Distribution network cloud-edge unloading method and system based on end-side data storage perception

Abstract

The present disclosure provides a distribution network cloud-edge unloading method and system based on end-side data storage perception, and the scheme includes: obtaining relevant information of distribution network services; based on the relevant information, classifying the distribution network services according to a pre-established classification model; wherein the classification model classifies the services based on bandwidth requirements, computing resource requirements, and latency requirements of the distribution network services; based on the obtained distribution network service type, and comprehensively considering relevant environmental status indicators of the end side, edge nodes, and cloud servers, determining a server selection index value; when the server selection index value is greater than a server selection index threshold, unloading the data to the cloud server for processing; otherwise, unloading the data to the cloud server for processing.

Description

Technical Field

[0001] The present disclosure belongs to the field of edge computing technology, and in particular, relates to a distribution network cloud-edge unloading method and system based on terminal-side data storage perception. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] With the development of new power systems and the large-scale access of intelligent power distribution and utilization terminals, the types of distribution network services and business data are increasing, and business needs are gradually showing differentiated characteristics.

[0004] Traditional cloud computing offloads all computing tasks of terminal devices to the cloud center for centralized processing, which will cause network congestion and large transmission delays, and cannot meet the differentiated data processing needs of multiple services in the distribution network. Edge computing uses a distributed computing method, in which multiple servers distributed in the network handle computing tasks, reducing the need for devices to upload data to cloud servers and reducing network congestion. At the same time, by deploying servers on the edge side, computing power is decentralized and the transmission time of power business data is reduced. Under the cloud-edge collaborative mechanism, the terminal can offload end-side data to the edge server or cloud server for processing, thereby improving data processing efficiency. However, the inventors found that the traditional offloading method lacks end-side data storage awareness, making it difficult to adapt cloud-edge offloading decisions to the differentiated processing needs of end-side data, resulting in poor cloud-edge offloading performance of power business data. Summary of the invention

[0005] In order to solve the above-mentioned problems, the present disclosure provides a distribution network cloud-edge unloading method and system based on end-side data storage perception. The scheme classifies multiple services according to the bandwidth requirements, computing requirements, latency requirements and other characteristics of different services of the distribution network, and effectively identifies the data type by perceiving the end-side distribution network service data, thereby realizing the adaptation of the distribution network cloud-edge unloading strategy to differentiated service requirements.

[0006] According to a first aspect of an embodiment of the present disclosure, a distribution network cloud edge unloading method based on device-side data storage perception is provided, including:

[0007] Obtain relevant information about distribution network services;

[0008] Based on the relevant information, the distribution network services are classified according to a pre-established classification model; wherein the classification model is based on bandwidth requirements, computing resource requirements, and latency requirements of the distribution network services for classification;

[0009] Based on the obtained distribution network service type and the relevant environmental status indicators of the end side, edge node and cloud server, the server selection index value is determined;

[0010] When the server selection index value is greater than the server selection index threshold, the data is unloaded to the edge server for processing; otherwise, the data is unloaded to the cloud server for processing.

[0011] Furthermore, the server selection index value is obtained by specifically adopting the following formula:

[0012]

[0013] Among them, D i,j is the storage volume of the jth service data at the end side of the i-th time slot, E i,j is the remaining power on the end side of the i-th time slot, B i,j,1 is the bandwidth allocated to the jth service between the ends of the i-th time slot, B i,j,2 is the bandwidth allocated between the end and the cloud for the jth service in the i-th time slot, Q i,j,1 is the backlog of the j-th service data queue at the i-th time slot side, Q i,j,2 is the backlog of the j-th service data queue on the cloud side in the i-th time slot, f i,j,1 The calculation speed of processing the jth service data on the side of the i-th time slot, f i,j,2 is the computing speed of the jth service data processed by the cloud side in the i-th time slot, T i,j,1 is the end-to-edge transmission delay of the jth service data in the ith time slot, T i,j,2 is the end-to-cloud transmission delay of the jth service data in the ith time slot, τ i,j,1 The side delay calculation for the jth service data in the ith time slot is τ i,j,1 is the cloud-side calculation delay of the jth service data in the i-th time slot, α i,j , β i,j , χ i,j ,δ i,j , ε i,j ,φ i,j , They are the weight parameters of data storage capacity, remaining power on the end side, bandwidth, queue backlog, computing speed, transmission delay, and computing delay.

