Business data management method and system based on container cloud

By analyzing data flow and network status in the container cloud, identifying abnormal behaviors and adjusting security policies, the problem of difficulty in real-time data monitoring and dynamic security management in the existing technology is solved, and the security and adaptability of data transmission and storage are improved. .

CN118802376BActive Publication Date: 2024-11-22BEIJING ZHONGDIAN HUITONG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411274677.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-12
Publication Date
2024-11-22
Estimated Expiration
2044-09-12

AI Technical Summary

Technical Problem

Existing data security technologies are difficult to implement real-time data monitoring and dynamic security management in large-scale distributed network environments, and static security rules are difficult to capture subtle exceptions and adapt to new threat patterns, resulting in data integrity and security issues.

Method used

Adopting a business data management method based on container cloud, by collecting and analyzing data flows and network status, calculating data transmission frequency and duration, filtering active and inactive data channels, establishing network behavior patterns, identifying high traffic and key dependency nodes, and analyzing Key dependency graphs, identify abnormal behavior metrics, and adjust network slicing configuration and security policies.

Benefits of technology

Real-time monitoring and dynamic security management of data transmission between multiple nodes is realized, the control accuracy of network traffic distribution and dependencies is improved, security risks are detected early, and the security and adaptability of data transmission and storage are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118802376B_ABST
    Figure CN118802376B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data security technology, specifically to a business data management method and system based on a container cloud, comprising the following steps: collecting business data flows and network status in a container cloud, analyzing data transmission security records between multiple nodes, calculating data transmission frequency and duration, screening active and inactive data channels, and obtaining network behavior patterns. In the present invention, by analyzing the business data flows and network status in a container cloud, further insight into network behavior patterns is obtained, data access patterns between nodes are analyzed, and key dependency graphs are established, which greatly enhances the visualization and prediction capabilities of network management, provides higher control accuracy for network traffic distribution and dependencies, and effectively identifies abnormal increases and decreases and changes in flow direction in data flows through analysis of key dependency graphs, and detects security risks or data integrity issues early, thereby improving the security and adaptability of data transmission and storage by adjusting network slice configurations and updating security policies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data security technology, and in particular to a business data management method and system based on container cloud. Background Art

[0002] The field of data security technology focuses on protecting data from unauthorized access, use, disclosure, destruction, modification, or destruction. This field includes a variety of technologies and practices such as data encryption, access control, authentication, network security policies, and data integrity maintenance. The implementation of data security is critical to preventing data breaches, protecting individual privacy, and maintaining the reputation and compliance of an organization. As technology advances and the amount of data generated increases, the field of data security continues to evolve, including the introduction of more advanced encryption technologies, cloud security solutions, and AI-driven security tools.

[0003] Among them, business data management methods is a topic involving strategies and technologies for effectively managing and using data in an organization, including all aspects of data collection, storage, access and analysis, ensuring the accuracy and availability of data to support business decisions and operational efficiency. The purpose of this topic is to help companies better understand and utilize their data resources, optimize business processes, and improve the quality of decision-making, while ensuring data security and compliance with relevant regulatory requirements. Business data management achieves these goals by establishing a clear data governance framework, implementing data quality control, and providing cross-departmental data access.

[0004] Although existing data security technologies have made progress in preventing data leakage and protecting personal privacy, they are obviously insufficient in dealing with real-time data monitoring and dynamic security management in large-scale distributed network environments. Traditional methods make it difficult to conduct effective real-time monitoring and instant analysis of widely dispersed data streams, which affects the response speed and timeliness of data management in a rapidly changing network environment. In addition, existing technologies rely on static security rules to detect data flow anomalies. When faced with complex and changing transmission and access patterns, this method often has difficulty capturing subtle anomalies or adapting to new threat patterns. These deficiencies may lead to data integrity and security issues, affecting the overall operating efficiency and protection capabilities of the system. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a business data management method and system based on container cloud.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a service data management method based on container cloud, comprising the following steps:

[0007] S1: Collect business data flows and network status in the container cloud, analyze data transmission security records between multiple nodes, calculate data transmission frequency and duration, filter active and inactive data channels, and obtain network behavior patterns;

[0008] S2: Based on the network behavior pattern, analyze the data access pattern between nodes, identify the nodes with high traffic and key dependencies, and obtain the key dependency graph by establishing the relationship between network traffic and node dependency;

[0009] S3: Analyze the key dependency graph, identify abnormal increases and decreases and changes in flow direction in the data flow by comparing the deviations between the current data integrity of multiple nodes and the historical data flow, record the deviation values, and obtain abnormal behavior indicators;

[0010] S4: Based on the abnormal behavior indicators, adjust the affected network slice configuration, reset data storage security measures, update the security policy of the network slice, record the operations and results of multiple adjustments, establish a policy adjustment log, and perform container cloud data storage.

