System resource dynamic allocation method and optimization system in cloud network convergence environment

Abstract

The invention relates to a system resource dynamic allocation method and optimization system in a cloud network convergence environment, and the method comprises the following steps: 1, capturing the resource use data of each node in the cloud network convergence environment in real time through an intelligent monitoring unit; 2, analyzing the collected data, identifying a peak period and a valley period of resource use, and predicting a resource demand trend in a period of time in the future; 3, formulating a resource allocation strategy based on data analysis and demand prediction results in combination with the total resource amount and the current use condition of the system; step 4, according to the formulated resource allocation strategy, automatically triggering a resource adjustment action; and 5, establishing a real-time feedback mechanism, collecting performance data and user feedback in the resource allocation process, and regularly evaluating the effectiveness of a resource allocation strategy. The problems that system resource allocation is not flexible enough, the intelligent degree is low, and the resource utilization rate and the service quality are difficult to guarantee at the same time in the cloud network convergence environment are solved.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to a method and an optimization system for dynamically allocating system resources in a cloud-network fusion environment. Background Art

[0002] With the deep integration of cloud computing and the Internet, the cloud-network integration environment has become an important trend in the development of information technology. In this environment, the allocation and management of system resources are particularly important. At present, related technologies mainly focus on static resource allocation or dynamic allocation based on simple rules, which lacks flexibility and intelligence.

[0003] In existing technologies, a preset resource allocation strategy is usually used to make a certain degree of dynamic adjustment based on the real-time load of the system. However, this method often cannot adapt to the complex and changeable cloud-network integration environment, resulting in low resource utilization or reduced service quality. In addition, this method lacks comprehensive resource demand prediction, refined resource allocation strategy, and real-time resource adjustment mechanism, making it difficult to maximize resource utilization efficiency while ensuring service quality. Summary of the invention

[0004] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a system resource dynamic allocation method and optimization system in a cloud-network converged environment, which solves the problems of insufficient flexibility in system resource allocation, low intelligence, and difficulty in simultaneously ensuring resource utilization and service quality in a cloud-network converged environment.

[0005] The above-mentioned object of the present invention is achieved through the following technical solutions: The method for dynamically allocating system resources in a cloud-network fusion environment includes the following steps: Step 1: Real-time monitoring and data collection: Using intelligent monitoring units, real-time capture of resource usage data of each node in the cloud-network converged environment, including key indicators such as CPU utilization, memory usage, and network bandwidth; Step 2: Data analysis and demand forecasting: Analyze the collected data, identify the peak and trough periods of resource usage, reveal the spatiotemporal distribution characteristics of resource demand, and use machine learning algorithms such as time series analysis or neural networks to predict resource demand trends in the future. Step 3: Dynamic resource allocation strategy formulation: Based on the results of data analysis and demand forecasting, combined with the total amount of system resources and current usage, a resource allocation strategy is formulated; Step 4: Resource adjustment and optimization execution: According to the formulated resource allocation strategy, resource adjustment actions are automatically triggered, such as increasing or decreasing the resource quota of the node, optimizing the scheduling order of resources, and ensuring that resources can be allocated on demand; Step 5: Feedback mechanism and continuous optimization. Establish a real-time feedback mechanism to collect performance data and user feedback during the resource allocation process, regularly evaluate the effectiveness of the resource allocation strategy, and adjust and optimize it based on the feedback data to adapt to changes in system load and changes in user needs.

[0006] As a further technical solution of the present invention: in step 1, the intelligent monitoring unit includes a resource usage monitoring module, a performance indicator monitoring module, a log analysis module and an application layer monitoring module; The resource usage monitoring module captures the resource usage of each computing node, storage node and network node in real time by deploying a monitoring agent; The performance indicator monitoring module monitors key performance indicators, such as response time, throughput, and error rate, to ensure that system performance meets expectations; The log analysis module is used to collect and analyze system logs to discover potential problems, abnormal behaviors or performance bottlenecks; The application layer monitoring module is used to monitor applications running in a cloud-network converged environment, including application status, service availability, and transaction processing.

