Cloud platform-based computing power data security management method and system

Abstract

The invention relates to the field of mobile communication, and discloses a computing power data security management method and system based on a cloud platform, and the method comprises the steps: firstly collecting user access data: obtaining the access data of each user through a cloud platform background, the user access data comprising user login I P, user request frequency and user passing identity verification time; analyzing user access security; receiving user access data, establishing a risk level evaluation for each access user, and executing an access control strategy for the access user according to a risk level evaluation result; analyzing and judging the risk of the cloud platform: inputting historical data and real-time access data of the cloud platform into a cloud platform state analysis model, and judging the current dangerous state of the cloud platform; the historical data of the cloud platform comprises historical access amount, historical request frequency and historical high-risk access user amount; and feedback regulation and control: adjusting the security mode of the cloud platform according to the fed-back current dangerous state of the cloud platform.

Description

Technical Field

[0001] The present invention relates to the field of mobile communications, and in particular to a computing power data security management method and system based on a cloud platform. Background Art

[0002] The computing power data of the cloud platform is an important part of cloud computing services. It refers to the use of cloud computing technology to distribute computing tasks on a resource pool composed of a large number of computers, so that various application systems can obtain computing power, storage space and various software services as needed. With the continuous development of cloud computing technology, the computing power data of the cloud platform has become an important support for enterprises to carry out digital transformation. Through the computing power data of the cloud platform, enterprises can achieve flexible expansion, efficient management, pay-as-you-go and other service models, thereby improving the production efficiency and innovation capabilities of enterprises.

[0003] The computing power data of the cloud platform mainly comes from distributed computing technology, which can efficiently integrate a large number of computing resources, realize load balancing and dynamic resource scheduling, and enable various application systems to obtain the best computing resources and storage services as needed. At the same time, the computing power data of the cloud platform is also highly reliable, secure and scalable, which can ensure the security and privacy protection of enterprise data, as well as achieve rapid response and efficient processing in various business scenarios.

[0004] At present, the computing power data of cloud platforms have been widely used in various industries, such as finance, medical care, education, manufacturing, etc. Through the computing power data of cloud platforms, enterprises can quickly build various application systems, reduce IT costs and maintenance costs, and improve business competitiveness and market response speed. At the same time, the computing power data of cloud platforms can also provide enterprises with flexible scalability and innovation space, promoting the digital transformation and upgrading of enterprises.

[0005] However, a large amount of user data is stored in the cloud platform. If data security measures are not in place, it may lead to data leakage, threatening user privacy and corporate secrets. In order to protect the data security of the cloud platform, many companies ignore the user's access needs and insist on using the first verification plus the second verification identity authentication method for all user access. Although it can improve data security and avoid data leakage, it obviously has the problem of low access efficiency for different user groups and needs. Summary of the invention

[0006] The purpose of the present invention is to provide a cloud platform-based computing power data security management method and system to solve the above technical problems.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A computing power data security management method based on a cloud platform includes the following steps:

[0009] Step S11: Collecting user access data: Obtaining each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency, and the time when the user passed the identity authentication;

[0010] Step S12: Analyze user access security; receive user access data, and establish a risk level evaluation for each access user, and execute access control policies for the access users according to the results of the risk level evaluation;

[0011] Step S13: Analyze and determine the cloud platform risk: input the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform; the cloud platform historical data includes historical access volume, historical request frequency and historical high-risk access user volume;

[0012] Step S14: Feedback control: adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

[0013] As a further technical solution, the specific method for evaluating the risk level of each accessing user includes:

[0014] By formula:

[0015]

[0016] Calculate and obtain the danger level Q;

[0017] The calculated danger level Q is compared with the preset threshold Q 阈 Make comparisons;

[0018] If Q>Q 阈 , then the access user's risk level is judged to be 1;

[0019] Otherwise, the access user's risk level is judged to be 0;

[0020] When the danger level of the access user is 1, the access user is judged to be in a high-risk state, and when the access user level is 0, the access user is judged to be in a low-risk state;

[0021] Where σ is the security coefficient of the access user IP. When the access user IP is identified as safe, σ = 0, otherwise, σ = 1; q i is the access frequency of the user visiting in the i-th time period; q 0 is the access frequency reference value of the user visiting the i-th time period, which is selected and determined according to the average reference value of the historical access frequency data; T i is the time when the user who accesses the i-th time period passes the authentication, T 0is the reference value of the time it takes for the visiting user to pass the identity authentication in the i-th time period, which is determined based on the average time it takes for the user to pass the identity authentication. Y is the request frequency trend coefficient during the visiting user's access process, and X is the time trend coefficient for the visiting user to pass the identity authentication during the access process.

