An optimization method for distributed edge computing networks

Abstract

The present invention discloses an optimization method for a distributed edge computing network, which relates to the field of network transmission optimization technology, and solves the technical problem that there is often a lack of comprehensive perception of the network environment and targeted optimization strategies, making it difficult to meet the strict requirements of distributed edge computing networks for efficient and stable data transmission. The present invention accurately classifies transmission paths into stable transmission paths and unstable transmission paths, and performs data transmission planning for different types of paths respectively. For stable transmission paths, a reasonable transmission strategy is formulated based on the comparison result between the data capacity of high-demand data and the transmission capacity per unit time of the path, so as to achieve efficient transmission of high-demand data; for unstable transmission paths, edge node storage or optimized transmission is performed based on the access frequency of low-demand data, so as to give full play to the advantages of each transmission path and avoid unreasonable utilization of the transmission path.

Description

Technical Field

[0001] The present invention relates to the technical field of network transmission optimization, and specifically to an optimization method for a distributed edge computing network. Background Art

[0002] With the rapid development of information technology, distributed edge computing networks have been widely used in many fields, such as industrial Internet, intelligent transportation, telemedicine, etc. In these application scenarios, the quality of network performance directly affects the overall operating efficiency and service quality of the system.

[0003] The patent with publication number CN104348745A discloses a method for optimizing data transmission in a wireless network system and a related wireless network system. In the wireless network system, a first channel is established between a user device and a third generation mobile communication alliance architecture network, and a second channel is established between the user device and an Internet protocol architecture network. The user device measures its transmission state and calculates a path maximum transmission unit / cut size for communication with the core network. The core network calculates an optimized maximum transmission unit / cut size based on the measured transmission state and the calculated path maximum transmission unit / cut size, and adjusts its coding scheme based on the optimized maximum transmission unit / cut size.

[0004] However, distributed edge computing networks face complex and ever-changing network environments, including different network topologies, diverse data types, and unstable network connections. Traditional network management methods often lack comprehensive perception of the network environment and targeted optimization strategies, making it difficult to meet the strict requirements of distributed edge computing networks for efficient and stable data transmission. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an optimization method for a distributed edge computing network, which solves the problem that there is often a lack of comprehensive perception of the network environment and targeted optimization strategies, and it is difficult to meet the strict requirements of the distributed edge computing network for efficient and stable data transmission.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for optimizing a distributed edge computing network, the method specifically comprising the following steps:

[0007] Step 1: Obtain the data transmission protocol of the network, and at the same time obtain the network environment of the current network, and determine whether the transmission protocol needs to be replaced and adjusted according to the network environment, and generate adjustment information;

[0008] Step 2: Classify the transmission data according to the data type, and select the corresponding compression method to process the classified transmission data to obtain pre-processed data;

[0009] Step 3: Analyze the transmission paths of the distributed network, classify the transmission paths by calculating the path state values ​​of the transmission paths, and obtain stable transmission paths and unstable transmission paths;

[0010] Step 4: Analyze the classified stable transmission paths, classify the preprocessed data to obtain high-demand data and low-demand data, perform transmission analysis on the high-demand data, and generate transmission optimization information;

[0011] Step 5: Analyze the unstable transmission paths obtained through classification, analyze the low-demand data, and optimize the transmission according to the data properties of the low-demand data to generate transmission optimization information.

[0012] As a further solution of the present invention, the specific method of generating the adjustment information in step 1 is:

[0013] Get the packet loss rate and broadband delay of the current network, and then judge the packet loss rate and broadband delay. If the packet loss rate and broadband delay are normal, it means that the current network environment is stable. On the contrary, if there is any abnormality in the packet loss rate or broadband delay, it means that the current network environment is unstable.

[0014] Then, the data transmission protocol is obtained, and the network environment corresponding to the data protocol is judged. If the network environment corresponding to the data protocol does not match the current network environment, it means that the data transmission protocol needs to be adjusted, and the standard data transmission protocol is selected for adjustment to generate adjustment information. Conversely, if the network environment corresponding to the data protocol matches the current network environment, it means that the data transmission protocol does not need to be adjusted.