[0011] As a further solution of the present invention, the step of calculating the data transmission frequency and duration is specifically:

[0012] S111: Collect the service data flow and network status of the container cloud, filter the metadata of the data transmission record, extract the timestamp and duration information of the data transmission from the filtered metadata, calculate the transmission start and end time difference of each data channel, and obtain the duration of each data channel , the reference formula is as follows:

[0013] ,

[0014] in, is the duration of the data channel, and are the end and start time of data transmission, respectively, and t represents the time interval;

[0015] S112: Calculate the data transmission frequency, where F is the number of data transmissions completed per unit time, according to the following formula:

[0016] ,

[0017] Then summarize the data transmission frequency and duration of each data channel;

[0018] Where F is the data transmission frequency, represents the amount of data transmitted for the i-th time, T is the total observation time, is the duration of the data channel, is the weight coefficient, is a coefficient to ensure the stability of calculation, is the attenuation coefficient, is an exponential decay.

[0019] As a further solution of the present invention, the step of screening active and inactive data channels is specifically:

[0020] S121: Using the calculation results of the data transmission frequency and duration, setting an active threshold and inactive threshold ,in, Higher than average data transmission frequency, Lower than average data transmission frequency;

[0021] S122: For each data channel, compare the data transmission frequency F. If F is greater than , it is marked as an active channel. If F is less than , it is marked as an inactive channel;

[0022] S123: Count the marking results of the data channels to obtain the number of active and inactive data channels in the network, with reference to the following formula:

[0023] ,

[0024] ,

[0025] in, is the number of active data channels, is the number of inactive data channels, represents the data transmission frequency of the i-th data channel, is the active threshold, which is set to a value higher than the average data transmission frequency. is the inactivity threshold, which is set to a value lower than the average data transmission frequency. , is the improved indicator function, , is the flexibility parameter.

[0026] As a further solution of the present invention, the step of acquiring the network behavior pattern is specifically as follows:

[0027] S131: Analyze the active and inactive data channels, monitor the behavior characteristics and transmission modes of the differentiated data channels, perform time series analysis on the active data channels, extract key activity time periods and peak transmission periods, and analyze frequency data on the inactive data channels;

[0028] S132: Based on the time series and frequency data, a network behavior model is constructed, referring to the following formula:

[0029] ,

[0030] Among them, P represents the behavior pattern of the entire network, and They are frequency indicators of active and inactive data channels, , is the weight factor, , is the scaling factor, , are the growth rate and decay rate, respectively. , is the frequency index of active (a) and inactive (n) data channels, , is an exponential function.

[0031] As a further solution of the present invention, the step of identifying nodes with high traffic and key dependencies is specifically:

[0032] S211: Based on the network behavior pattern, identify the data access pattern, calculate the total amount of data transmission between nodes, and set the multiple nodes as and , the data transmission volume is , for each node , calculate the total outflow and total inflow, refer to the following formula:

[0033] ,

[0034] ,

[0035] in, For Node The total outflow rate, For Node The total inflow, , Representative Node To Node and nodes To Node The amount of data transfer, and is the weight assigned according to the quality or importance of the connection between nodes. and represents the distance between nodes, is the attenuation coefficient;

[0036] S212: By comparing each node and , set the threshold Determine the high traffic nodes. or , then the node are classified as high traffic nodes, where Set thresholds for node traffic;

[0037] By analyzing the total amount of data transmission between nodes, the calculation node The degree of dependence on other nodes is calculated using the dependency relationship, with the reference formula as follows:

[0038] ,

[0039] What you get The high dependency is represented as a key dependency node;

[0040] in, For Node The dependence of The weight representing the strength of the dependencies between nodes.

[0041] As a further solution of the present invention, the step of obtaining the key dependency graph is specifically as follows:

[0042] S221: Analyze the interactive data traffic between nodes to determine the traffic-intensive relationship using the high-traffic nodes and key dependent nodes information, and calculate the traffic contribution of each node pair to other nodes, referring to the following formula:

[0043] ,

[0044] in, Represents a slave node To Node A measure of the traffic density between Is a node arrive The amount of data transfer, Is a node The total outflow rate, is a weight factor, which is assigned according to the traffic contribution and node importance;

[0045] S222: By comparing each node pair value, set a threshold ,like , then the node and There is a traffic-intensive relationship between them, and the relationship between network traffic and node dependency is established to obtain the key dependency graph.