[0022] As a further technical solution, the method for obtaining the request frequency trend coefficient Y is:

[0023] By formula: get,

[0024] Among them, the number of requests for a user to access the process is m+1, j∈[1,m], q j The interval between each request of the access user. is the average interval duration of all requests during a visit by a visiting user; θ is the conversion coefficient, which is determined based on historical experience data and experimental data.

[0025] As a further technical solution, the method for obtaining the identity authentication time trend coefficient is:

[0026] By formula: Get X;

[0027] Among them, the number of identity authentications performed by the accessing user during one access is R+1, k∈[1, R], q k The interval between each authentication of the access user. is the average duration of all identity authentications during a user's visit; μ is the conversion coefficient, which is determined based on historical experience data and experimental data.

[0028] As a further technical solution, the specific method of executing the access control policy on the access user according to the result of the risk level evaluation is:

[0029] Get the danger level Q of all daily access users on the cloud platform;

[0030] Set the authentication mode for access users with a risk level of 0 to log in after the first authentication.

[0031] The identity verification mode for access users with a risk level of 1 is set to include first verification plus second verification;

[0032] The number of access users with a danger level of 1 is fitted with a curve of the number of access users with a danger level of 1, with the danger level Q as the horizontal axis and the number of access users corresponding to each danger level Q as the vertical axis;

[0033] Divide the curve of the number of dangerous level access users into several sub-areas according to the preset dangerous level interval, and obtain the area of ​​each sub-area;

[0034] By formula Calculate the user access risk factor D i , when the interval where the maximum area value of the curve of the number of dangerous level access users is located is within the preset dangerous level interval, then γ=1, otherwise, γ=0; is the sum of the areas of all sub-regions, G max is the maximum area value of the sub-region.

[0035] As a further technical solution, the high-risk user access risk coefficient D i Access to the preset danger zone Make comparisons;

[0036] like The authentication mode for high-risk access users is improved, adding three SMS verifications on top of the first and second verifications;

[0037] like The login privileges of high-risk users are suspended, and the identity authentication mode of high-risk users is turned off;

[0038] like Maintain the authentication mode for high-risk access users;

[0039] Get the access risk factor D of the user within a work cycle i , if the access risk factor D of high-risk users is i Lower than the average value of the previous working cycle This reduces the authentication mode for high-risk access users.

[0040] As a further technical solution, the process of establishing the cloud platform status analysis model is as follows:

[0041] Based on historical access data, obtain the daily access volume change curve F(t), the daily access user IP number change curve S(t) and the daily identity authentication time change curve H(t);

[0042] By formula: Calculate E Status ; and obtain the cloud platform status analysis model E Status (t);

[0043] in,

[0044]

[0045] Among them, ΔF, ΔS and ΔH are respectively 1 , t2 ], the reference value of daily visit volume, the reference value of daily visitor IP number and the reference value of daily identity verification time, α 1 , α 2 and α 3 is the preset weight coefficient, which is selected and determined based on historical experience data; F 0 (t) is the standard curve of daily visits changing with time, S 0 (t) is the standard curve of the number of daily visiting user IPs changing over time, H 0 (t) is the standard curve of the daily identity authentication time varying with time, ε is the reference value of the number of risky users, N i is the number of high-risk users visiting daily, F i is the actual number of visits, n is the historical time period [t 1 , t 2 ] is the total number, Δp is the preset reference value, which is selected and determined based on historical experience data.

