmitted, if the analysis fails, the data is transmitted in the ratio of 50%, 60%, 70%, 80%..., so the embodiment of the present application realizes progressive transmission.

[0091] That is to say, before starting the transmission, the client needs to preset the initial data transmission ratio (such as 20%), the preset increase ratio (such as 20%), the maximum transmission ratio (such as 100% or set according to actual conditions), and a preliminary assessment of different data analysis instructions, so as to clarify which instructions may only require a small amount of data for successful analysis and which require more data.

[0092] In addition, the client can dynamically adjust the preset increase ratio or maximum transmission ratio based on the historical transmission and analysis success rate to optimize transmission efficiency and accuracy. Before sending data, the client can evaluate the data analysis instruction, predict the amount of data it may need, and determine the initial transmission ratio and increase ratio based on the predicted amount of data. For example, for instructions that require more data to successfully analyze (such as specific person recognition in face recognition), a higher initial transmission ratio or a smaller increase ratio can be preset.

[0093] In one embodiment of the present application, the acceleration processing method based on progressive transmission also includes: if the client's information to be transmitted includes the original data to be transmitted, the corresponding data analysis instructions and the specified acceleration analysis algorithm, the client determines the target data type of the original data to be transmitted; the client searches for the first correspondence between the preset data type and the compression method, and obtains the target compression method corresponding to the target data type; the client compresses the transmitted original data according to the target compression method to obtain compressed data, compresses the data analysis instructions and the specified acceleration analysis algorithm to obtain compression instruction parameters; the client transmits the compressed data of the preset data transmission ratio and the compression instruction parameters to the computing end; the computing end decompresses the compression instruction parameters to obtain the data analysis code and the target acceleration algorithm, and performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm and the data analysis code, and feeds back the acceleration analysis result to the client.

[0094] Specifically, in the embodiment of the present application, the user can also specify the acceleration analysis algorithm by himself, so that the computing end uses the specified acceleration analysis algorithm to perform acceleration analysis, thereby improving the customization performance of the present method.

[0095] In an embodiment of the present application, the accelerated processing method based on progressive transmission includes: if there is only original data to be transmitted in the information to be transmitted of the client, the client determines the target data type of the original data to be transmitted; the client searches for a first correspondence between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type; the client compresses the transmitted original data according to the target compression method to obtain compressed data, and transmits the compressed data to the computing end.

[0096] In an embodiment of the present application, the user can also transmit only the original data to be transmitted, without transmitting the data analysis instructions or the specified acceleration analysis algorithm. At this time, it is only necessary to use a suitable compression algorithm to compress and transmit all compressed data to the computing end for storage.

[0097] The accelerated processing method based on progressive transmission provided by the embodiment of the present application can save computing time and improve computing efficiency. For example, in the convolution algorithm and FFT (Fast Fourier Transform) algorithm, when the method of the present application is combined with different hardware operations, the average computing time used is shorter, such as Figure 4 shown.

[0098] For different algorithm flows, the average operation time used by the method of the present application in combination with different hardware is shown in Table 1:

[0099] Table 1

[0100]

[0101] In a specific application, electrocardiogram (ECG) can diagnose various heart diseases by analyzing ECG signals. The method provided in the present application is used to analyze ECG signals, and the time data of the ECG analysis steps before and after acceleration are shown in Table 2.

[0102] Table 2

[0103]

[0104] In one embodiment, if Figure 5 As shown, based on the above-mentioned accelerated processing method based on progressive transmission, the present invention also provides an accelerated processing system based on progressive transmission, including:

[0105] The client 10 is used to compress the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compressed instructions, and transmit the compressed data with a preset data transmission ratio and the compressed instructions to the computing end;

[0106] The computing end 20 is used to determine a target acceleration algorithm according to the compression instruction, perform acceleration analysis on the compressed data of a preset data transmission ratio according to the target acceleration algorithm, and feed back the acceleration analysis result to the client;

[0107] The client 10 is also used to adjust the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission if the analysis fails.

[0108] In one embodiment, the present application also provides a progressive transmission method, which includes:

[0109] The original data to be transmitted and the corresponding data analysis instructions are respectively compressed to obtain compressed data and compressed instructions;

[0110] Transmitting the compressed data of the preset data transmission ratio and the compression instruction to the computing end, so that the computing end performs analysis to obtain an accelerated analysis result;

[0111] If the analysis at the computing end fails, the preset data transmission ratio is adjusted according to the accelerated analysis result to complete the progressive transmission.