[0046] As a further solution of the present invention, the step of obtaining the abnormal behavior indicator is specifically as follows:

[0047] S311: Analyze the nodes in the key dependency graph, calculate the current data integrity, and set each node The current data integrity of , the amount of real-time data obtained through real-time monitoring and expected data volume Calculation, refer to the following formula:

[0048] ,

[0049] in, For Node The current data integrity, is the adjustment parameter, For Node The amount of real-time data at the current time t, For Node The expected amount of data at the current time t;

[0050] S312: Calculate the historical data flow integrity based on the average integrity of the data of the same node in the past cycle, with reference to the following formula:

[0051] ,

[0052] in, For historical data flow integrity, It is a parameter that enhances the influence of historical data;

[0053] S313: Comparing the data integrity Integrity with historical data flow , calculate the deviation value , is the difference between the current data integrity and the historical data stream integrity, and the reference formula is as follows:

[0054] ,

[0055] in, is the adjustment factor for the deviation value, is the deviation between the current data integrity and the historical data stream integrity, through Parameters are adjusted;

[0056] S314: Setting the abnormality determination threshold ,like , it is considered abnormal, the deviation value is recorded, and the abnormal behavior index is obtained.

[0057] As a further solution of the present invention, the steps of establishing a policy adjustment log and performing container cloud data storage are specifically as follows:

[0058] S411: Collect abnormal behavior indicators and apply network slicing adjustment formula:

[0059] ,

[0060] Update configuration;

[0061] in, Represents the new network slice configuration, For the original configuration, The abnormal behavior indicator processing result, , , It is to adjust the parameters to reflect the intensity and adaptability of the impact of abnormal indicators on the configuration, and obtain the pre-adjusted network slice configuration;

[0062] S412: Verify the adaptability and stability of the pre-adjusted configuration, using the effect verification formula:

[0063] ,

[0064] in, Calculate the relative change of the configuration adjustment and establish the adjusted network slice configuration;

[0065] S413: According to the adjustment process and verification results, record the adjustment details and verification data of each configuration, establish a policy adjustment log, and store container cloud data.

[0066] A business data management system based on a container cloud, wherein the business data management system based on a container cloud is used to execute the above-mentioned business data management method based on a container cloud, and the system includes:

[0067] The data collection and analysis module uses container cloud monitoring to capture business data flows and network status in real time, combines inter-node communication data, calculates data transmission frequency and duration, and screens active and inactive data channels to integrate them into network behavior data graphs.

[0068] The network behavior analysis module identifies the data access frequency and traffic size between multiple nodes by analyzing the network behavior data graph, identifies and marks nodes with high traffic and key dependent nodes, establishes the relationship between network traffic and node dependency, and obtains a key dependency graph;

[0069] The abnormal behavior analysis module compares the integrity of the current node data with the historical records based on the key dependency graph, identifies abnormal increases and decreases in data flow and changes in flow direction, records and analyzes deviation values, and generates abnormal behavior indicators;

[0070] The security policy adjustment module optimizes and adjusts the network slice configuration based on the abnormal behavior indicators, updates the security measures for data storage and transmission, records operations and adjustment results, establishes a policy adjustment log, and performs container cloud data storage.

[0071] Compared with the prior art, the advantages and positive effects of the present invention are:

[0072] In the present invention, by analyzing the business data flow and network status in the container cloud, the data transmission security records between multiple nodes are monitored in real time, the data transmission frequency and duration are effectively calculated, the active and inactive data channels are screened out, and further insights into network behavior patterns are gained. By analyzing the data access patterns between nodes and establishing key dependency graphs, the visualization and prediction capabilities of network management are greatly enhanced, providing higher control accuracy for network traffic distribution and dependencies. In addition, through the analysis of key dependency graphs, abnormal increases and decreases in data flows and changes in flow direction can be effectively identified, and security risks or data integrity issues can be discovered early, thereby improving the security and adaptability of data transmission and storage by adjusting network slice configurations and updating security policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0074] Figure 2 A flow chart of the steps of calculating the frequency and duration of data transmission according to the present invention;

[0075] Figure 3 A flowchart of the steps of screening active and inactive data channels of the present invention;

[0076] Figure 4 A flowchart of the steps for obtaining the network behavior model of the present invention;

[0077] Figure 5 A flowchart of the steps of identifying high-traffic and critically dependent nodes of the present invention;

[0078] Figure 6 A flowchart of the steps for obtaining the key dependency graph of the present invention;

[0079] Figure 7 A flowchart of the steps for obtaining abnormal behavior indicators of the present invention;

[0080] Figure 8 This is a flow chart of the steps of establishing a strategy to adjust logs and storing container cloud data in the present invention. DETAILED DESCRIPTION

[0081] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0082] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0083] Embodiment 1:

[0084] See also Figure 1 The present invention provides a technical solution: a service data management method based on container cloud, comprising the following steps:

[0085] S1: Collect business data flows and network status in the container cloud, analyze data transmission security records between multiple nodes, calculate data transmission frequency and duration, filter active and inactive data channels, and obtain network behavior patterns;

[0086] S2: Based on the network behavior pattern, analyze the data access pattern between nodes, identify the nodes with high traffic and key dependencies, and obtain the key dependency graph by establishing the relationship between network traffic and node dependency;

[0087] S3: Analyze the key dependency graph, identify abnormal increases and decreases and changes in the data flow and flow direction by comparing the deviations between the current data integrity of multiple nodes and the historical data flow, record the deviation values, and obtain abnormal behavior indicators;

[0088] S4: Based on abnormal behavior indicators, adjust the affected network slice configuration, reset data storage security measures, update the security policy of the network slice, record the operations and results of multiple adjustments, establish a policy adjustment log, and perform container cloud data storage.