[0014] Furthermore, based on the bandwidth requirements, computing resource requirements and latency index requirements of the distribution network services, the distribution network services are classified into large-flow, high-complexity, low-latency services, large-flow, low-complexity, low-latency services, large-flow, low-complexity, non-real-time services, large-flow, high-complexity, non-real-time services, small-flow, high-complexity, low-latency services, small-flow, low-complexity, low-latency services, small-flow, low-complexity, non-real-time services and small-flow, high-complexity, non-real-time services.

[0015] Furthermore, the classification model divides the distribution network services into categories based on the bandwidth requirements, computing resource requirements and delay demand indicators of the distribution network services. Specifically, the proportion of bandwidth requirements, computing resource requirements and delay indicator requirements in the sum of various demand indicators are calculated respectively, and the classification of different distribution network service types is realized based on the obtained indicator proportions and their corresponding preset thresholds.

[0016] Furthermore, the relevant information of the distribution network service includes the service bandwidth requirement, computing resource requirement and latency requirement of the distribution network.

[0017] Furthermore, the relevant environmental status indicators include end-side data storage, channel conditions between ends and edges, channel conditions between ends and clouds, end-side power, edge-side computing power, edge-side queue backlog, cloud-side computing power, cloud-side queue backlog, transmission delay, and computing delay.

[0018] According to a second aspect of an embodiment of the present disclosure, a distribution network cloud edge unloading system based on terminal-side data storage perception is provided, including:

[0019] A data acquisition unit, which is used to obtain relevant information of the distribution network business;

[0020] A service classification unit, which is used to classify the distribution network services according to a pre-established classification model based on the relevant information; wherein the classification model classifies the services based on bandwidth requirements, computing resource requirements, and latency index requirements of the distribution network services;

[0021] A threshold comparison unit, which is used to determine a server selection index value based on the obtained distribution network service type and the relevant environmental status indicators of the terminal side, edge node and cloud server;

[0022] A cloud edge unloading unit is used to unload data to an edge server for processing when the server selection index value is greater than a server selection index threshold; otherwise, the data is unloaded to a cloud server for processing.

[0023] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored and running on the memory, wherein when the processor executes the program, a distribution network cloud-edge unloading method based on end-side data storage perception is implemented.

[0024] According to a fourth aspect of an embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the distribution network cloud-edge unloading method based on end-side data storage perception is implemented.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] (1) The present disclosure provides a distribution network cloud-edge unloading method and system based on edge data storage perception. The scheme classifies multiple services according to the bandwidth requirements, computing requirements, latency requirements and other characteristics of different services of the distribution network. By perceiving the edge distribution network service data, the scheme effectively identifies the data type and realizes the adaptation of the distribution network cloud-edge unloading strategy to differentiated service requirements.

[0027] (2) The solution disclosed in the present invention comprehensively considers the impact of the end-side business data type and environmental status indicators such as end-side data, end-side channels, and queue backlog on different business data offloading decisions. When unloading data, the server is selected based on the environmental status and the end-side business data type to adapt to the differentiated needs of distribution network services and improve the cloud-edge offloading performance of power business data.

[0028] Advantages of additional aspects of the present disclosure will be given in part in the following description and in part will become apparent from the following description or will be learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

[0030] Figure 1 It is a schematic diagram of the overall architecture of the distribution network cloud edge unloading method based on end-side data storage perception described in the embodiment of the present disclosure;

[0031] Figure 2 This is a flow chart of the distribution network cloud-edge unloading method based on end-side data storage perception described in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0033] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0034] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0035] In the absence of conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other.

[0036] Terminology explanation:

[0037] Terminal side: terminal side, including but not limited to distribution network control terminal, intelligent inspection robot and load control terminal;

[0038] Side: edge computing nodes;

[0039] Cloud side: cloud server.

[0040] Embodiment 1:

[0041] The purpose of this embodiment is to provide a distribution network cloud-edge unloading method based on edge data storage perception.