[0007] As a further technical solution of the present invention: in step 2, using a machine learning algorithm to predict resource demand specifically includes the following steps: collecting historical resource usage data and constructing a training set, then selecting a long short-term memory network LSTM machine learning algorithm for training to establish a model between resource usage and influencing factors, and finally predicting future resource demand based on the model.

[0008] As a further technical solution of the present invention: in the step 3, the processing monitoring data is dynamically allocated to obtain a first dynamic resource allocation strategy, the first dynamic resource allocation strategy is trained for overhead balance twice to obtain a second dynamic resource allocation strategy, and the second dynamic resource allocation strategy is optimized for multi-dimensional indicator balance to obtain a multi-dimensional dynamic resource allocation strategy.

[0009] As a further technical solution of the present invention: Step 3 comprises the following steps: Step S31: Analyze the node performance degradation factors, degradation characteristics, and resource proportions of the processed monitoring data, thereby obtaining resource utilization data; Step S32: performing resource cost trend analysis on the load change feature comparison data and the resource utilization data to obtain resource cost trend data, and performing clustering interval division on the resource cost trend data to obtain cost trend interval data; Step S33: Performing dynamic allocation strategy training on the overhead trend interval data to obtain a first dynamic resource allocation strategy, and performing communication overhead distribution analysis on the processed monitoring data to obtain communication overhead distribution data; Step S34: Perform distribution characteristic analysis on the communication overhead distribution data to obtain communication distribution characteristic data, perform overhead balance secondary training on the first dynamic resource allocation strategy based on the communication distribution characteristic data to obtain a second dynamic resource allocation strategy, and perform multi-dimensional indicator balance optimization on the second dynamic resource allocation strategy to obtain a multi-dimensional dynamic resource allocation strategy.

[0010] As a further technical solution of the present invention: Step 4 specifically includes the following steps: the resource monitoring module monitors the resource usage and business needs in real time; the policy engine module analyzes the data collected by the resource monitoring module and generates resource adjustment instructions according to the preset or dynamically generated resource allocation strategy; the action execution module receives the resource adjustment instruction and automatically triggers the resource adjustment action, such as increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; the feedback learning module collects the effect data after the resource adjustment and optimizes the resource allocation strategy through machine learning technology.

[0011] As a further technical solution of the present invention: the method for generating the resource allocation strategy includes determining target resource indicator data in the target object whose resource indicator data meets the resource preset conditions when it is determined that the monitoring results of the object indicator data for the target object meet the object preset conditions, wherein the target object includes at least one target resource, and determining the target influencing factor associated with the target resource indicator data according to a causal relationship diagram between the preset resource indicator data and the initial influencing factor, and generating the resource allocation strategy for the target object according to the target influencing factor.

[0012] As a further technical solution of the present invention: determining the target influencing factor associated with the target resource indicator data based on the causal relationship diagram between the preset resource indicator data and the initial influencing factors, includes: traversing the causal relationship diagram between the preset resource indicator data and the initial influencing factors, determining at least two initial influencing factors associated with the target resource indicator data, determining the influence value of the first initial influencing factor on the target resource indicator data, and determining the target influencing factor from the at least two initial influencing factors based on the influence value, wherein the first initial influencing factor is one of the at least two initial influencing factors.

[0013] The present invention also discloses a system resource dynamic allocation optimization system in a cloud-network fusion environment, comprising: Monitoring module: collects and analyzes system resource usage and performance indicators in real time; Decision-making module: Based on monitoring data and combined with preset resource allocation strategies, dynamically generate resource adjustment plans; Execution module: automatically triggers resource adjustment actions based on the resource adjustment plan generated by the decision module, including increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; Feedback module: Feedback the execution results to the decision-making module so as to adjust the resource allocation strategy in real time.