[0046] As a further technical solution, the specific method of conducting dangerous state analysis through the cloud platform state analysis model is as follows:

[0047] S131: The cloud platform dangerous state coefficient E obtained through the cloud platform state analysis model Status (t n ) and the preset dangerous state coefficient threshold interval Make comparisons;

[0048] like Then suspend the access and login to the cloud platform;

[0049] like Then keep the cloud platform access and login normally;

[0050] like This will improve the security protection level of the cloud platform.

[0051] A computing power data security management system based on a cloud platform, comprising:

[0052] The data collection module is used to collect user access data and obtain each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency and the time when the user passes the identity authentication;

[0053] The user access analysis module is used to analyze the user access security, receive user access data, and establish a risk level evaluation for each access user, and execute access control policies on the access users according to the results of the risk level evaluation;

[0054] The cloud platform risk assessment module is used to analyze and determine the risks of the cloud platform. It inputs the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform. The cloud platform historical data includes historical access volume, historical request frequency, and historical high-risk access user volume.

[0055] The control module is used to adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

[0056] Beneficial effects of the present invention:

[0057] (1) The risk level of a single access user is divided according to the IP address, user request frequency and user identity verification time of the cloud platform. The user identity verification time can be used to determine whether the operation is a robot, which assists in determining the risk level of the access user. At the same time, the cloud platform's historical access volume, historical request frequency and historical high-risk access user volume can be used to conduct an overall analysis of the risk level of each access user. The current risk status of the cloud platform can be obtained through the analysis results, so as to adjust the cloud platform security mode and adjust the cloud platform security mode in a timely manner to ensure the data security of the cloud platform.

[0058] (2) The danger level of the access user is obtained based on historical access data, and access control policies are implemented on the access user based on the results of the danger level evaluation, so as to establish different identity authentication modes for different access user danger levels, and the identity authentication mode can be adaptively adjusted to achieve the purpose of changing the number of identity authentications, shortening the identity authentication time for users with different needs, and improving the efficiency of users calling cloud platform data. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further described below in conjunction with the accompanying drawings.

[0060] Figure 1 is a step diagram of the present invention;

[0061] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0062] 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.

[0063] See also Figure 1-Figure 2 As shown, the present invention is a computing power data security management method based on a cloud platform, comprising the following steps:

[0064] Step S11: Collecting user access data: Obtaining each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency, and the time when the user passed the identity authentication;

[0065] Step S12: Analyze user access security; receive user access data, and establish a risk level evaluation for each access user, and execute access control policies for the access users according to the results of the risk level evaluation;

[0066] Step S13: Analyze and determine the cloud platform risk: input the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform; the cloud platform historical data includes historical access volume, historical request frequency and historical high-risk access user volume;

[0067] Step S14: Feedback control: adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

[0068] In this embodiment, the risk level of a single accessing user is divided according to the IP address of the cloud platform, the frequency of user requests and the time of user identity verification. The user identity verification time can be used to determine whether the operation is a robot, thereby assisting in determining the danger level of the accessing user. At the same time, an overall analysis can be conducted based on the historical access volume, historical request frequency and historical high-risk access user volume of the cloud platform and the danger level of each accessing user. The current danger status of the cloud platform can be obtained through the analysis results to adjust the cloud platform security mode, and the cloud platform security mode can be adjusted in a timely manner to ensure the data security of the cloud platform.

[0069] The specific method for evaluating the risk level of each access user includes:

[0070] By formula: Calculate and obtain the danger level Q;

[0071] The calculated danger level Q is compared with the preset threshold Q 阈 Compare; if Q>Q 阈 , then the access user's risk level is judged to be 1; otherwise, the access user's risk level is judged to be 0; when the access user's risk level is 1, the access user is judged to be in a high-risk state, and when the access user's risk level is 0, the access user is judged to be in a low-risk state;

[0072] Where σ is the security coefficient of the access user IP. When the access user IP is identified as safe, σ = 0, otherwise, σ = 1; q i is the access frequency of the user visiting in the i-th time period; q 0is the access frequency reference value of the user visiting the i-th time period, which is selected and determined according to the average reference value of the historical access frequency data; T i is the time when the user who accesses the i-th time period passes the authentication, T 0 is the reference value of the time it takes for the visiting user to pass the identity authentication in the i-th time period, which is determined based on the average time it takes for the user to pass the identity authentication. Y is the request frequency trend coefficient during the visiting user's access process, and X is the time trend coefficient for the visiting user to pass the identity authentication during the access process.