[0042] A distribution network cloud edge unloading method based on edge data storage perception, comprising:

[0043] Obtain relevant information about distribution network services;

[0044] Based on the relevant information, the distribution network services are classified according to a pre-established classification model; wherein the classification model is based on bandwidth requirements, computing resource requirements, and latency requirements of the distribution network services for classification;

[0045] Based on the obtained distribution network service type and the relevant environmental status indicators of the end side, edge node and cloud server, the server selection index value is determined;

[0046] When the server selection index value is greater than the server selection index threshold, the data is unloaded to the edge server for processing; otherwise, the data is unloaded to the cloud server for processing.

[0047] Furthermore, the server selection index value is obtained by specifically adopting the following formula:

[0048]

[0049] Among them, D i,j is the storage volume of the jth service data at the end side of the i-th time slot, E i,j is the remaining power on the end side of the i-th time slot, B i,j,1 is the bandwidth allocated to the jth service between the ends of the i-th time slot, B i,j,2 is the bandwidth allocated between the end and the cloud for the jth service in the i-th time slot, Q i,j,1 is the backlog of the j-th service data queue at the i-th time slot side, Q i,j,2 is the backlog of the j-th service data queue on the cloud side in the i-th time slot, f i,j,1 The calculation speed of processing the jth service data on the side of the i-th time slot, fi,j,2 is the computing speed of the jth service data processed by the cloud side in the i-th time slot, T i,j,1 is the end-to-edge transmission delay of the jth service data in the ith time slot, T i,j,2 is the end-to-cloud transmission delay of the jth service data in the ith time slot, τ i,j,1 The side delay calculation for the jth service data in the ith time slot is τ i,j,1 is the cloud-side calculation delay of the jth service data in the i-th time slot, α i,j , β i,j , χ i,j ,δ i,j , ε i,j ,φ i,j , They are the weight parameters of data storage capacity, remaining power on the end side, bandwidth, queue backlog, computing speed, transmission delay, and computing delay.

[0050] Furthermore, based on the bandwidth requirements, computing resource requirements and latency index requirements of the distribution network services, the distribution network services are classified into large-flow, high-complexity, low-latency services, large-flow, low-complexity, low-latency services, large-flow, low-complexity, non-real-time services, large-flow, high-complexity, non-real-time services, small-flow, high-complexity, low-latency services, small-flow, low-complexity, low-latency services, small-flow, low-complexity, non-real-time services and small-flow, high-complexity, non-real-time services.

[0051] Furthermore, the classification model divides the distribution network services into categories based on the bandwidth requirements, computing resource requirements and delay demand indicators of the distribution network services. Specifically, the proportion of bandwidth requirements, computing resource requirements and delay indicator requirements in the sum of various demand indicators are calculated respectively, and the classification of different distribution network service types is realized based on the obtained indicator proportions and their corresponding preset thresholds.

[0052] Furthermore, the relevant information of the distribution network service includes the service bandwidth requirement, computing resource requirement and latency requirement of the distribution network.

[0053] Furthermore, the relevant environmental status indicators include end-side data storage, channel conditions between ends and edges, channel conditions between ends and clouds, end-side power, edge-side computing power, edge-side queue backlog, cloud-side computing power, cloud-side queue backlog, transmission delay, and computing delay.

[0054] Specifically, for ease of understanding, the solution described in this embodiment is described in detail below with reference to the accompanying drawings:

[0055] The solution described in this embodiment mainly solves the following problems:

[0056] (1) Existing offloading methods do not classify various types of distribution network services according to bandwidth, computing, latency and other requirements, cannot perceive and identify the data stored in the distribution network on the edge, and cannot adapt the cloud-edge offloading strategy to the differentiated processing requirements of the end-side business data.

[0057] (2) Traditional offloading methods do not consider the impact of end-side business data types and status indicators such as end-side data processing, cloud-edge channels, queue backlogs, and computing backlogs on different business data offloading decisions. When unloading data, it is impossible to optimize server selection based on environmental status and end-side business data types, and it is difficult to adapt to the differentiated needs of distribution network services, resulting in poor cloud-edge offloading performance of power business data.