[0014] In summary, the present invention includes at least one of the following beneficial technical effects: 1. The present invention discloses a method and optimization system for dynamic allocation of system resources in a cloud-network fusion environment. Compared with the prior art, this method introduces intelligent resource demand prediction and dynamic resource allocation strategy. This technology can more accurately predict future resource demand and perform refined resource allocation and adjustment according to actual demand. This can not only significantly improve resource utilization and reduce operation and maintenance costs, but also improve the flexibility and scalability of the system while ensuring service quality.

[0015] 2. The present invention can respond to changes in the resource requirements of the system in real time, dynamically adjust resource quotas and optimize resource scheduling sequences, thereby significantly improving the performance, stability and resource utilization of the system.

[0016] 3. The present invention realizes on-demand resource allocation through automated resource adjustment and optimization execution, improves resource utilization, and meets complex and changing business needs. At the same time, the resource allocation strategy is continuously optimized through the feedback learning mechanism, which improves the intelligence level and adaptability of the system.

[0017] 4. Through the real-time feedback mechanism and continuous optimization process, the present invention can significantly improve the response speed and accuracy of resource allocation, reduce the risk of uneven system load and unmet user needs, thereby improving the overall performance of the system and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The figure is a simplified flow chart of the method of the present invention.

[0019] Figure 2 The figure is a simplified structural diagram of the optimization system of the present invention. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application; obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments, and all other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application without making creative work are within the scope of protection of the present application.

[0021] Embodiment 1: Reference Figure 1 , is a method for dynamically allocating system resources in a cloud-network fusion environment disclosed in the present invention, comprising the following steps: Step 1: Real-time monitoring and data collection: Using intelligent monitoring units, real-time capture of resource usage data of each node in the cloud-network converged environment, including key indicators such as CPU utilization, memory usage, and network bandwidth; Step 2: Data analysis and demand forecasting: Analyze the collected data, identify the peak and trough periods of resource usage, reveal the spatiotemporal distribution characteristics of resource demand, and use machine learning algorithms such as time series analysis or neural networks to predict resource demand trends in the future. Step 3: Dynamic resource allocation strategy formulation: Based on the results of data analysis and demand forecasting, combined with the total amount of system resources and current usage, a resource allocation strategy is formulated; Step 4: Resource adjustment and optimization execution: According to the formulated resource allocation strategy, resource adjustment actions are automatically triggered, such as increasing or decreasing the resource quota of the node, optimizing the scheduling order of resources, and ensuring that resources can be allocated on demand; Step 5: Feedback mechanism and continuous optimization. Establish a real-time feedback mechanism to collect performance data and user feedback during the resource allocation process, regularly evaluate the effectiveness of the resource allocation strategy, and adjust and optimize it based on the feedback data to adapt to changes in system load and changes in user needs.

[0022] In step 1, the intelligent monitoring unit includes a resource usage monitoring module, a performance indicator monitoring module, a log analysis module and an application layer monitoring module; the resource usage monitoring module captures the resource usage of each computing node, storage node and network node in real time by deploying a monitoring agent; the performance indicator monitoring module monitors key performance indicators such as response time, throughput and error rate to ensure that the system performance meets expectations; the log analysis module is used to collect and analyze system logs to discover potential problems, abnormal behaviors or performance bottlenecks; the application layer monitoring module is used to monitor applications running in a cloud-network converged environment, including application status, service availability and transaction processing.

[0023] In step 2, using machine learning algorithms to predict resource demand specifically includes the following steps: collecting historical resource usage data and building a training set, then selecting a long short-term memory network (LSTM) machine learning algorithm for training to establish a model between resource usage and influencing factors, and finally predicting future resource demand based on the model.