[0015] As a further solution of the present invention, the specific method of obtaining the preprocessed data in step 2 is:

[0016] Acquire all transmission data and the data type corresponding to the transmission data, then classify the transmission data according to the data type to obtain transmission data of the same type, compress the obtained transmission data of the same type, select a compression method according to the data type of the transmission data of the same type, and compress according to the selected compression method to obtain preprocessed data.

[0017] As a further solution of the present invention, the specific method of analyzing the transmission path in step 3 is:

[0018] Get all the transmission paths of the distributed network and label them as i, where i=1, 2, ..., j, where j represents the number of transmission paths. Then get the broadband delay and packet loss rate corresponding to the transmission path i, and substitute the obtained broadband delay and packet loss rate into the formula The path state value Qi corresponding to the transmission path i is calculated, where and Respectively represent the weight coefficients of broadband delay Ki and packet loss rate Di, and satisfy , the specific value is set by the operator;

[0019] Then the calculated path state value Qi is compared with the preset value Qy, and the specific value of the preset value Qy is set by the operator. If the path state value Qi is greater than the preset value Qy, the corresponding transmission path is classified as a stable transmission path, otherwise the corresponding transmission path is classified as an unstable transmission path.

[0020] As a further solution of the present invention, the specific method of analyzing the stable transmission path in step 4 is:

[0021] The classified stable transmission path is obtained and recorded as a, where a=1, 2, ..., b, where b represents the number of stable transmission paths. All preprocessed data are obtained and classified according to transmission requirements to obtain high-demand data and low-demand data, and the data capacity corresponding to all high-demand data is calculated and recorded as L. Then, the transmission capacity per unit time P corresponding to the stable transmission path a with the largest path state value is obtained, and the data capacity L is compared with the transmission capacity per unit time P.

[0022] If the data capacity L is smaller than the transmission capacity per unit time P, the data is directly transmitted and transmission optimization information is generated. Otherwise, a secondary analysis signal is generated and processed.

[0023] As a further solution of the present invention, the specific method of processing the secondary analysis signal in step 4 is:

[0024] All stable transmission paths a are obtained, and the path state mean corresponding to the stable transmission path a is calculated. The stable transmission paths are screened based on the path state mean to obtain secondary screening paths labeled o, where o=1, 2, ..., h. Then, the data capacity L is evenly divided based on the number of secondary screening paths to obtain evenly divided data packets. All secondary screening paths o are obtained, and the evenly divided data packets are distributed and transmitted according to the labels of the secondary screening paths o.

[0025] As a further solution of the present invention, the specific method of analyzing the unstable transmission path in step 5 is:

[0026] Obtain all unstable transmission paths, then obtain all low-demand data and label them as f, where f=1, 2, ..., g, where g represents the number of low-demand data, and obtain the number of accesses to the low-demand data f within the time period T as Cf, and compare the number of accesses Cf with the access preset value Cy;

[0027] If the number of access times Cf is greater than the preset access value Cy, the corresponding low-demand data will be marked as high-frequency access data, otherwise the corresponding low-demand data will be marked as low-frequency access data. At the same time, the marked high-frequency access data will be stored in the edge node, and transmission optimization information will be generated. The low-frequency access data will be analyzed according to the number of unstable transmission paths.

[0028] As a further solution of the present invention, the specific method of performing transmission analysis on low-frequency access data in step 5 is:

[0029] All low-frequency access data are obtained, and the low-frequency access data are integrated to obtain the data to be transmitted. At the same time, the data to be transmitted are evenly divided according to the data capacity, and then the evenly divided data to be transmitted are transmitted through different unstable transmission paths respectively, and transmission optimization information is generated.