[0058] In order to solve the above problems, this embodiment proposes a distribution network cloud edge unloading method and system based on terminal side data storage perception, such as Figure 1 As shown in , its main technical concepts include: first, classify the distribution network services according to the different distribution network service bandwidth, computing resources, delay and other requirements, and realize the end-side business data perception and identification by analyzing the bandwidth demand indicators, computing resource demand indicators and delay demand indicators of the end-side data. Then, according to the identified business type, the server selection index is calculated by comprehensively considering the end-side data storage, channel conditions between the end and the edge, channel conditions between the end and the cloud, end-side power, edge-side computing power, edge-side queue backlog, cloud-side computing power, cloud-side queue backlog, end-to-end transmission delay, computing delay and other status indicators; finally, the cloud-edge unloading strategy is determined according to the server selection indicators and thresholds. As shown in Figure 2 As shown, the specific steps are as follows:

[0059] S1: According to the resource requirements of different distribution network services such as bandwidth, computing resources, and latency, distribution network services are divided into the following eight types of services, including large-flow, high-complexity, low-latency services, large-flow, low-complexity, low-latency services, large-flow, low-complexity, non-real-time services, large-flow, high-complexity, non-real-time services, small-flow, high-complexity, low-latency services, small-flow, low-complexity, low-latency services, small-flow, low-complexity, non-real-time services, and small-flow, high-complexity, non-real-time services.

[0060] S2: Perceive and identify the terminal-side data based on the above distribution network service classification.

[0061] S2.1: Normalize the end-side data demand indicators. Comprehensively consider the demand for bandwidth, computing resources, latency and other resources of the distribution network business, and normalize the bandwidth demand indicators, computing resource demand indicators, and latency demand indicators of the end-side data.

[0062] S2.2: End-side data demand index analysis. Score the end-side data based on the end-side data bandwidth demand, computing resource demand, and latency demand normalized indicators, and calculate the proportion of the end-side data bandwidth demand index, computing resource demand index, and latency demand index according to the following formula:

[0063]

[0064] Among them, ω is the normalized index of bandwidth demand, ξ is the normalized index of computing resource demand, ψ is the normalized index of delay demand, and θ 1 ,θ 2 ,θ 3 are their respective proportions.

[0065] S2.3: Based on the terminal data demand indicator ratio obtained in step S2.2, identify the terminal data according to Table 1 to determine the service type to which it belongs.

[0066] Table 1 End-side data service type identification table

[0067]

[0068] Among them, θ in Table 1 1,max ,θ 2,max ,θ 3,max They are respectively the thresholds of the proportions of the preset normalized bandwidth requirement index, the normalized computing resource requirement index, and the normalized latency requirement index.

[0069] S3: Based on the service type determined in step S2, the data cloud-edge offloading strategy is determined by comprehensively perceiving environmental status indicators such as end-side data storage, channel conditions between ends and edges, channel conditions between ends and clouds, end-side power, edge-side computing power, edge-side queue backlog, cloud-side computing power, cloud-side computing backlog, transmission delay, computing delay, and other environmental status indicators and preset thresholds. In order to more clearly reflect the above environmental status indicators, the scheme described in this embodiment quantifies the end-side data storage situation as the amount of data to be uploaded stored in the terminal, quantifies the channel conditions between ends and edges as the amount of bandwidth that can be allocated between the end and edges, quantifies the channel conditions between ends and clouds as the amount of bandwidth that can be allocated between the end and clouds, quantifies the edge-side computing power as the computing speed of edge-side data processing, that is, the time required for the edge server to process 1 bit of data, quantifies the edge-side queue backlog as the amount of unprocessed data stored in the edge server, quantifies the cloud-side computing power as the computing speed of cloud-side data processing, that is, the time required for the cloud server to process 1 bit of data, and quantifies the cloud-side computing backlog as the amount of unprocessed data stored in the cloud server.

[0070] S3.1: For the jth service, consider the end-side data storage, the channel status between the end and the edge, the channel status between the end and the cloud, the end-side power, the edge-side computing power, the edge-side queue backlog, the cloud-side computing power, the cloud-side queue backlog, the transmission delay, the computing delay and other status indicators, and define the server selection index. It is expressed as

[0071]

