Computer software security encryption and decryption management system

Abstract

The invention discloses a computer software security encryption and decryption management system, which belongs to the technical field of computer software, and comprises a collaborative computing platform, a collaborative management module, a system management module and a resource monitoring management module, and the collaborative management module, the system management module and the resource monitoring management module are all in network connection with the collaborative computing platform. And the collaborative management module comprises an encryption and decryption mechanism cracking module and an encryption and decryption management module. In the invention, through a heuristic algorithm and a machine learning algorithm, the system can dynamically allocate and adjust the tasks to the FPGA or the GPU according to the current resource use condition and task requirements, so that not only can the load balance between the FPGA server and the GPU be ensured, but also the condition that some equipment is overloaded and other equipment is idle can be avoided, and the reliability of the system is improved. According to the method, hardware resources of the FPGA server and the GPU can be utilized more effectively, the overall efficiency of the system is improved, the execution speed of tasks can be increased, and the reliability of the system and the satisfaction degree of users can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer software, and in particular relates to a computer software security encryption and decryption management system. Background Art

[0002] With the development of information technology, encryption and decryption technology is increasingly used in information systems. The persons under investigation by law enforcement agencies in the process of collecting electronic evidence generally begin to use encryption and decryption technology to protect sensitive documents, and it is becoming increasingly difficult to obtain useful evidence from them. In the process of supporting judicial practice, we found that there are many types of encryption and decryption data set using the encryption and decryption functions of application software; The prior art discloses some invention patents in the field of computer software technology, among which the invention patent with application number CN109684792B discloses a computer software security encryption and decryption management system, which belongs to the field of computer software technology, including a collaborative computing platform, a collaborative management module and a system management module, wherein the collaborative management module and the system management module are both connected to the collaborative computing platform network, the collaborative management module includes an encryption and decryption mechanism cracking module and an encryption and decryption management module, the encryption and decryption mechanism cracking module includes a cluster hybrid heterogeneous processing architecture and a task creation module, the cluster hybrid heterogeneous processing architecture is connected to the task creation module, and the cluster hybrid heterogeneous processing architecture includes a task master control module and an FPGA computing management server. The present invention effectively improves the computing power of large-scale encryption and decryption mechanisms through a cluster hybrid heterogeneous processing architecture, and selects the required existing mainstream password algorithm or customized password algorithm according to the actual encryption and decryption mechanism through an algorithm self-selection module.

[0003] Although existing technologies can improve the computing power of large-scale encryption and decryption mechanisms through cluster hybrid heterogeneous processing architectures, they cannot dynamically allocate and adjust tasks to FPGAs or GPUs based on current resource usage and task requirements, resulting in an imbalance in the load between some FPGA servers and GPUs.

[0004] Based on this, the present invention designs a computer software security encryption and decryption management system to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to propose a computer software security encryption and decryption management system to solve the problem that it is impossible to dynamically allocate and adjust tasks to FPGA or GPU according to current resource usage and task requirements, resulting in an imbalance in the load between some FPGA servers and GPUs.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A computer software security encryption and decryption management system comprises: a collaborative computing platform, a collaborative management module, a system management module and a resource monitoring management module, wherein the collaborative management module, the system management module and the resource monitoring management module are all connected to the collaborative computing platform network, the collaborative management module comprises an encryption and decryption mechanism cracking module and an encryption and decryption management module, the encryption and decryption mechanism cracking module comprises a cluster hybrid heterogeneous processing architecture and a task creation module for adding tasks to the collaborative computing platform, the cluster hybrid heterogeneous processing architecture comprises a task master control module and an FPGA computing management server, the resource monitoring module is used for real-time tracking of the load conditions, memory usage, processor temperature, power consumption and other key indicators of the FPGA computing management server and GPU, the resource monitoring module comprises a data collection module, a data preprocessing module and a data storage module, the collaborative management module is connected to a scheduling strategy module, the collaborative management module performs analysis by establishing a data prediction model and performs scheduling through a scheduling strategy module, the scheduling strategy module is connected to a feedback mechanism module, and the task creation module identifies task characteristics through a heuristic algorithm.

[0007] As a further description of the above technical solution: The encryption and decryption mechanism cracking module is used to quickly analyze and crack the encryption and decryption mechanism. The FPGA computing management server includes an FPGA server management terminal and several FPGA servers. All of the FPGA servers are connected to the FPGA server management terminal. The task main control module includes several GPUs. The GPUs are all connected to the FPGA servers. The encryption and decryption mechanism cracking module also includes a computing node customization module and a task creation module connected to the collaborative computing platform. The computing node customization module is used to select professionally customized computing nodes to crack the encryption and decryption mechanism.