[0112] In one embodiment, the original data to be transmitted and the corresponding data analysis instructions are compressed respectively to obtain compressed data and compressed instructions, including:

[0113] Acquire information to be transmitted, and if the information to be transmitted includes original data to be transmitted and corresponding data analysis instructions, determine the target data type of the original data to be transmitted;

[0114] Finding a first corresponding relationship between a preset data type and a compression method to obtain a target compression method corresponding to the target data type;

[0115] Compressing the transmission original data according to the target compression mode to obtain compressed data, and compressing the data analysis instruction to obtain a compression instruction;

[0116] The original data to be transmitted is pictures, videos or audios.

[0117] In one embodiment, if the operation end analysis fails, the preset data transmission ratio is adjusted according to the acceleration analysis result to complete the progressive transmission, including:

[0118] If the analysis fails, the preset data transmission ratio is adjusted according to the preset increase ratio to obtain an adjusted data transmission ratio;

[0119] The compressed data with the adjusted data transmission ratio is transmitted to the computing end, so that the computing end performs accelerated analysis on the compressed data with the adjusted data transmission ratio until a successful accelerated analysis result is obtained, thereby completing the progressive transmission.

[0120] In one embodiment, the progressive transmission method further includes:

[0121] If the information to be transmitted includes the original data to be transmitted, the corresponding data analysis instruction and the specified accelerated analysis algorithm, determining the target data type of the original data to be transmitted;

[0122] Finding a first corresponding relationship between a preset data type and a compression method to obtain a target compression method corresponding to the target data type;

[0123] Compressing the transmitted original data according to the target compression method to obtain compressed data, compressing the data analysis instruction and the specified acceleration analysis algorithm to obtain compression instruction parameters;

[0124] The compressed data of the preset data transmission ratio and the compression instruction parameters are transmitted to the computing end, so that the computing end decompresses the compression instruction parameters to obtain the data analysis code and the target acceleration algorithm, and performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm and the data analysis code.

[0125] In one embodiment, the progressive transmission method further includes:

[0126] If the information to be transmitted only contains original data to be transmitted, determining the target data type of the original data to be transmitted;

[0127] Finding a first corresponding relationship between a preset data type and a compression method to obtain a target compression method corresponding to the target data type;

[0128] The transmission original data is compressed according to the target compression method to obtain compressed data, and the compressed data is transmitted to the operation end.

[0129] In one embodiment, the present application also provides an electronic device, including a data transmission processing module, the data including but not limited to one or more of ultrasonic imaging, electrocardiogram, infrasound, and spectrogram for emotion prediction, the electronic device implements the accelerated processing method based on progressive transmission as described above through the data transmission processing module.

[0130] This embodiment further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program can be executed to implement the steps of the progressive transmission method as described above.

[0131] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0132] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0133] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes additional implementations, in which the order shown or discussed may not be followed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can read instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or N wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary and then storing it in a computer memory.

[0135] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0136] A person of ordinary skill in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0137] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0138] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

[0139] In summary, the present invention discloses an accelerated processing method, system, electronic device and medium based on progressive transmission, the method comprising: the client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compression instruction; the client transmits the compressed data of the preset data transmission ratio and the compression instruction to the computing end; the computing end determines the target acceleration algorithm according to the compression instruction; the computing end performs accelerated analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the accelerated analysis result to the client; if the analysis fails, the client adjusts the preset data transmission ratio according to the accelerated analysis result to complete progressive transmission. The present invention saves transmission time by transmitting compressed data of the preset data transmission ratio, further selects an acceleration algorithm for the data transmitted to the computing end, and then performs accelerated analysis processing, which saves analysis time and improves the efficiency of data transmission and analysis.

[0140] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. An accelerated processing method based on progressive transmission, characterized in that: include: The client compresses the original data to be transmitted and the corresponding data analysis instructions respectively to obtain compressed data and compressed instructions; The client transmits the compressed data of the preset data transmission ratio and the compression instruction to the computing end; The computing end determines a target acceleration algorithm according to the compression instruction; The computing end performs acceleration analysis on the compressed data of the preset data transmission ratio according to the target acceleration algorithm, and feeds back the acceleration analysis result to the client; If the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission; The computing end determines a target acceleration algorithm according to the compression instruction, including: The computing end decompresses the compressed instruction to obtain a data analysis code, and performs dependency analysis on the data analysis code to obtain a dependency relationship; The computing end determines a target acceleration algorithm according to the dependency relationship; The computing end determines a target acceleration algorithm according to the dependency, including: If the dependency is an instruction-level dependency, the operation end determines the target instruction-level dependency type of the data analysis code, searches a preset acceleration algorithm library, and obtains a target acceleration algorithm corresponding to the target instruction-level dependency type, wherein the acceleration algorithm library includes: a correspondence between the instruction-level dependency type and the acceleration algorithm; If the dependency is a loop-level dependency, the computing end determines the execution mode in the data analysis code. If the execution mode in the data analysis code includes serial execution, the computing end adjusts the serially executed data analysis code to a parallel executed data analysis code to execute a parallel acceleration algorithm. If the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission, including: If the analysis fails, the client adjusts the preset data transmission ratio according to the preset increase ratio to obtain an adjusted data transmission ratio; The client transmits the compressed data with the adjusted data transmission ratio to the computing end, and the computing end performs acceleration analysis on the compressed data with the adjusted data transmission ratio according to the target acceleration algorithm until a successful acceleration analysis result is obtained, thereby completing the progressive transmission.