[0024] In step 3, dynamic allocation strategy training is performed on the processed monitoring data to obtain a first dynamic resource allocation strategy, overhead balance training is performed on the first dynamic resource allocation strategy to obtain a second dynamic resource allocation strategy, and multi-dimensional indicator balance optimization is performed on the second dynamic resource allocation strategy to obtain a multi-dimensional dynamic resource allocation strategy.

[0025] Further, step 3 includes the following steps: Step S31: Analyze the node performance degradation factors, degradation characteristics, and resource proportions of the processed monitoring data, thereby obtaining resource utilization data; Step S32: performing resource cost trend analysis on the load change feature comparison data and the resource utilization data to obtain resource cost trend data, and performing clustering interval division on the resource cost trend data to obtain cost trend interval data; Step S33: Performing dynamic allocation strategy training on the overhead trend interval data to obtain a first dynamic resource allocation strategy, and performing communication overhead distribution analysis on the processed monitoring data to obtain communication overhead distribution data; Step S34: Perform distribution characteristic analysis on the communication overhead distribution data to obtain communication distribution characteristic data, perform overhead balance secondary training on the first dynamic resource allocation strategy based on the communication distribution characteristic data to obtain a second dynamic resource allocation strategy, and perform multi-dimensional indicator balance optimization on the second dynamic resource allocation strategy to obtain a multi-dimensional dynamic resource allocation strategy.

[0026] Step 4 specifically includes the following steps: the resource monitoring module monitors resource usage and business needs in real time; the policy engine module analyzes the data collected by the resource monitoring module and generates resource adjustment instructions based on preset or dynamically generated resource allocation strategies; the action execution module receives the resource adjustment instructions and automatically triggers resource adjustment actions, such as increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; the feedback learning module collects the effect data after resource adjustment and optimizes the resource allocation strategy through machine learning technology.

[0027] The method for generating a resource allocation strategy includes determining target resource indicator data in the target object whose resource indicator data meets the resource preset conditions when it is determined that the monitoring results of the object indicator data for the target object meet the object preset conditions, wherein the target object includes at least one target resource, determining the target influencing factor associated with the target resource indicator data based on a causal relationship diagram between the preset resource indicator data and the initial influencing factor, and generating a resource allocation strategy for the target object based on the target influencing factor.

[0028] According to a causal relationship diagram between preset resource indicator data and initial influencing factors, a target influencing factor associated with the target resource indicator data is determined, including: traversing the causal relationship diagram between the preset resource indicator data and the initial influencing factors, determining at least two initial influencing factors associated with the target resource indicator data, determining an impact value of a first initial influencing factor on the target resource indicator data, and determining the target influencing factor from at least two initial influencing factors according to the impact value, wherein the first initial influencing factor is one of the at least two initial influencing factors.

[0029] The application scenarios of this embodiment include but are not limited to: cloud service providers dynamically allocate computing resources, storage resources, and network resources in a cloud-network converged environment to meet the service level agreement (SLA) requirements of different customers; in the internal IT systems of large enterprises, dynamically adjust resource allocation according to business needs and system load to improve overall operation and maintenance efficiency and service quality.

[0030] Embodiment 2: With the rapid development of cloud computing technology, resource management and scheduling have become the key to ensuring the efficient operation of cloud services. At present, cloud resource management systems usually have basic resource allocation and scheduling functions, but they often lack flexibility and intelligence, and it is difficult to make dynamic adjustments based on resource usage and business needs in real time. Existing technologies usually use preset resource allocation rules and scheduling algorithms for resource management. However, this method often cannot achieve refined and dynamic management of resources when facing complex and changing business scenarios.

[0031] Existing technologies lack an intelligent resource adjustment mechanism and are unable to automatically trigger resource adjustment actions based on real-time resource usage and business needs, such as increasing or decreasing node resource quotas, optimizing resource scheduling sequences, etc., resulting in low resource utilization or insufficient resources.