[0008] As a further description of the above technical solution: The data collection module collects data on the execution of a series of tasks, including the characteristics of each task and the corresponding execution time. The data preprocessing module includes data cleaning, feature engineering and data segmentation. Data cleaning is to remove or correct outliers and missing values; feature engineering: encoding categorical data such as task type (such as one-hot encoding), and standardizing or normalizing numerical data such as data size and complexity; data segmentation: dividing the data set into a training set and a test set.

[0009] As a further description of the above technical solution: The establishment of the data prediction model includes the following steps: Model construction: Select a multivariate linear regression model for intelligent analysis of tasks and configure hyperparameters to predict the trend of execution time. Model training, using training set data to train the model, usually using the least squares method to estimate the parameter β; Model evaluation, using the test set to evaluate the performance of the model; Model prediction, using the trained model for prediction; The model is applied and the predicted results are transmitted to the collaborative management module.

[0010] As a further description of the above technical solution: The data prediction model is specifically expressed by the multivariate linear regression formula as follows: Execution time = β0 + β1 × task type + β2 × data size + β3 × complexity + … + ϵ; Among them, β0 is the intercept term; β1, β2, and β3 are the coefficients of the respective variables; ϵ is the error term; …: This ellipsis indicates that there may be other independent variables in the model; Model estimation, using the least squares method to estimate the model parameters, that is, to find the coefficients; β0, β1, β2, β3 minimize the sum of squares of the difference (residual) between the predicted value and the actual execution time; Model explanation; Intercept term (β0): The estimated value of execution time when all independent variables are zero, which may not be meaningful in practice because task characteristics cannot all be zero.

[0011] As a further description of the above technical solution: The hyperparameters include task type, data size and complexity, etc. Task type coefficient: indicates the impact of task type on execution time when other variables remain unchanged. If the coefficient is positive, it means that this type of task requires longer execution time than the reference type; Data size coefficient: indicates the average increase in execution time for every unit increase in data size; Complexity coefficient: indicates the average increase in execution time for every unit increase in complexity.

[0012] As a further description of the above technical solution: The heuristic algorithm consists of the following steps: Analyze the computing characteristics of the task, such as data parallelism, computing density, memory access pattern, etc., to determine the most suitable hardware resources, and then allocate resources based on the urgency, importance or user-specified priority of the task.

[0013] As a further description of the above technical solution: Model evaluation evaluates model performance through the following indicators: R-squared (coefficient of determination): measures the degree of fit of the model to the data; residual analysis: checks whether the residuals are randomly distributed to verify the model assumptions; F-test: tests the overall significance of the model; t-test: tests the significance of each coefficient.

[0014] Model application: If the model passes the evaluation, it can be used to predict the execution time of new tasks, or to perform task optimization to reduce the execution time by adjusting task characteristics.

[0015] As a further description of the above technical solution: The scheduling strategy module includes dynamic scheduling, load balancing and resource reservation. Dynamic scheduling allocates and adjusts tasks to FPGA or GPU according to current resource usage and task requirements. Load balancing dynamically allocates and adjusts tasks to FPGA or GPU. Resource reservation ensures load balancing between FPGA server and GPU. Resource reservation avoids overloading of some devices while other devices are idle. It reserves certain resources for critical tasks to ensure that they can obtain the necessary computing power.

[0016] As a further description of the above technical solution: The feedback mechanism module includes performance feedback and error handling. Performance feedback is to collect performance data of task execution for adjusting scheduling strategies, while error handling is to monitor errors in task execution and respond quickly, such as rescheduling or allocating resources.

[0017] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: In the present invention, by combining heuristic algorithms and machine learning algorithms, the system can dynamically allocate and adjust tasks to FPGA or GPU according to current resource usage and task requirements, which can not only ensure load balancing between FPGA servers and GPUs, but also avoid the situation where some devices are overloaded while other devices are idle. The hardware resources of FPGA servers and GPUs can be more effectively utilized to improve the overall efficiency of the system, which can not only speed up the execution of tasks, but also improve the reliability of the system and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the system composition of a computer software security encryption and decryption management system proposed by the present invention; Figure 2 A schematic diagram of the model establishment process of a computer software security encryption and decryption management system proposed by the present invention; DETAILED DESCRIPTION

[0019] 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.

[0020] Please see attached Figure 1 -Attached Figure 2 The present invention provides a technical solution: a computer software security encryption and decryption management system, comprising: a collaborative computing platform, a collaborative management module, a system management module and a resource monitoring management module, wherein the collaborative management module, the system management module and the resource monitoring management module are all connected to the collaborative computing platform network, the collaborative management module comprises an encryption and decryption mechanism cracking module and an encryption and decryption management module, the encryption and decryption mechanism cracking module comprises a cluster hybrid heterogeneous processing architecture and a task creation module for adding tasks to the collaborative computing platform, the cluster hybrid heterogeneous processing architecture comprises a task master control module and an FPGA computing management server, the resource monitoring module is used to track the load of the FPGA computing management server and GPU, memory usage, processor temperature, power consumption and other key indicators in real time, the resource monitoring module comprises a data collection module, a data preprocessing module and a data storage module, the collaborative management module is connected to a scheduling strategy module, the collaborative management module analyzes by establishing a data prediction model, and schedules through a scheduling strategy module, the scheduling strategy module is connected to a feedback mechanism module, and the task creation module identifies task characteristics through a heuristic algorithm.