[0072] Among them, D i,j is the storage volume of the jth service data at the end side of the i-th time slot, E i,j is the remaining power on the end side of the i-th time slot, B i,j,1 is the bandwidth allocated to the jth service between the ends of the i-th time slot, B i,j,2 is the bandwidth allocated between the end and the cloud for the jth service in the i-th time slot, Q i,j,1 is the backlog of the j-th service data queue at the i-th time slot side, Q i,j,2 is the backlog of the j-th service data queue on the cloud side in the i-th time slot, f i,j,1 The calculation speed of processing the jth service data on the side of the i-th time slot, f i,j,2 is the computing speed of the jth service data processed by the cloud side in the i-th time slot, T i,j,1 is the end-to-edge transmission delay of the jth service data in the ith time slot, T i,j,2 is the end-to-cloud transmission delay of the jth service data in the ith time slot, τ i,j,1 The side delay calculation for the jth service data in the ith time slot is τ i,j,1 is the cloud-side calculation delay of the jth service data in the i-th time slot, α i,j , β i,j , χ i,j ,δ i,j , ε i,j ,φ i,j , They are the weight parameters of the preset data storage capacity, remaining power on the terminal side, bandwidth, queue backlog, computing speed, transmission delay, and computing delay;

[0073] Different weight parameters are set according to the business type identified in step S1 to reflect the differentiated needs of different businesses for the above resources. The scheme described in this embodiment compares the edge-side index and the cloud-side index by subtraction, wherein the edge-side status index is proportional to the server selection index, and the cloud-side status index is inversely proportional to the server selection index. When the edge-side bandwidth and computing speed are large, and the latency and queue backlog are small, the algorithm tends to select the edge server, otherwise it tends to select the cloud server. In particular, the server selection index is inversely proportional to the remaining power on the end side. When the remaining power on the end side is small, it tends to select the edge server to reduce the power consumption of terminal data transmission, otherwise it selects the cloud server. The server selection index is also inversely proportional to the amount of data stored on the end side. When the amount of data stored on the end side is large, the communication overhead is large and the transmission time is long. At this time, it tends to select the edge server to reduce the communication overhead, otherwise it selects the cloud server.

[0074] S3.2: Judgment Where X is the server selection index threshold, which is used to determine whether the data is offloaded to the edge server for processing. If it is satisfied, go to step S3.4, otherwise go to step S3.3.

[0075] S3.3: Offload data to the cloud server for processing.

[0076] S3.4: Offload data to edge servers for processing.

[0077] S3.5: Update the end-side data demand indicators and cloud-edge status indicators.

[0078] Embodiment 2:

[0079] The purpose of this embodiment is to provide a distribution network cloud-edge unloading system based on edge data storage perception.

[0080] A distribution network cloud-edge unloading system based on edge data storage perception, comprising:

[0081] A data acquisition unit, which is used to obtain relevant information of the distribution network business;

[0082] A service classification unit, which is used to classify the distribution network services according to a pre-established classification model based on the relevant information; wherein the classification model classifies the services based on bandwidth requirements, computing resource requirements, and latency index requirements of the distribution network services;

[0083] A threshold comparison unit, which is used to determine a server selection index value based on the obtained distribution network service type and the relevant environmental status indicators of the terminal side, edge node and cloud server;

[0084] A cloud edge unloading unit is used to unload data to an edge server for processing when the server selection index value is greater than a server selection index threshold; otherwise, the data is unloaded to a cloud server for processing.

[0085] Furthermore, the system described in this embodiment corresponds to the method described in Example 1, and its technical details are described in detail in Example 1, so they are not repeated here.

[0086] In further embodiments, there is also provided:

[0087] An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method described in Embodiment 1 is performed. For the sake of brevity, no further description is given here.

[0088] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0089] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0090] A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in embodiment 1 is completed.

[0091] The method in the first embodiment can be directly embodied as a hardware processor, or a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0092] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in the present embodiment can be implemented in electronic hardware or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0093] The above-mentioned embodiments provide a distribution network cloud-edge unloading method and system based on end-side data storage perception, which can be implemented and has broad application prospects.