[0032] Reference Figure 2 The present invention also discloses a system resource dynamic allocation optimization system in a cloud-network fusion environment, including: Monitoring module: collects and analyzes system resource usage and performance indicators in real time; Decision-making module: Based on monitoring data and combined with preset resource allocation strategies, dynamically generate resource adjustment plans; Execution module: automatically triggers resource adjustment actions based on the resource adjustment plan generated by the decision module, including increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; Feedback module: Feedback the execution results to the decision-making module so as to adjust the resource allocation strategy in real time.

[0033] The present invention is applicable to resource management systems in cloud computing environments, and is particularly applicable to scenarios requiring real-time dynamic adjustment of resource quotas and scheduling sequences, such as large enterprise-level cloud service providers, high-concurrency online application platforms, and the like.

[0034] Compared with the existing technology, the technical effects of this technical solution are as follows: through automated resource adjustment and optimization execution, on-demand resource allocation is achieved, resource utilization is improved, and complex and changing business needs are met. At the same time, the resource allocation strategy is continuously optimized through the feedback learning mechanism, which improves the intelligence level and adaptability of the system.

[0035] The implementation principle of the present invention is as follows: the present invention discloses a method and optimization system for dynamic allocation of system resources in a cloud-network fusion environment. Compared with the prior art, this method introduces intelligent resource demand prediction and dynamic resource allocation strategy. This technology can more accurately predict future resource demand and perform refined resource allocation and adjustment according to actual demand. This can not only significantly improve resource utilization and reduce operation and maintenance costs, but also improve the flexibility and scalability of the system while ensuring service quality.

[0036] The embodiments of this specific implementation method are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for dynamically allocating system resources in a cloud-network fusion environment, characterized in that: The following steps are involved: Step 1: Real-time monitoring and data collection: Using intelligent monitoring units, real-time capture of resource usage data of each node in the cloud-network converged environment, including key indicators such as CPU utilization, memory usage, and network bandwidth; Step 2: Data analysis and demand forecasting: Analyze the collected data, identify the peak and trough periods of resource usage, reveal the spatiotemporal distribution characteristics of resource demand, and use machine learning algorithms such as time series analysis or neural networks to predict resource demand trends in the future. Step 3: Dynamic resource allocation strategy formulation: Based on the results of data analysis and demand forecasting, combined with the total amount of system resources and current usage, a resource allocation strategy is formulated; Step 4: Resource adjustment and optimization execution: According to the formulated resource allocation strategy, resource adjustment actions are automatically triggered, such as increasing or decreasing the resource quota of the node, optimizing the scheduling order of resources, and ensuring that resources can be allocated on demand; Step 5: Feedback mechanism and continuous optimization. Establish a real-time feedback mechanism to collect performance data and user feedback during the resource allocation process, regularly evaluate the effectiveness of the resource allocation strategy, and adjust and optimize it based on the feedback data to adapt to changes in system load and changes in user needs.

2. The method for dynamically allocating system resources in a cloud-network fusion environment according to claim 1, characterized in that: In step 1, the intelligent monitoring unit includes a resource usage monitoring module, a performance indicator monitoring module, a log analysis module and an application layer monitoring module; The resource usage monitoring module captures the resource usage of each computing node, storage node and network node in real time by deploying a monitoring agent; The performance indicator monitoring module monitors key performance indicators, such as response time, throughput, and error rate, to ensure that system performance meets expectations; The log analysis module is used to collect and analyze system logs to discover potential problems, abnormal behaviors or performance bottlenecks; The application layer monitoring module is used to monitor applications running in a cloud-network converged environment, including application status, service availability, and transaction processing.

3. The method for dynamically allocating system resources in a cloud-network fusion environment according to claim 1, characterized in that: In step 2, using a machine learning algorithm to predict resource demand specifically includes the following steps: collecting historical resource usage data and constructing a training set, then selecting a long short-term memory network (LSTM) machine learning algorithm for training to establish a model between resource usage and influencing factors, and finally predicting future resource demand based on the model.