[0073] The requested frequency trend coefficient is calculated by the formula: Get, where the number of requests for a user to access the process is m+1, j∈[1,m], q j The interval between each request of the access user. is the average interval duration of all requests during a visit by a visiting user; θ is the conversion coefficient, which is determined based on historical experience data and experimental data;

[0074] The authentication time trend coefficient is calculated by the formula: Get, where the number of identity authentications performed by the accessing user during one access is R+1, k∈[1, R], q k The interval between each authentication of the access user. is the average duration of all identity authentications during a user's visit; μ is the conversion coefficient, which is determined based on historical experience data and experimental data.

[0075] The specific method of implementing access control policies on access users based on the results of risk level assessment is as follows:

[0076] Get the danger level Q of all daily access users on the cloud platform;

[0077] Set the authentication mode for access users with a risk level of 0 to log in after the first authentication.

[0078] The identity authentication mode for access users with a risk level of 1 is set to include first verification plus second verification;

[0079] The number of access users with a danger level of 1 is fitted with a curve of the number of access users with a danger level of 1, with the danger level Q as the horizontal axis and the number of access users corresponding to each danger level Q as the vertical axis;

[0080] Divide the curve of the number of dangerous level access users into several sub-areas according to the preset dangerous level interval, and obtain the area of ​​each sub-area;

[0081] By formula The user access risk coefficient is calculated. When the interval where the maximum area value of the curve of the number of dangerous level access users is located is within the preset dangerous level interval, γ=1, otherwise, γ=0; is the sum of the areas of all sub-regions, G max is the maximum area value of the sub-region;

[0082] The high-risk user access risk factor D i Access to the preset danger zone Make comparisons;

[0083] like The authentication mode for high-risk access users is improved, adding three SMS verifications on top of the first and second verifications;

[0084] like The login privileges of high-risk users are suspended, and the identity authentication mode of high-risk users is turned off;

[0085] like The authentication mode for high-risk access users is maintained;

[0086] Get the access risk factor D of the user within a work cycle i , if the access risk factor D of high-risk users is i Lower than the average value of the previous working cycle The identity verification mode of high-risk access users is reduced. In this embodiment, the risk level of each access user is evaluated by combining the IP address of the cloud platform, the frequency of user requests, and the time of user identity verification, and a simulation curve is established according to the number of users evaluated by the risk level and the risk level. The risk coefficient of high-risk users' access is obtained by analyzing and processing the curve. The risk coefficient of high-risk users' access is compared with the preset risk interval, and the access control strategy is executed according to the comparison result, that is, the identity verification is increased, maintained, or suspended on the basis of the original identity verification mode, thereby taking into account both data security and shortening the time of identity verification, and improving the efficiency of users' access to the cloud platform.

[0087] The process of establishing the cloud platform status analysis model is as follows:

[0088] Based on historical access data, obtain the daily access volume change curve F(t), the daily access user IP number change curve S(t) and the daily identity authentication time change curve H(t);

[0089] By formula: Calculate E Status ; It is also possible to obtain the cloud platform status analysis model E Status (t);

[0090] in,

[0091]

[0092] Among them, ΔF, ΔS and ΔH are respectively 1 , t 2 ], the reference value of daily visit volume, the reference value of daily visitor IP number and the reference value of daily identity verification time, α 1 , α 2 and α 3 is the preset weight coefficient, which is selected and determined based on historical experience data; F 0 (t) is the standard curve of daily visits changing with time, S 0 (t) is the standard curve of the number of daily visiting user IPs changing over time, H 0 (t) is the standard curve of the daily identity authentication time varying with time, ε is the reference value of the number of risky users, N i is the number of high-risk users visiting daily, F i is the actual number of visits, n is the historical time period [t 1 , t 2 ] is the total number, Δp is the preset reference value, which is selected and determined based on historical experience data.