[0089] Network behavior patterns include frequency analysis, channel activity, and channel idle status. Key dependency graphs include node traffic distribution, dependency strength mapping, and key node identification. Abnormal behavior indicators include traffic deviation detection, abnormal channel identification, and flow abnormality indication. Policy adjustment logs include security configuration updates, access policy modifications, and isolation policy adjustment details.

[0090] See also Figure 2 , the steps for calculating the frequency and duration of data transmission are as follows:

[0091] S111: Collect the service data flow and network status of the container cloud, filter the metadata of the data transmission record, extract the timestamp and duration information of the data transmission from the filtered metadata, calculate the transmission start and end time difference of each data channel, and obtain the duration of each data channel , the reference formula is as follows:

[0092] ,

[0093] in, is the duration of the data channel, and are the end and start time of data transmission, respectively, and t represents the time interval;

[0094] S112: Calculate the data transmission frequency, where F is the number of data transmissions completed per unit time, according to the following formula:

[0095] ,

[0096] Then summarize the data transmission frequency and duration of each data channel;

[0097] Where F is the data transmission frequency, represents the amount of data transmitted for the i-th time, T is the total observation time, is the duration of the data channel, is the weight coefficient, is a coefficient to ensure the stability of calculation, is the attenuation coefficient, is an exponential decay.

[0098] Parameter Description:

[0099] : the amount of data transmitted for the i-th time; T: total observation time; : The average duration of the data channel, calculated as ,in and are the end and start times of the transmission; :adjust The amplification factor of the function increases the sensitivity of duration to frequency; : Stability coefficient to ensure that the denominator is non-zero; , : Forms a nonlinear adjustment to the duration, making the frequency calculation more sensitive to changes in duration.

[0100] Formula derivation and examples:

[0101] Assume that within a certain observation time T, data transmission is recorded 5 times, and the amount of data at each time They are 10MB, 15MB, 20MB, 10MB, and 25MB respectively; the total observation time T is 100 minutes; the average duration Assume 5 minutes.

[0102] Data volume calculation: ,

[0103] Basic frequency calculation: ,

[0104] Application formula:

[0105] Assumptions =2, =1, =3, =0.5;

[0106] ,

[0107] Calculation process overview:

[0108] Calculate the total amount of data , calculate the unadjusted fundamental frequency ;

[0109] use , , , Parameter to adjust the base frequency to reflect the impact of data transmission duration.

[0110] Through nonlinear function and an exponential decay function Combined with the adjustment coefficient, the impact of duration on transmission frequency is comprehensively considered.

[0111] This computational flow shows how, by combining different mathematical tools and parameter adjustments, the frequency of data transmission can be described precisely and flexibly, taking into account the actual influence of the duration on the transmission frequency.

[0112] See also Figure 3 ,The steps for screening active and inactive data channels are as follows:

[0113] S121: Using the calculation results of data transmission frequency and duration, setting an active threshold and inactive threshold ,in, Higher than average data transmission frequency, Lower than average data transmission frequency;

[0114] S122: For each data channel, compare the data transmission frequency F. If F is greater than , it is marked as an active channel. If F is less than , it is marked as an inactive channel;

[0115] S123: Count the marking results of the data channels to obtain the number of active and inactive data channels in the network, with reference to the following formula:

[0116] ,

[0117] ,

[0118] in, is the number of active data channels, is the number of inactive data channels, represents the data transmission frequency of the i-th data channel, is the active threshold, which is set to a value higher than the average data transmission frequency. is the inactivity threshold, which is set to a value lower than the average data transmission frequency. , is the improved indicator function, , is the flexibility parameter.

[0119] Parameter description and source:

[0120] : The data transmission frequency of the ith data channel. Assume that the frequency data of each channel is directly obtained from the network log;

[0121] : Active threshold, which is set based on the average data transmission frequency of the network plus a certain standard deviation. For example, assuming the average transmission frequency of the network is 50 times / hour and the standard deviation is 10 times / hour, then =60 times / hour;

[0122] : Inactivity threshold, usually set to the average transmission frequency minus a certain standard deviation; in the same example, =40 times / hour;

[0123] and : Adjustment parameters to adjust the sensitivity of threshold judgment, which can be adjusted according to the seasonality or periodicity of network behavior; assuming that the setting .

[0124] Formula derivation and calculation process:

[0125] Determine the threshold:

[0126] Calculate the average frequency and standard deviation based on historical data, set and .