[0094] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A distribution network cloud edge unloading method based on edge data storage perception, characterized in that: include: Obtain relevant information about distribution network services; Based on the relevant information, the distribution network services are classified according to a pre-established classification model; wherein the classification model is based on bandwidth requirements, computing resource requirements, and latency requirements of the distribution network services for classification; Based on the obtained distribution network service type and the relevant environmental status indicators of the end side, edge node and cloud server, the server selection index value is determined; The server selection index value is obtained by using the following formula: in, For the The end side of the time slot The amount of business data storage, For the The remaining power on the end side of each time slot is For the The time slot ends are The bandwidth allocated to each service, For the The time slot between the end and the cloud is The bandwidth allocated to each service, For the Time slot side The business data queue is backlogged. For the Time slot cloud side The business data queue is backlogged. For the Time slot side processing The speed of calculating business data, For the The cloud side processes the The speed of calculating business data, For the Time slot The end-to-edge transmission delay of the service data. For the Time slot The end-to-cloud transmission delay of business data For the Time slot The edge computing latency of the service data is For the Time slot The cloud computing latency of various types of business data, , , , , , , Respectively i Time slot j The data storage capacity, remaining power on the end side, bandwidth, queue backlog, computing speed, transmission delay, and weight parameters of computing delay corresponding to each service; When the server selection index value is greater than the server selection index threshold, the data is unloaded to the edge server for processing; otherwise, the data is unloaded to the cloud server for processing.

2. A distribution network cloud edge unloading method based on end-side data storage perception as claimed in claim 1, characterized in that: The distribution network services are classified based on their bandwidth requirements, computing resource requirements and latency index requirements, and are divided into large-flow, high-complexity, low-latency services, large-flow, low-complexity, low-latency services, large-flow, low-complexity, non-real-time services, large-flow, high-complexity, non-real-time services, small-flow, high-complexity, low-latency services, small-flow, low-complexity, low-latency services, small-flow, low-complexity, non-real-time services and small-flow, high-complexity, non-real-time services.

3. A distribution network cloud edge unloading method based on end-side data storage perception as claimed in claim 1, characterized in that: The classification model is based on the bandwidth demand, computing resource demand and delay demand indicators of the distribution network business. Specifically, the proportion of bandwidth demand, computing resource demand and delay indicator demand in the sum of various demand indicators is calculated respectively, and the classification of different distribution network business types is realized based on the obtained indicator proportions and their corresponding preset thresholds.

4. A distribution network cloud edge unloading method based on end-side data storage perception as claimed in claim 1, characterized in that: The relevant information of the distribution network service includes the service bandwidth requirement, computing resource requirement and latency requirement of the distribution network.

5. A distribution network cloud edge unloading method based on end-side data storage perception as claimed in claim 1, characterized in that: The relevant environmental status indicators include end-side data storage, channel conditions between ends and edges, channel conditions between ends and clouds, end-side power, edge-side computing power, edge-side queue backlog, cloud-side computing power, cloud-side queue backlog, transmission delay, and computing delay.

6. A distribution network cloud edge unloading system based on edge data storage perception, characterized in that: include: A data acquisition unit, which is used to obtain relevant information of the distribution network business; A service classification unit, which is used to classify the distribution network services according to a pre-established classification model based on the relevant information; wherein the classification model classifies the services based on bandwidth requirements, computing resource requirements, and latency index requirements of the distribution network services; A threshold comparison unit, which is used to determine a server selection index value based on the obtained distribution network service type and the relevant environmental status indicators of the terminal side, edge node and cloud server; The server selection index value is obtained by using the following formula: in, For the The time slot end side The amount of business data storage, For the The remaining power on the end side of each time slot is For the The time slot ends are The bandwidth allocated to each service, For the The time slot between the end and the cloud is The bandwidth allocated to each service, For the Time slot side The business data queue is backlogged. For the Time slot cloud side The business data queue is backlogged. For the Time slot side processing The speed of calculating business data, For the The cloud side processes the The speed of calculating business data, For the Time slot The end-to-edge transmission delay of the service data. For the Time slot The end-to-cloud transmission delay of business data For the Time slot The edge computing latency of the service data is For the Time slot The cloud computing latency of various types of business data, , , , , , , Respectively i Time slot j The data storage capacity, remaining power on the end side, bandwidth, queue backlog, computing speed, transmission delay, and weight parameters of computing delay corresponding to each service; A cloud edge unloading unit is used to unload data to an edge server for processing when the server selection index value is greater than a server selection index threshold; otherwise, the data is unloaded to a cloud server for processing.

7. An electronic device comprising a memory, a processor and a computer program stored and running on the memory, characterized in that: When the processor executes the program, it implements a distribution network cloud-edge unloading method based on end-side data storage perception as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, it implements a distribution network cloud-edge unloading method based on end-side data storage perception as described in any one of claims 1 to 5.

Images