[0093] Specific methods for analyzing dangerous states through the cloud platform state analysis model:

[0094] S131: The cloud platform dangerous state coefficient E obtained through the cloud platform state analysis model Status (t n ) and the preset dangerous state coefficient threshold interval Make comparisons;

[0095] like Then suspend the access and login to the cloud platform;

[0096] like Then keep the cloud platform access and login normally;

[0097] like This will improve the security protection level of the cloud platform.

[0098] In this embodiment, the overall danger state of the cloud platform is comprehensively judged by the historical deviation status of the number of access users, the historical deviation status of the IP data of access users, the historical deviation status of the identity authentication time and the number of high-risk users, so as to quickly make a corresponding response according to the danger state of the cloud platform; when the cloud platform danger state coefficient is lower than the preset danger state coefficient threshold interval, indicating that the overall danger state of the cloud platform is in a relatively low danger state, and the data of the cloud platform is in a safe state, then the normal cloud platform access login is maintained; when the cloud platform danger state coefficient exceeds the preset danger state coefficient threshold interval, indicating that the overall danger state of the cloud platform is in a relatively high danger state, and it is facing the risk of data leakage at any time, then the cloud platform access login is suspended; when the cloud platform danger state coefficient is in the preset danger state coefficient threshold interval, indicating that the overall danger state of the cloud platform is in a normal state, and the cloud platform data may be at risk of leakage, then the cloud platform security protection level is increased to reduce the cloud platform data risk.

[0099] A computing power data security management system based on a cloud platform, comprising:

[0100] The data collection module is used to collect user access data and obtain each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency and the time when the user passes the identity authentication;

[0101] The user access analysis module is used to analyze the user access security, receive user access data, and establish a risk level evaluation for each access user, and execute access control policies on the access users according to the results of the risk level evaluation;

[0102] The cloud platform risk assessment module is used to analyze and determine the risks of the cloud platform. It inputs the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform. The cloud platform historical data includes historical access volume, historical request frequency, and historical high-risk access user volume.

[0103] The control module is used to adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

[0104] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A cloud platform-based computing power data security management method, characterized in that: The following steps are involved: Step S11: Collecting user access data: Obtaining each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency, and the time when the user passed the identity authentication; Step S12: Analyze user access security; receive user access data, and establish a risk level evaluation for each access user, and execute access control policies for the access users according to the results of the risk level evaluation; Step S13: Analyze and determine the cloud platform risk: input the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform; the cloud platform historical data includes historical access volume, historical request frequency and historical high-risk access user volume; Step S14: Feedback control: adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

2. The cloud platform-based computing power data security management method according to claim 1 is characterized in that: The specific methods for evaluating the risk level of each access user include: By formula: Calculate and obtain the danger level Q; The calculated danger level Q is compared with the preset threshold Q 阈 Make comparisons; If Q>Q 阈 , then the access user's risk level is judged to be 1; Otherwise, the access user's risk level is judged to be 0; When the danger level of the access user is 1, the access user is judged to be in a high-risk state, and when the access user level is 0, the access user is judged to be in a low-risk state; Where σ is the security coefficient of the access user IP. When the access user IP is identified as safe, σ = 0, otherwise, σ = 1; q i is the access frequency of the user visiting the i-th time period; q0 is the access frequency reference value of the user visiting the i-th time period, which is selected and determined according to the average reference value of the historical access frequency data; T i is the time for the visiting user to pass the identity authentication in the i-th time period, T0 is the reference value of the time for the visiting user to pass the identity authentication in the i-th time period, which is determined based on the average time for the user to pass the identity authentication, Y is the request frequency trend coefficient during the visiting user's access process, and X is the time trend coefficient for passing the identity authentication during the visiting user's access process.

3. The cloud platform-based computing power data security management method according to claim 2 is characterized in that: The method for obtaining the request frequency trend coefficient Y is: By formula: get, Among them, the number of requests for a user to access the process is m+1, j∈[1,m], q j The interval between each request of the access user. is the average interval duration of all requests during a visit by a visiting user; θ is the conversion coefficient, which is determined based on historical experience data and experimental data.