[0127] Calculate the indicator function:

[0128] The function is used to identify the active channel and is defined as:

[0129] ,

[0130] The function is used to identify inactive channels and is defined as:

[0131] ,

[0132] Data channel evaluation:

[0133] For each data channel in the network, calculate its transmission frequency .

[0134] application and Function to determine whether each channel is active or inactive.

[0135] Calculation example:

[0136] Assume that the network has three data channels, and their transmission frequencies are , , times / hour.

[0137] for :

[0138] ,

[0139] ,

[0140] for :

[0141] ,

[0142] ,

[0143] for :

[0144] ,

[0145] ,

[0146] Through this process, it is possible to clearly identify whether each data channel is active or inactive, and to count it accordingly. and middle.

[0147] See also Figure 4 ,The specific steps for obtaining the network behavior pattern are:

[0148] S131: Analyze active and inactive data channels, monitor the behavior characteristics and transmission modes of differentiated data channels, perform time series analysis on active data channels, extract key activity time periods and peak transmission periods, and analyze frequency data on inactive data channels;

[0149] S132: Based on time series and frequency data, construct network behavior patterns, refer to the following formula:

[0150] ,

[0151] Among them, P represents the behavior pattern of the entire network, and They are frequency indicators of active and inactive data channels, , is the weight factor, , is the scaling factor, , are the growth rate and decay rate, respectively. , is the frequency index of active (a) and inactive (n) data channels, , is an exponential function.

[0152] Parameter Description:

[0153] and : Represents the frequency index of active and inactive data channels respectively; and : represents the weights of active and inactive data channels in the network behavior pattern; and : Adjust the total contribution of active and inactive channels; and : Adjusts the growth and decay rates of the effects of active and inactive frequencies.

[0154] Calculation example and derivation process:

[0155] Assume there are 3 active channels and 2 inactive channels. The frequency of active channels They are 2, 3, and 4 respectively. The frequency of the inactive channel They are 0.5 and 1 respectively; weights , Set to 1 respectively; set =0.6, =0.4; =1, =1.

[0156] Calculation steps:

[0157] Active channel contribution calculation:

[0158] ,

[0159] ;

[0160] Inactive channel contribution calculation:

[0161] ,

[0162] ;

[0163] Total network behavior pattern calculation: ;

[0164] Calculation process overview:

[0165] First, the exponential contribution of active and inactive data channels is calculated, and then the adjustment parameters are set according to and These contributions are adjusted and finally, the active and inactive contributions are added together to get the total network behavior pattern value.

[0166] This process demonstrates how to combine different mathematical tools and parameter adjustments to accurately describe and quantify network behavior patterns, especially in data channels with different activity levels.

[0167] See also Figure 5 ,The steps to identify high traffic and critical dependency nodes are:

[0168] S211: Based on the network behavior pattern, identify the data access pattern and calculate the total amount of data transmission between nodes. Suppose multiple nodes are and , the data transmission volume is , for each node , calculate the total outflow and total inflow, refer to the following formula:

[0169] ,

[0170] ,

[0171] in, For Node The total outflow rate, For Node The total inflow, , Representative Node To Node and nodes To Node The amount of data transfer, and is the weight assigned according to the quality or importance of the connection between nodes. and represents the distance between nodes, is the attenuation coefficient;

[0172] S212: By comparing each node and , set the threshold Determine the high traffic nodes. or , then the node are classified as high traffic nodes, where Set thresholds for node traffic;

[0173] By analyzing the total amount of data transmission between nodes, the calculation node The degree of dependence on other nodes is calculated using the dependency relationship, with the reference formula as follows:

[0174] ,

[0175] What you get The high dependency is represented as a key dependency node;

[0176] in, For Node The dependence of The weight representing the strength of the dependencies between nodes.

[0177] Parameter Description:

[0178] :node To Node The amount of data transmission;

[0179] , : weights of the traffic, these weights take into account the quality of the connection or the importance between nodes, and increase the importance of more important connections by an exponent of 1.2;

[0180] , : The physical or logical distance between nodes;

[0181] : Attenuation coefficient, which adjusts the effect of distance on weight so that the influence of farther nodes is reduced.

[0182] Hypothetical example and calculation process

[0183] Assume there are three nodes , , ;The data transmission volume and distance between nodes are as follows:

[0184] ,

[0185] ,

[0186] Weight , , ;

[0187] Attenuation coefficient =0.1;

[0188] calculate and :

[0189] Outflow:

[0190] ,

[0191] Inflow:

[0192] ,

[0193] This calculation flow shows how to calculate the inbound and outbound traffic of a node by applying the specific application of transmission volume, weight, distance and attenuation coefficient. This method can effectively identify high-traffic nodes in the network for network optimization or security analysis.