[0030] The present invention provides an optimization method for a distributed edge computing network. Compared with the prior art, it has the following beneficial effects:

[0031] The present invention obtains network packet loss rate, broadband delay and other indicators, and then dynamically adjusts the protocol according to the matching situation between the network environment and the data transmission protocol, so as to ensure that the data transmission protocol always adapts to the current network environment, improve the data transmission efficiency and stability, effectively solve the problem of poor adaptability to the network environment, carefully classify the transmission data according to the data type, and select the appropriate compression method for different types, so as to reduce the network transmission burden, improve the utilization rate of network resources, and overcome the lack of pertinence in data processing;

[0032] Accurately classify the transmission paths into stable transmission paths and unstable transmission paths, and plan data transmission for different types of paths. For stable transmission paths, formulate reasonable transmission strategies based on the comparison results between the data capacity of high-demand data and the transmission capacity per unit time of the path to achieve efficient transmission of high-demand data; for unstable transmission paths, perform edge node storage or optimized transmission based on the access frequency of low-demand data, give full play to the advantages of each transmission path, and avoid unreasonable utilization of the transmission path;

[0033] On the stable transmission path, high-demand data and low-demand data are reasonably classified and analyzed for transmission. When the capacity of high-demand data does not match the capacity of the stable transmission path corresponding to the maximum path state value, the stable transmission path is screened twice and the data is evenly processed and distributed for transmission to ensure the reliable transmission of high-demand data. On the unstable transmission path, according to the access frequency characteristics of low-demand data, high-frequency access data is stored on edge nodes, and low-frequency access data is integrated and evenly divided and transmitted through different unstable transmission paths, effectively improving the overall efficiency of data transmission and solving the problem of imperfect data transmission strategy.

[0034] For low-demand data on unstable transmission paths, by analyzing the relationship between its access times and preset values, the low-demand data is further marked as high-frequency access data and low-frequency access data, and edge node storage and equal transmission strategies based on the number of unstable transmission paths are adopted respectively. This can improve the pertinence and effectiveness of data transmission on unstable transmission paths, improve the situation of improper handling of unstable transmission paths, and enhance the performance and reliability of the entire distributed edge computing network. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a diagram of the steps of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] For example, see Figure 1 , the present application provides a method for optimizing a distributed edge computing network, the method specifically comprising the following steps:

[0038] Step 1: Obtain the data transmission protocol of the network, and at the same time obtain the network environment of the current network, and determine whether the transmission protocol needs to be replaced and adjusted according to the network environment, and generate adjustment information.

[0039] The packet loss rate and broadband delay of the current network are obtained, and then the packet loss rate and broadband delay are judged, and the packet loss rate and broadband delay are specifically compared with normal values. The normal value is set by the operator, and the difference between the packet loss rate and the normal value of the packet loss rate, and the broadband delay and the normal value of the broadband delay are calculated respectively. If the packet loss rate and the broadband delay are normal, it means that the current network environment is stable. On the contrary, if there is any set of abnormalities in the packet loss rate or the broadband delay, it means that the current network environment is unstable. If the specific difference is within the allowable range, it means normal. If it exceeds the allowable range, it means abnormal. The specific value of the allowable range is set by the operator. For example, in a distributed edge computing network of an enterprise office, the operator sets the normal value of the packet loss rate to be less than 5%, and the allowable range is set to fluctuate by 2%, that is, 3%-7% is an acceptable range; the normal value of the broadband delay is set to be less than 30 milliseconds, and the allowable range is set to fluctuate by 10 milliseconds, that is, 20 milliseconds-40 milliseconds is normal. If the actual measured packet loss rate is 4% and the broadband delay is 25 milliseconds, both of which are within their respective allowable ranges, then the network environment of the enterprise office network can be determined to be stable.

[0040] On the contrary, if either the packet loss rate or the broadband delay exceeds the allowable range set by the operator, it means that the current network environment is unstable. For example, in the above-mentioned corporate office network, if the actual measured packet loss rate is 8%, even if the broadband delay is 22 milliseconds (within the normal range), the network environment is judged to be unstable because the packet loss rate exceeds the set allowable range of 3% to 7%.