[0021] The encryption and decryption mechanism cracking module is used to quickly analyze and crack the encryption and decryption mechanism. The FPGA computing management server includes an FPGA server management terminal and several FPGA servers. All of the FPGA servers are connected to the FPGA server management terminal. The task main control module includes several GPUs. The GPUs are all connected to the FPGA servers. The encryption and decryption mechanism cracking module also includes a computing node customization module and a task creation module connected to the collaborative computing platform. The computing node customization module is used to select professionally customized computing nodes to crack the encryption and decryption mechanism.

[0022] The data collection module collects data on the execution of a series of tasks, including the characteristics of each task and the corresponding execution time. The data preprocessing module includes data cleaning, feature engineering and data segmentation. Data cleaning is to remove or correct outliers and missing values; feature engineering: encoding categorical data such as task type (such as one-hot encoding), and standardizing or normalizing numerical data such as data size and complexity; data segmentation: dividing the data set into a training set and a test set.

[0023] The establishment of the data prediction model includes the following steps: Model construction: Select a multivariate linear regression model for intelligent analysis of tasks and configure hyperparameters to predict the trend of execution time. Model training, using training set data to train the model, usually using the least squares method to estimate the parameter β; Model evaluation, using the test set to evaluate the performance of the model; Model prediction, using the trained model for prediction; The model is applied and the predicted results are transmitted to the collaborative management module.

[0024] The data prediction model is specifically expressed by the multivariate linear regression formula as follows: Execution time = β0 + β1 × task type + β2 × data size + β3 × complexity + … + ϵ; Among them, β0 is the intercept term; β1, β2, and β3 are the coefficients of the respective variables; ϵ is the error term; …: This ellipsis indicates that there may be other independent variables in the model; Model estimation, using the least squares method to estimate the model parameters, that is, to find the coefficients; β0, β1, β2, β3 minimize the sum of squares of the difference (residual) between the predicted value and the actual execution time; Model explanation; Intercept term (β0): The estimated value of execution time when all independent variables are zero, which may not be meaningful in practice because task characteristics cannot all be zero.

[0025] The hyperparameters include task type, data size and complexity, etc. Task type coefficient: indicates the impact of task type on execution time when other variables remain unchanged. If the coefficient is positive, it means that this type of task requires longer execution time than the reference type; Data size coefficient: indicates the average increase in execution time for every unit increase in data size; Complexity coefficient: indicates the average increase in execution time for every unit increase in complexity.

[0026] The heuristic algorithm consists of the following steps: Analyze the computing characteristics of the task, such as data parallelism, computing density, memory access pattern, etc., to determine the most suitable hardware resources, and then allocate resources based on the urgency, importance or user-specified priority of the task.

[0027] Model evaluation evaluates model performance through the following indicators: R-squared (coefficient of determination): measures the degree of fit of the model to the data; residual analysis: checks whether the residuals are randomly distributed to verify the model assumptions; F-test: tests the overall significance of the model; t-test: tests the significance of each coefficient.

[0028] Model application: If the model passes the evaluation, it can be used to predict the execution time of new tasks, or to perform task optimization to reduce the execution time by adjusting task characteristics.

[0029] The scheduling strategy module includes dynamic scheduling, load balancing and resource reservation. Dynamic scheduling allocates and adjusts tasks to FPGA or GPU according to current resource usage and task requirements. Load balancing dynamically allocates and adjusts tasks to FPGA or GPU. Resource reservation ensures load balancing between FPGA server and GPU. Resource reservation avoids overloading of some devices while other devices are idle. It reserves certain resources for critical tasks to ensure that they can obtain the necessary computing power.

[0030] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A computer software security encryption and decryption management system, characterized in that: include: A collaborative computing platform, a collaborative management module, a system management module and a resource monitoring management module, wherein the collaborative management module, the system management module and the resource monitoring management module are all connected to the collaborative computing platform network, the collaborative management module includes an encryption and decryption mechanism cracking module and an encryption and decryption management module, the encryption and decryption mechanism cracking module includes a cluster hybrid heterogeneous processing architecture and a task creation module for adding tasks to the collaborative computing platform, the cluster hybrid heterogeneous processing architecture includes a task master control module and an FPGA computing management server, the resource monitoring module is used to track in real time the load of the FPGA computing management server and GPU, memory usage, processor temperature, power consumption and other key indicators, the resource monitoring module includes a data collection module, a data preprocessing module and a data storage module, the collaborative management module is connected to a scheduling strategy module, the collaborative management module analyzes by establishing a data prediction model and schedules through a scheduling strategy module, the scheduling strategy module is connected to a feedback mechanism module, and the task creation module identifies task characteristics through a heuristic algorithm.