[0194] See also Figure 6 , the steps to obtain the key dependency graph are as follows:

[0195] S221: Analyze the interactive data traffic between nodes using the information of high-traffic nodes and key dependent nodes to determine the traffic-intensive relationship, and calculate the traffic contribution of each node pair to other nodes, referring to the following formula:

[0196] ,

[0197] in, Represents a slave node To Node A measure of the traffic density between Is a node arrive The amount of data transfer, Is a node The total outflow rate, is a weight factor, which is assigned according to the traffic contribution and node importance;

[0198] S222: By comparing each node pair value, set a threshold ,like , then the node and There is a traffic-intensive relationship between them, and the relationship between network traffic and node dependency is established to obtain the key dependency graph.

[0199] Parameter Description:

[0200] :node To Node The data transmission volume represents the direct data flow between two nodes;

[0201] :node The total outflow rate is expressed as The total amount of data sent to all other nodes;

[0202] : Weights assigned based on traffic contribution and importance between nodes are used to adjust and highlight important traffic relationships;

[0203] : Threshold used to determine significant traffic density relationships.

[0204] Hypothetical example and calculation process

[0205] Assume there are three nodes , , ; Data transmission volume and total outbound traffic are as follows:

[0206] ,

[0207] Total outflow ,

[0208] Weight ,

[0209] calculate and :

[0210] Compute Node arrive Traffic density:

[0211] ,

[0212] Compute Node arrive Traffic density:

[0213] ,

[0214] Determine significant traffic density relationships

[0215] Set a threshold =0.5, based on this threshold, we can judge:

[0216] Greater than ,therefore and There is a significant traffic density relationship between them.

[0217] Less than ,therefore and The traffic density relationship between them is not significant.

[0218] See also Figure 7 ,The specific steps for obtaining abnormal behavior indicators are:

[0219] S311: Analyze the nodes in the key dependency graph, calculate the current data integrity, and set each node The current data integrity of , the amount of real-time data obtained through real-time monitoring and expected data volume Calculation, refer to the following formula:

[0220] ,

[0221] in, For Node The current data integrity, is the adjustment parameter, For Node The amount of real-time data at the current time t, For Node The expected amount of data at the current time t;

[0222] S312: Calculate the historical data flow integrity based on the average integrity of the data of the same node in the past cycle, with reference to the following formula:

[0223] ,

[0224] in, For historical data flow integrity, It is a parameter that enhances the influence of historical data;

[0225] S313: Comparing the data integrity Integrity with historical data flow , calculate the deviation value , is the difference between the current data integrity and the historical data stream integrity, and the reference formula is as follows:

[0226] ,

[0227] in, is the adjustment factor for the deviation value, is the deviation between the current data integrity and the historical data stream integrity, through Parameters are adjusted;

[0228] S314: Setting the abnormality determination threshold ,like , it is considered abnormal, the deviation value is recorded, and the abnormal behavior index is obtained.

[0229] Parameter Description:

[0230] :node The amount of real-time data at the current time t; :node The expected amount of data at the current time t; , : An index used to adjust the ratio of real-time and historical data so that fluctuations or continuous consistency of data are more significant in the analysis; : Adjust the coefficient of the final deviation impact to amplify or reduce the final deviation value to facilitate capturing anomalies; : Anomaly determination threshold, used to determine when a deviation value indicates an anomaly.

[0231] Example demonstration:

[0232] Suppose there is a node , the amount of real-time data at a certain moment 120MB, expected data volume is 150MB, the cumulative real-time data volume in the past 10 days is 1100MB, and the cumulative expected data volume is 1300MB; set =1.1, =1.1, =1.5.

[0233] Calculation process:

[0234] Calculate current data integrity :

[0235] ,

[0236] Calculate historical data flow integrity :

[0237] ,

[0238] Calculate the deviation value :

[0239] ,

[0240] Determine abnormality:

[0241] if (Assumption =10), then -12.03 indicates an anomaly.

[0242] See also Figure 8 , the specific steps to establish a policy adjustment log and store container cloud data are as follows:

[0243] S411: Collect abnormal behavior indicators and apply network slicing adjustment formula:

[0244] ,

[0245] Update configuration;

[0246] in, Represents the new network slice configuration, For the original configuration, The abnormal behavior indicator processing result, , , It is to adjust the parameters to reflect the intensity and adaptability of the impact of abnormal indicators on the configuration, and obtain the pre-adjusted network slice configuration;

[0247] S412: Verify the adaptability and stability of the pre-adjusted configuration, using the effect verification formula:

[0248] ,

[0249] in, Calculate the relative change of the configuration adjustment and establish the adjusted network slice configuration;

[0250] S413: According to the adjustment process and verification results, record the adjustment details and verification data of each configuration, establish a policy adjustment log, and store container cloud data.