4. The method for dynamically allocating system resources in a cloud-network fusion environment according to claim 1, characterized in that: In step 3, dynamic allocation strategy training is performed on the processed monitoring data to obtain a first dynamic resource allocation strategy, overhead balance secondary training is performed on the first dynamic resource allocation strategy to obtain a second dynamic resource allocation strategy, and multi-dimensional indicator balance optimization is performed on the second dynamic resource allocation strategy to obtain a multi-dimensional dynamic resource allocation strategy.

5. The method for dynamically allocating system resources in a cloud-network fusion environment according to claim 4, characterized in that: The step 3 comprises the following steps: Step S31: Analyze the node performance degradation factors, degradation characteristics, and resource proportions of the processed monitoring data, thereby obtaining resource utilization data; Step S32: performing resource cost trend analysis on the load change feature comparison data and the resource utilization data to obtain resource cost trend data, and performing clustering interval division on the resource cost trend data to obtain cost trend interval data; Step S33: Performing dynamic allocation strategy training on the overhead trend interval data to obtain a first dynamic resource allocation strategy, and performing communication overhead distribution analysis on the processed monitoring data to obtain communication overhead distribution data; Step S34: Perform distribution characteristic analysis on the communication overhead distribution data to obtain communication distribution characteristic data, perform overhead balance secondary training on the first dynamic resource allocation strategy based on the communication distribution characteristic data to obtain a second dynamic resource allocation strategy, and perform multi-dimensional indicator balance optimization on the second dynamic resource allocation strategy to obtain a multi-dimensional dynamic resource allocation strategy.

6. The method for dynamic allocation of system resources in a cloud-network fusion environment according to claim 1, characterized in that: The step 4 specifically includes the following steps: the resource monitoring module monitors the resource usage and business needs in real time; the policy engine module analyzes the data collected by the resource monitoring module and generates resource adjustment instructions according to the preset or dynamically generated resource allocation strategy; the action execution module receives the resource adjustment instruction and automatically triggers the resource adjustment action, such as increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; the feedback learning module collects the effect data after the resource adjustment and optimizes the resource allocation strategy through machine learning technology.

7. The method for dynamic allocation of system resources in a cloud-network fusion environment according to claim 6, characterized in that: The method for generating the resource allocation strategy includes determining target resource indicator data in the target object whose resource indicator data meets the resource preset conditions when it is determined that the monitoring results of the object indicator data for the target object meet the object preset conditions, wherein the target object includes at least one target resource, determining the target influencing factor associated with the target resource indicator data based on a causal relationship diagram between the preset resource indicator data and the initial influencing factor, and generating the resource allocation strategy for the target object based on the target influencing factor.

8. The method for dynamically allocating system resources in a cloud-network fusion environment according to claim 7, characterized in that: The method of determining the target influencing factor associated with the target resource indicator data based on a causal relationship diagram between preset resource indicator data and initial influencing factors includes: traversing the causal relationship diagram between preset resource indicator data and initial influencing factors, determining at least two initial influencing factors associated with the target resource indicator data, determining an influence value of a first initial influencing factor on the target resource indicator data, and determining the target influencing factor from the at least two initial influencing factors based on the influence value, wherein the first initial influencing factor is one of the at least two initial influencing factors.

9. A system for dynamically allocating and optimizing system resources in a cloud-network fusion environment, characterized in that it includes: Monitoring module: collects and analyzes system resource usage and performance indicators in real time; Decision-making module: Based on monitoring data and combined with preset resource allocation strategies, dynamically generate resource adjustment plans; Execution module: automatically triggers resource adjustment actions based on the resource adjustment plan generated by the decision module, including increasing or decreasing the resource quota of the node and optimizing the scheduling order of resources; Feedback module: Feedback the execution results to the decision-making module so as to adjust the resource allocation strategy in real time.

Images