[0041] Then, the data transmission protocol is obtained, and the network environment corresponding to the data protocol is judged. If the network environment corresponding to the data protocol does not match the current network environment, it means that the data transmission protocol needs to be adjusted, and the standard data transmission protocol is selected for adjustment to generate adjustment information. On the contrary, if the network environment corresponding to the data protocol matches the current network environment, it means that the data transmission protocol does not need to be adjusted, and the standard data transmission protocol is indicated as a transmission protocol that adapts to the current network environment and adapts to the most data formats for transmission.

[0042] Step 2: Classify the transmission data according to the data type, and select the corresponding compression method to process the classified transmission data to obtain pre-processed data.

[0043] All transmission data are obtained, and the data types corresponding to the transmission data are obtained, such as image data, audio data, multimedia data, and text data, etc. Then, the transmission data is classified according to the data type to obtain the same type of transmission data, and the obtained same type of transmission data is compressed, and the compression method is selected according to the data type of the same type of transmission data. For text data, lossless compression algorithms such as Huffman coding and Lempel-Ziv-Welch (LZW) algorithm can be used. For multimedia data, a suitable compression method is selected according to the requirements for data quality. For example, for video surveillance data, on the premise of ensuring that the surveillance object can be identified, video compression standards such as H.264 or H.265 can be used, and compression processing is performed according to the selected compression method to obtain preprocessed data.

[0044] Step three: Analyze the transmission paths of the distributed network, classify the transmission paths by calculating the path state values ​​of the transmission paths, and obtain stable transmission paths and unstable transmission paths.

[0045] Get all the transmission paths of the distributed network and label them as i, where i=1, 2, ..., j, where j represents the number of transmission paths. Then get the broadband delay and packet loss rate corresponding to the transmission path i, and the broadband delay and packet loss rate obtained here are the averages within time t. Substitute the obtained broadband delay and packet loss rate into the formula The path state value Qi corresponding to the transmission path i is calculated, where and Respectively represent the weight coefficients of broadband delay Ki and packet loss rate Di, and satisfy , the specific value is set by the operator;

[0046] Then, the calculated path state value Qi is compared with the preset value Qy, and the specific value of the preset value Qy is set by the operator. If the path state value Qi is greater than the preset value Qy, the corresponding transmission path is classified as a stable transmission path. Conversely, if the path state value Qi is less than the preset value Qi, the corresponding transmission path is classified as an unstable transmission path.

[0047] Step 4: Analyze the stable transmission path obtained by classification, obtain high-demand data and low-demand data by classifying the preprocessed data, and perform transmission analysis on the high-demand data to generate transmission optimization information.

[0048] The classified stable transmission path is obtained and recorded as a, where a=1, 2, ..., b, where b represents the number of stable transmission paths. Then all preprocessed data are obtained and classified according to the transmission requirements. The transmission requirements here represent the requirements for corresponding data transmission. For example, financial data transmission has extremely high requirements for data integrity. For example, text data does not have high requirements during transmission. High-demand data and low-demand data are obtained, and the high-demand data are labeled as n, where n=1, 2, ..., m, where m represents the number of high-demand data. At the same time, the data capacity corresponding to all high-demand data is calculated and recorded as L, and the data capacity here is the total capacity value and of all high-demand data. Then the transmission capacity per unit time P corresponding to the stable transmission path a with the largest path state value is obtained, and the data capacity L is compared with the transmission capacity per unit time P.

[0049] If the data capacity L is less than the transmission capacity per unit time P, it means that the stable transmission path corresponding to the maximum path state value can transmit the high-demand data at one time, and directly transmit it, generating transmission optimization information. On the contrary, if the data capacity L is greater than the transmission capacity per unit time P, it means that it cannot be transmitted at one time, and a secondary analysis signal is generated;

[0050] The generated secondary analysis signal is processed to obtain all stable transmission paths a, and the path state mean corresponding to the stable transmission path a is calculated. The stable transmission path is screened based on the path state mean to obtain a secondary screening path number denoted as o, and o=1, 2, ..., h, and the secondary screening path here is represented as the corresponding path state value greater than the path state mean. Then, the data capacity L is evenly divided based on the number of secondary screening paths as the segmentation standard to obtain an evenly divided data packet, and the evenly divided data packet here includes at least one group of high-demand data. The specific even division is divided according to the original number, and the evenly divided data packet is subjected to transmission analysis, and the specific transmission analysis method is as follows:

[0051] Obtain all secondary screening paths o, distribute and transmit the equally divided data packets according to the labels of the secondary screening paths o, and generate transmission optimization information.