[0251] Explanation and derivation of network slice adjustment formula:

[0252] formula: ,

[0253] Parameter explanation:

[0254] : The original network slice configuration, assumed to be 100 (the unit can be any measure of network resources, such as bandwidth Mbps);

[0255] : Adjusted network slice configuration;

[0256] : The result of abnormal behavior indicator processing, assumed to be 5 (which can be understood as the abnormal degree of network traffic, such as the percentage of sudden increase in traffic);

[0257] , , : Adjust the parameters, assuming =0.2, =0.1, =0.05, these parameters define the sensitivity and adaptability of abnormal indicators to the impact of network configuration.

[0258] Derivation process and examples:

[0259] First calculate the numerator:

[0260] ,

[0261] ,

[0262] ,

[0263] The sum of the numerators = 1.0 + 0.179 = 1.179.

[0264] Next, calculate the denominator:

[0265] ,

[0266] ,

[0267] The sum of the denominators = 1 + 0.00035 ≈ 1.00035.

[0268] Substitute the results for the numerator and denominator into the adjustment formula:

[0269] ,

[0270] It shows that under the given abnormal behavior indicator, the original configuration increases from 100 to about 117.87 to cope with the impact of abnormal behavior.

[0271] Explanation and derivation of configuration adjustment effect verification formula

[0272] formula: ,

[0273] Parameter explanation:

[0274] : The original configuration, same as above, is 100;

[0275] : The adjusted configuration obtained from the previous step is 117.865.

[0276] Derivation process and examples:

[0277] calculate :

[0278] ,

[0279] Calculate the effect verification value :

[0280] ,

[0281] This represents an increase of approximately 17.9% in the adjusted configuration over the original configuration, thus confirming the amount of change in the configuration.

[0282] A business data management system based on container cloud, which is used to execute the above-mentioned business data management method based on container cloud, includes:

[0283] The data collection and analysis module uses container cloud monitoring to capture business data flows and network status in real time, combines inter-node communication data, calculates data transmission frequency and duration, and screens active and inactive data channels to integrate them into network behavior data graphs.

[0284] The network behavior analysis module analyzes the network behavior data graph to identify the data access frequency and traffic size between multiple nodes, identify and mark nodes with high traffic and key dependent nodes, establish the relationship between network traffic and node dependency, and obtain the key dependency graph;

[0285] The abnormal behavior analysis module compares the integrity of the current node data with the historical records based on the key dependency graph, identifies abnormal increases and decreases in data flow and changes in flow direction, records and analyzes deviation values, and generates abnormal behavior indicators;

[0286] The security policy adjustment module optimizes and adjusts the network slice configuration based on abnormal behavior indicators, updates security measures for data storage and transmission, records operations and adjustment results, establishes policy adjustment logs, and performs container cloud data storage.