2. The accelerated processing method based on progressive transmission according to claim 1, characterized in that: The client compresses the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compressed instructions, including: If the information to be transmitted of the client includes the original data to be transmitted and the corresponding data analysis instruction, the client determines the target data type of the original data to be transmitted; The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type; The client compresses the transmission original data according to the target compression mode to obtain compressed data, and compresses the data analysis instruction to obtain a compression instruction; The original data to be transmitted is pictures, videos or audios.

3. The accelerated processing method based on progressive transmission according to claim 1, characterized in that: The accelerated processing method based on progressive transmission also includes: If the information to be transmitted of the client includes the original data to be transmitted, the corresponding data analysis instruction and the specified acceleration analysis algorithm, the client determines the target data type of the original data to be transmitted; The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type; The client compresses the transmitted original data according to the target compression method to obtain compressed data, and compresses the data analysis instruction and the specified acceleration analysis algorithm to obtain compression instruction parameters; The client transmits the compressed data of the preset data transmission ratio and the compression instruction parameters to the computing end; The computing end decompresses the compression instruction parameters to obtain a data analysis code and a target acceleration algorithm, and performs acceleration analysis on the compressed data of a preset data transmission ratio according to the target acceleration algorithm and the data analysis code, and feeds back the acceleration analysis results to the client.

4. The accelerated processing method based on progressive transmission according to claim 1, characterized in that: The accelerated processing method based on progressive transmission includes: If the information to be transmitted of the client only contains original data to be transmitted, the client determines the target data type of the original data to be transmitted; The client searches for a first corresponding relationship between a preset data type and a compression method, and obtains a target compression method corresponding to the target data type; The client compresses the transmission original data according to the target compression method to obtain compressed data, and transmits the compressed data to the computing end.

5. An accelerated processing system based on progressive transmission, characterized in that: include: The client is used to compress the original data to be transmitted and the corresponding data analysis instruction respectively to obtain compressed data and compressed instructions, and transmit the compressed data with a preset data transmission ratio and the compressed instructions to the computing end; A computing end, for determining a target acceleration algorithm according to the compression instruction, performing acceleration analysis on the compressed data of a preset data transmission ratio according to the target acceleration algorithm, and feeding back the acceleration analysis result to the client; The client is further configured to adjust the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission if the analysis fails; The computing end determines a target acceleration algorithm according to the compression instruction, including: The computing end decompresses the compressed instruction to obtain a data analysis code, and performs dependency analysis on the data analysis code to obtain a dependency relationship; The computing end determines a target acceleration algorithm according to the dependency relationship; The computing end determines a target acceleration algorithm according to the dependency, including: If the dependency is an instruction-level dependency, the operation end determines the target instruction-level dependency type of the data analysis code, searches a preset acceleration algorithm library, and obtains a target acceleration algorithm corresponding to the target instruction-level dependency type, wherein the acceleration algorithm library includes: a correspondence between the instruction-level dependency type and the acceleration algorithm; If the dependency is a loop-level dependency, the computing end determines the execution mode in the data analysis code. If the execution mode in the data analysis code includes serial execution, the computing end adjusts the serially executed data analysis code to a parallel executed data analysis code to execute a parallel acceleration algorithm. If the analysis fails, the client adjusts the preset data transmission ratio according to the acceleration analysis result to complete the progressive transmission, including: If the analysis fails, the client adjusts the preset data transmission ratio according to the preset increase ratio to obtain an adjusted data transmission ratio; The client transmits the compressed data with the adjusted data transmission ratio to the computing end, and the computing end performs acceleration analysis on the compressed data with the adjusted data transmission ratio according to the target acceleration algorithm until a successful acceleration analysis result is obtained, thereby completing the progressive transmission.

6. An electronic device, characterized in that: It includes a data transmission processing module, and the electronic device implements the accelerated processing method based on progressive transmission as described in any one of claims 1 to 4 through the data transmission processing module.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program can be executed to implement the steps of the accelerated processing method based on progressive transmission as described in any one of claims 1 to 4.

Images