4. The cloud platform-based computing power data security management method according to claim 2 is characterized in that: The method for obtaining the identity authentication time trend coefficient is: By formula: Get X; Among them, the number of identity authentications performed by the accessing user during one access is R+1, k∈[1, R], q k The interval between each authentication of the access user. is the average duration of all identity authentications during a user's visit; μ is the conversion coefficient, which is determined based on historical experience data and experimental data.

5. The cloud platform-based computing power data security management method according to claim 2 is characterized in that: The specific method of implementing access control policies on access users based on the results of risk level assessment is as follows: Get the danger level Q of all daily access users on the cloud platform; Set the authentication mode for access users with a risk level of 0 to log in after the first authentication. The identity verification mode for access users with a risk level of 1 is set to include first verification plus second verification; The number of access users with a danger level of 1 is fitted with a curve of the number of access users with a danger level of 1, with the danger level Q as the horizontal axis and the number of access users corresponding to each danger level Q as the vertical axis; Divide the curve of the number of dangerous level access users into several sub-areas according to the preset dangerous level interval, and obtain the area of ​​each sub-area; By formula Calculate the user access risk factor D i , when the interval where the maximum area value of the curve of the number of dangerous level access users is located is within the preset dangerous level interval, then γ=1, otherwise, γ=0; is the sum of the areas of all sub-regions, G max is the maximum area value of the sub-region.

6. The cloud platform-based computing power data security management method according to claim 5 is characterized in that: The high-risk user access risk factor D i Access to the preset danger zone Make comparisons; like The authentication mode for high-risk access users is improved, adding three SMS verifications on top of the first and second verifications; like The login privileges of high-risk users are suspended, and the identity authentication mode of high-risk users is turned off; like Maintain the authentication mode for high-risk access users; Get the access risk factor D of the user within a work cycle i , if the access risk factor D of high-risk users is i Lower than the average value of the previous working cycle This reduces the authentication mode for high-risk access users.

7. The cloud platform-based computing power data security management method according to claim 1 is characterized in that: The process of establishing the cloud platform status analysis model is as follows: Based on historical access data, obtain the daily access volume change curve F(t), the daily access user IP number change curve S(t) and the daily identity authentication time change curve H(t); By formula: Calculate E Status ; and obtain the cloud platform status analysis model E Status (t); in, Among them, ΔF, ΔS and ΔH are the reference values ​​of daily visits, daily visiting user IP number and daily identity verification time in the time period [t1, t2] respectively, α1, α2 and α3 are preset weight coefficients, which are selected and determined according to historical experience data; F0(t) is the standard curve of daily visits changing with time, S0(t) is the standard curve of daily visiting user IP number changing with time, H0(t) is the standard curve of daily identity verification time changing with time, ε is the reference value of the number of risky users, N i is the number of high-risk users visiting daily, F i is the actual number of visits, n is the total number of historical time periods [t1, t2], and Δp is a preset reference value, which is selected and determined based on historical experience data.

8. The cloud platform-based computing power data security management method according to claim 2 is characterized in that: Specific methods for analyzing dangerous states through the cloud platform state analysis model: S131: The cloud platform dangerous state coefficient E obtained through the cloud platform state analysis model Status (t n ) and the preset dangerous state coefficient threshold interval Make comparisons; like Then suspend the access and login to the cloud platform; like Then keep the cloud platform access and login normally; like This improves the security protection level of the cloud platform.

9. A cloud platform-based computing power data security management system, the management system is applicable to the cloud platform-based computing power data security management method according to any one of claims 1 to 8, characterized in that: include: The data collection module is used to collect user access data and obtain each user's access data through the cloud platform background. The user access data includes the user's login IP, user request frequency and the time when the user passes the identity authentication; The user access analysis module is used to analyze the user access security, receive user access data, and establish a risk level evaluation for each access user, and execute access control policies on the access users according to the results of the risk level evaluation; The cloud platform risk assessment module is used to analyze and determine the risks of the cloud platform. It inputs the cloud platform historical data and real-time access data into the cloud platform status analysis model to determine the current dangerous status of the cloud platform. The cloud platform historical data includes historical access volume, historical request frequency, and historical high-risk access user volume. The control module is used to adjust the cloud platform security mode according to the feedback of the current dangerous state of the cloud platform.

Images