[0287] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A business data management method based on container cloud, characterized in that: The following steps are involved: Collect business data flows and network status in the container cloud, analyze data transmission security records between multiple nodes, calculate data transmission frequency and duration, filter active and inactive data channels, and obtain network behavior patterns; Based on the network behavior pattern, the data access pattern between nodes is analyzed, the nodes with high traffic and key dependencies are identified, and the key dependency graph is obtained by establishing the relationship between network traffic and node dependency; Analyze the key dependency graph, identify abnormal increases and decreases and changes in flow direction in the data flow by comparing the deviations between the current data integrity of multiple nodes and the historical data flow, record the deviation values, and obtain abnormal behavior indicators; Based on the abnormal behavior indicators, adjust the affected network slice configuration, reset data storage security measures, update the security policy of the network slice, record the operations and results of multiple adjustments, establish a policy adjustment log, and perform container cloud data storage; The step of calculating the data transmission frequency and duration is specifically as follows: Collect the business data flow and network status of the container cloud, filter the metadata of the data transmission record, extract the timestamp and duration information of the data transmission from the filtered metadata, calculate the transmission start and end time difference of each data channel, and obtain the duration of each data channel ; Calculate the data transmission frequency, F is the number of data transmissions completed per unit time, refer to the formula as follows: , Then summarize the data transmission frequency and duration of each data channel; Where F is the data transmission frequency, represents the amount of data transmitted for the i-th time, T is the total observation time, is the duration of the data channel, is the weight coefficient, is a coefficient to ensure the stability of calculation, is the attenuation coefficient, is exponential decay; The steps of screening active and inactive data channels are specifically as follows: Using the calculation results of the data transmission frequency and duration, an activity threshold is set. and inactive threshold ,in, Higher than average data transmission frequency, Lower than average data transmission frequency; For each data channel, compare the data transmission frequency F. If F is greater than , it is marked as an active channel. If F is less than , it is marked as an inactive channel; The marking results of the data channels are counted to obtain the number of active and inactive data channels in the network, with reference to the following formula: , , in, is the number of active data channels, is the number of inactive data channels, represents the data transmission frequency of the i-th data channel, is the active threshold, which is set to a value higher than the average data transmission frequency. is the inactivity threshold, which is set to a value lower than the average data transmission frequency. , is the improved indicator function, , is the flexibility parameter; The steps for obtaining the network behavior pattern are specifically as follows: Analyze the active and inactive data channels, monitor the behavior characteristics and transmission modes of the differentiated data channels, perform time series analysis on the active data channels, extract key activity time periods and peak transmission periods, and analyze frequency data on the inactive data channels; Based on the time series and frequency data, the network behavior model is constructed according to the following formula: , Among them, P represents the behavior pattern of the entire network, and They are frequency indicators of active and inactive data channels, , is the weight factor, , is the scaling factor, , are the growth rate and decay rate, respectively. , is the frequency index of active (a) and inactive (n) data channels, , is an exponential function; The steps of identifying high traffic and key dependent nodes are specifically as follows: Based on the network behavior pattern, the data access pattern is identified and the total amount of data transmission between nodes is calculated. Suppose multiple nodes are and , the data transmission volume is , for each node , calculate the total outflow and total inflow, refer to the following formula: , , in, For Node The total outflow rate, For Node The total inflow, , Representative Node To Node and nodes To Node The amount of data transfer, and is the weight assigned according to the quality or importance of the connection between nodes. and represents the distance between nodes, is the attenuation coefficient; By comparing each node and , set the threshold Determine the high traffic nodes. or , then the node are classified as high traffic nodes, where Set thresholds for node traffic; By analyzing the total amount of data transmission between nodes, the node The degree of dependence on other nodes is calculated using the dependency relationship, with the reference formula as follows: , What you get The high dependency is represented as a key dependency node; in, For Node The dependence of The weight representing the strength of dependency between nodes; The steps for obtaining the key dependency graph are specifically as follows: By using the high-traffic nodes and key dependent nodes information, the interactive data traffic between nodes is analyzed to determine the traffic-intensive relationship, and the traffic contribution of each node pair to other nodes is calculated, referring to the following formula: , in, Represents a slave node To Node A measure of the traffic density between Is a node arrive The amount of data transfer, Is a node The total outflow rate, is a weight factor, which is assigned according to the traffic contribution and node importance; By comparing each node pair value, set a threshold ,like , then the node and There is a traffic-intensive relationship between them, and the relationship between network traffic and node dependency is established to obtain the key dependency graph; The steps for obtaining the abnormal behavior indicator are specifically as follows: Analyze the nodes in the key dependency graph, calculate the current data integrity, and set each node The current data integrity of , the amount of real-time data obtained through real-time monitoring and expected data volume Calculation, refer to the following formula: , in, For Node The current data integrity, is the adjustment parameter, For Node The amount of real-time data at the current time t, For Node The expected amount of data at the current time t; Based on the average integrity of the data of the same node in the past period, the historical data flow integrity is calculated according to the following formula: , in, For historical data flow integrity, It is a parameter that enhances the influence of historical data; Compare the data integrity Integrity with historical data flow , calculate the deviation value , is the difference between the current data integrity and the historical data stream integrity, and the reference formula is as follows: , in, is the adjustment factor for the deviation value, is the deviation between the current data integrity and the historical data stream integrity, through Parameters are adjusted; Set the abnormality judgment threshold ,like , it is considered abnormal, the deviation value is recorded, and the abnormal behavior index is obtained; The steps of establishing a policy adjustment log and storing container cloud data are specifically as follows: Collect abnormal behavior indicators and apply the network slice adjustment formula: , Update configuration; in, Represents the new network slice configuration, For the original configuration, is the abnormal behavior indicator processing result, , , It is to adjust the parameters to reflect the intensity and adaptability of the impact of abnormal indicators on the configuration, and obtain the pre-adjusted network slice configuration; Verify the adaptability and stability of the pre-adjusted configuration using the effect verification formula: , in, Calculate the relative change of the configuration adjustment and establish the adjusted network slice configuration; Based on the adjustment process and verification results, record the adjustment details and verification data of each configuration, establish a policy adjustment log, and store container cloud data.

2. A business data management system based on container cloud, characterized in that: The system for executing the container cloud-based business data management method according to claim 1 comprises: The data collection and analysis module uses container cloud monitoring to capture business data flows and network status in real time, combines inter-node communication data, calculates data transmission frequency and duration, and screens active and inactive data channels to integrate them into network behavior data graphs. The network behavior analysis module identifies the data access frequency and traffic size between multiple nodes by analyzing the network behavior data graph, identifies and marks nodes with high traffic and key dependent nodes, establishes the relationship between network traffic and node dependency, and obtains a key dependency graph; The abnormal behavior analysis module compares the integrity of the current node data with the historical records based on the key dependency graph, identifies abnormal increases and decreases in data flow and changes in flow direction, records and analyzes deviation values, and generates abnormal behavior indicators; The security policy adjustment module optimizes and adjusts the network slice configuration based on the abnormal behavior indicators, updates the security measures for data storage and transmission, records operations and adjustment results, establishes a policy adjustment log, and performs container cloud data storage.