2. A computer software security encryption and decryption management system according to claim 1, characterized in that: The encryption and decryption mechanism cracking module is used to quickly analyze and crack the encryption and decryption mechanism. The FPGA computing management server includes an FPGA server management terminal and several FPGA servers. All of the FPGA servers are connected to the FPGA server management terminal. The task main control module includes several GPUs. The GPUs are all connected to the FPGA servers. The encryption and decryption mechanism cracking module also includes a computing node customization module and a task creation module connected to the collaborative computing platform. The computing node customization module is used to select professionally customized computing nodes to crack the encryption and decryption mechanism.

3. A computer software security encryption and decryption management system according to claim 1, characterized in that: The data collection module collects data on the execution of a series of tasks, including the characteristics of each task and the corresponding execution time. The data preprocessing module includes data cleaning, feature engineering and data segmentation. Data cleaning is to remove or correct outliers and missing values; feature engineering: encoding categorical data such as task type (such as one-hot encoding), and standardizing or normalizing numerical data such as data size and complexity; data segmentation: dividing the data set into a training set and a test set.

4. A computer software security encryption and decryption management system according to claim 1, characterized in that: The establishment of the data prediction model includes the following steps: Model construction: Select a multivariate linear regression model for intelligent analysis of tasks and configure hyperparameters to predict the trend of execution time. Model training, using training set data to train the model, usually using the least squares method to estimate the parameter β; Model evaluation, using the test set to evaluate the performance of the model; Model prediction, using the trained model for prediction; The model is applied and the predicted results are transmitted to the collaborative management module.

5. A computer software security encryption and decryption management system according to claim 4, characterized in that: The data prediction model is specifically expressed by a multiple linear regression formula as follows: Execution time = β0 + β1 × task type + β2 × data size + β3 × complexity + … + ϵ; Among them, β0 is the intercept term; β1, β2, and β3 are the coefficients of the respective variables; ϵ is the error term; …: This ellipsis indicates that there may be other independent variables in the model; Model estimation, using the least squares method to estimate the model parameters, that is, to find the coefficients; β0, β1, β2, β3 minimize the sum of squares of the difference (residual) between the predicted value and the actual execution time; Model explanation; Intercept term (β0): The estimated value of execution time when all independent variables are zero, which may not be meaningful in practice because task characteristics cannot all be zero.

6. A computer software security encryption and decryption management system according to claim 4, characterized in that: The hyperparameters include task type, data size and complexity, etc. Task type coefficient: indicates the impact of task type on execution time when other variables remain unchanged. If the coefficient is positive, it means that this type of task requires longer execution time than the reference type; Data size coefficient: indicates the average increase in execution time for every unit increase in data size; Complexity coefficient: indicates the average increase in execution time for every unit increase in complexity.

7. A computer software security encryption and decryption management system according to claim 3, characterized in that: The heuristic algorithm consists of the following steps: Analyze the computing characteristics of the task, such as data parallelism, computing density, memory access pattern, etc., to determine the most suitable hardware resources, and then allocate resources based on the urgency, importance or user-specified priority of the task.

8. A computer software security encryption and decryption management system according to claim 7, characterized in that: Model evaluation evaluates model performance through the following indicators: R-squared (coefficient of determination): measures the fit of the model to the data; residual analysis: checks whether the residuals are randomly distributed to verify the model assumptions; F-test: tests the overall significance of the model; t-test: Test the significance of each coefficient; Model application: If the model passes the evaluation, it can be used to predict the execution time of new tasks, or to optimize tasks and reduce execution time by adjusting task characteristics.

9. A computer software security encryption and decryption management system according to claim 1, characterized in that: The scheduling strategy module includes dynamic scheduling, load balancing and resource reservation. Dynamic scheduling allocates and adjusts tasks to FPGA or GPU according to current resource usage and task requirements. Load balancing dynamically allocates and adjusts tasks to FPGA or GPU. Resource reservation ensures load balancing between FPGA server and GPU. Resource reservation avoids overloading of some devices while other devices are idle. It reserves certain resources for critical tasks to ensure that they can obtain the necessary computing power.

10. A computer software security encryption and decryption management system according to claim 1, characterized in that: The feedback mechanism module includes performance feedback and error handling. Performance feedback is to collect performance data of task execution for adjusting scheduling strategies, while error handling is to monitor errors in task execution and respond quickly, such as rescheduling or allocating resources.

Images