[0052] Step 5: Analyze the unstable transmission paths obtained through classification, analyze the low-demand data, and optimize the transmission according to the data properties of the low-demand data to generate transmission optimization information.

[0053] Obtain all unstable transmission paths, then obtain all low-demand data and label them as f, and f=1, 2, ..., g, where g represents the number of low-demand data, and at the same time obtain the number of accesses to the low-demand data f within the time period T and record it as Cf, and compare the number of accesses Cf with the access preset value Cy, and the value of the access preset value Cy here is set by the operator, if the number of accesses Cf is greater than the access preset value Cy, then the corresponding low-demand data is marked as high-frequency access data, if the number of accesses Cf is less than the access preset value Cy, then the corresponding low-demand data is marked as low-frequency access data, and the marked high-frequency access data is stored at the edge node, and transmission optimization information is generated, and the low-frequency access data is analyzed for transmission according to the number of unstable transmission paths;

[0054] All low-frequency access data are obtained, and the low-frequency access data are integrated to obtain the data to be transmitted. At the same time, the data to be transmitted is evenly divided according to the data capacity, and the criterion for equalization here is to divide it evenly according to the number of unstable transmission paths. Then, the evenly divided data to be transmitted are transmitted through different unstable transmission paths respectively, and transmission optimization information is generated.

[0055] Embodiment 2: This embodiment is implemented on the basis of Embodiment 1, and the difference from Embodiment 1 is as follows:

[0056] When judging the stability of the network environment, you can also judge it through the jitter indicators corresponding to the network environment. Specifically, you need to analyze the fluctuation of the delay and measure it through professional network monitoring tools. These tools will continuously record the delay data over a period of time and calculate the standard deviation of the delay to measure the degree of jitter. For real-time video and audio applications, jitter below 10ms is relatively good, and the network environment is relatively stable; if the jitter is between 10-30ms, it may cause occasional freezes or discontinuities in the video or audio; when the jitter exceeds 30ms, the quality of real-time applications will be severely reduced. For example, video calls will frequently experience deterioration in picture and voice quality, which indicates that the network environment is unstable.

[0057] Embodiment 3, as the embodiment 3 of the present invention, focuses on combining the implementation processes of embodiment 1 and embodiment 2 for implementation.

[0058] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0059] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A distributed edge computing network optimization method, characterized in that: The method specifically comprises the following steps: Step 1: Obtain the data transmission protocol of the network, and at the same time obtain the network environment of the current network, and determine whether the transmission protocol needs to be replaced and adjusted according to the network environment, and generate adjustment information; Step 2: Classify the transmission data according to the data type, and select the corresponding compression method to process the classified transmission data to obtain pre-processed data; Step 3: Analyze the transmission paths of the distributed network, classify the transmission paths by calculating the path state values ​​of the transmission paths, and obtain stable transmission paths and unstable transmission paths; Step 4: Analyze the classified stable transmission paths, classify the preprocessed data to obtain high-demand data and low-demand data, and perform transmission analysis on the high-demand data to generate transmission optimization information. The specific processing method is as follows: The classified stable transmission path is obtained and recorded as a, where a=1, 2, ..., b, where b represents the number of stable transmission paths. All preprocessed data are obtained and classified according to transmission requirements to obtain high-demand data and low-demand data, and the data capacity corresponding to all high-demand data is calculated and recorded as L. Then, the transmission capacity per unit time P corresponding to the stable transmission path a with the largest path state value is obtained, and the data capacity L is compared with the transmission capacity per unit time P. If the data capacity L is less than the transmission capacity per unit time P, the data is directly transmitted to generate transmission optimization information. Otherwise, a secondary analysis signal is generated and processed. Step 5: Analyze the unstable transmission paths obtained by classification, analyze the low-demand data, and optimize the transmission according to the data properties of the low-demand data to generate transmission optimization information. The specific processing method is: Obtain all unstable transmission paths, then obtain all low-demand data and label them as f, where f=1, 2, ..., g, where g represents the number of low-demand data, and obtain the number of accesses to the low-demand data f within the time period T as Cf, and compare the number of accesses Cf with the access preset value Cy; If the number of access times Cf is greater than the preset access value Cy, the corresponding low-demand data will be marked as high-frequency access data, otherwise the corresponding low-demand data will be marked as low-frequency access data. At the same time, the marked high-frequency access data will be stored in the edge node, and transmission optimization information will be generated. The low-frequency access data will be analyzed according to the number of unstable transmission paths.

2. The method for optimizing a distributed edge computing network according to claim 1, characterized in that: The specific method of generating the adjustment information in step 1 is: Get the packet loss rate and broadband delay of the current network, and then judge the packet loss rate and broadband delay. If the packet loss rate and broadband delay are normal, it means that the current network environment is stable. On the contrary, if there is any abnormality in the packet loss rate or broadband delay, it means that the current network environment is unstable. Then, the data transmission protocol is obtained, and the network environment corresponding to the data protocol is judged. If the network environment corresponding to the data protocol does not match the current network environment, it means that the data transmission protocol needs to be adjusted, and the standard data transmission protocol is selected for adjustment to generate adjustment information. Conversely, if the network environment corresponding to the data protocol matches the current network environment, it means that the data transmission protocol does not need to be adjusted.

3. The method for optimizing a distributed edge computing network according to claim 1, characterized in that: The specific method of obtaining the preprocessed data in step 2 is: Acquire all transmission data and the data type corresponding to the transmission data, then classify the transmission data according to the data type to obtain transmission data of the same type, compress the obtained transmission data of the same type, select a compression method according to the data type of the transmission data of the same type, and compress according to the selected compression method to obtain preprocessed data.

4. The method for optimizing a distributed edge computing network according to claim 1, characterized in that: The specific method of analyzing the transmission path in step 3 is: Get all the transmission paths of the distributed network and label them as i, where i=1, 2, ..., j, where j represents the number of transmission paths. Then get the broadband delay and packet loss rate corresponding to the transmission path i, and substitute the obtained broadband delay and packet loss rate into the formula The path state value Qi corresponding to the transmission path i is calculated, where and Respectively represent the weight coefficients of broadband delay Ki and packet loss rate Di, and satisfy , the specific value is set by the operator; Then the calculated path state value Qi is compared with the preset value Qy, and the specific value of the preset value Qy is set by the operator. If the path state value Qi is greater than the preset value Qy, the corresponding transmission path is classified as a stable transmission path, otherwise the corresponding transmission path is classified as an unstable transmission path.

5. The method for optimizing a distributed edge computing network according to claim 1, characterized in that: The specific method of processing the secondary analysis signal in step 4 is: All stable transmission paths a are obtained, and the path state mean corresponding to the stable transmission path a is calculated. The stable transmission paths are screened based on the path state mean to obtain secondary screening paths labeled o, where o=1, 2, ..., h. Then, the data capacity L is evenly divided based on the number of secondary screening paths to obtain evenly divided data packets. All secondary screening paths o are obtained, and the evenly divided data packets are distributed and transmitted according to the labels of the secondary screening paths o.

6. The method for optimizing a distributed edge computing network according to claim 1, characterized in that: The specific method of performing transmission analysis on low-frequency access data in step 5 is: All low-frequency access data are obtained, and the low-frequency access data are integrated to obtain the data to be transmitted. At the same time, the data to be transmitted are evenly divided according to the data capacity, and then the evenly divided data to be transmitted are transmitted through different unstable transmission paths respectively, and transmission optimization information is generated.

Images