A cloud computing-based nuclear power plant PaaS platform resource quota monitoring method and system

Abstract

This invention discloses a cloud-based nuclear power plant PaaS platform resource quota monitoring method and system, comprising a visual management platform portal, a cloud-based database, and a cloud-based resource scheduling center. Its beneficial effects include enabling more rational, flexible, and convenient management of all resource quotas within the cloud-based PaaS platform, specifically for nuclear power plant operations. This reduces manual operation and maintenance costs, enables automated forecasting and alerting, effectively integrates resource utilization, and improves cluster throughput.

Description

Technical Field

[0001] The present invention belongs to the technical field of cloud computing, and in particular relates to a method and system for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing. Background Art

[0002] Cloud computing is a distributed computing model with a wide variety of cloud computing resources. Cloud platforms are responsible for monitoring these resources. Improving virtual node resource utilization and achieving load balancing within cloud computing's virtual resources has been a key research topic for developers in recent years. In real-world cloud platform applications, users deploy and run applications in cloud environments. How can resources be allocated on demand? When resources need to be dispatched across multiple locations, how can they be efficiently coordinated to maintain load balancing? Real-time resource monitoring and effective load balancing are crucial.

[0003] With the vigorous development of my country's nuclear power industry, nuclear power plants will inevitably put forward new and higher requirements for the application of information technology in their physical businesses such as technology research and development, design, construction, commissioning, production, operation, maintenance, decommissioning, and technical services. Platform management based on standardization, specialization, and high availability is an urgent need in the current nuclear power information field.

[0004] Visual information management based on the cloud platform can meet the resource monitoring needs of various business scenarios with relatively low cost investment. This patent provides a method for adaptive monitoring rules of cloud platform resources, which judges the data obtained from resource monitoring, discovers and allocates resources in real time, and performs task scheduling and load balancing. Summary of the Invention

[0005] The purpose of the present invention is to provide a cloud computing-based nuclear power plant PaaS platform resource quota monitoring method and system, which can achieve more reasonable, flexible and convenient management of all resource quotas under the cloud computing PaaS platform with nuclear power plant business as the scope.

[0006] The technical solution of the present invention is as follows: a cloud computing-based nuclear power plant PaaS platform resource quota monitoring system, including a visual management platform portal, a cloud database, and a cloud resource scheduling center.

[0007] The visual management platform portal includes: data display module, records and alarms, which realize real-time monitoring and display of the operating status of key resources of physical hosts and key services of cloud platforms, including: CPU utilization, memory utilization, network port I / O and disk utilization data collection and storage, log recording of abnormal situations and alarm functions for abnormal resource usage.

[0008] The cloud database includes: data collection and data storage, obtaining and storing data information and resource quotas of resources in different regions, the data collection module transfers resources without adjustment according to the resource monitoring module, and dynamically adjusts the transmitted data through the resource scheduling module, and predicts whether the load quota needs to be dynamically adjusted through a custom algorithm to improve the load balancing efficiency of the application.

[0009] The cloud resource scheduling center includes: resource configuration module, resource monitoring module, resource policy control module, and resource scheduling module. The synergistic effect of each module can monitor data in real time and schedule resources to ensure that service resources can operate efficiently and correctly.

[0010] A cloud computing-based nuclear power plant PaaS platform resource quota monitoring method includes the following steps:

[0011] Step 1: Configure the host environment and build a Kubernetes-based cluster architecture to run multiple tenants with different load balancing requirements on the same virtual node.

[0012] Step 2: Configure load balancing data for users in multiple locations. You can use the default quota or manually adjust the quota.

[0013] Step 3: Monitor the resource usage of each service in the Pass cloud platform environment in real time, including the total resource usage and the resource usage of each tenant node;

[0014] Step 4: Use a custom algorithm to predict whether resource adjustments are needed based on the data obtained from resource monitoring;

[0015] Step 5: Obtain the resource services that need to be adjusted and determine whether to dynamically expand or shrink the resources by setting the load threshold;

[0016] Step 6: The adaptively dynamically adjusted resource services will be deployed to the target node.

[0017] The step 1 includes:

[0018] Step 11: Configure resource objects based on the Kubernetes cluster and provide a RESTful management interface to support operations such as adding, deleting, modifying, querying, and monitoring Kubernetes cluster resource objects.

[0019] Step 12: Create a visual user interface, configure the log module to record abnormal status, and configure the alarm module to send information about related resource abnormalities;

[0020] Step 13: Configure the functional plug-in for user authentication and authorization;

[0021] Step 14: Support HTTPS protocol for secure access.

[0022] Said step 2 comprises,

[0023] Step 21: Build a resource quota table and set default resource quotas based on the tenant's application service volume, including the maximum and minimum resource configuration values;

[0024] Step 22: For services with configured resource quotas, administrators can manually expand or shrink capacity and adjust resources based on actual resource requirements.

[0025] Said step 4 comprises,

[0026] Step 41: The judgment result of step 4 is a preliminary judgment. To ensure that the resource status can be adjusted effectively in real time, a second judgment is performed on the resource node obtained in the preliminary judgment to determine whether the obtained instance resource value falls within the maximum and minimum value ranges of its quota configuration. If not, it is determined that the instance resource information is "abnormal" and needs to be dynamically adjusted.

[0027] Step 42: The resource monitoring module transmits data to the resource policy control module and the data collection module in real time;

[0028] Step 43: The optimal scaling strategy in the resource strategy module includes: an optimal expansion strategy and an optimal contraction strategy. The optimal expansion strategy or the optimal contraction strategy is selected based on the resource status of the node.

[0029] Said step 5 comprises,

[0030] Step 51: In the judgment result of step 43, when the current value of the instance resource is less than the minimum value configured in the instance resource information, it indicates that the resource situation is good and flexible resource scaling can be performed. When the current value of the instance resource is greater than the maximum value configured in the instance resource information, the service instance is pre-expanded or expanded in real time according to the optimal expansion strategy rules. If the prediction result of the resource monitoring module shows that there is still insufficient resources at the next moment after the pre-expansion of the resource instance, vertical real-time expansion is required to maximize resource utilization.

[0031] The beneficial effects of the present invention are as follows: The cloud computing-based nuclear power plant PaaS platform resource quota monitoring method and system provided by the present invention can achieve more reasonable, flexible, and convenient management of all resource quotas under the cloud computing PaaS platform, focusing on nuclear power plant business. This reduces manual operation and maintenance costs, enables automated prediction and alarming, effectively integrates resource utilization, and improves cluster throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of the architecture of the cloud computing-based nuclear power plant PaaS platform resource quota monitoring method of the present invention;

[0033] Figure 2 This is a schematic diagram of a management device for a cloud computing-based nuclear power plant PaaS platform resource quota monitoring method according to the present invention;

[0034] Figure 3 This is a flow chart of a method for monitoring resource quotas on a PaaS platform for nuclear power plants based on cloud computing according to the present invention;

[0035] Figure 4 The present invention provides a flow chart of a resource monitoring and prediction method and a resource dynamic scaling method for a cloud computing-based nuclear power plant PaaS platform resource quota monitoring method. DETAILED DESCRIPTION

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] The present invention provides a cloud computing-based nuclear power plant PaaS platform resource quota monitoring system, comprising: a visual management platform portal, a cloud database, and a cloud resource scheduling center.

[0038] like Figure 1 The illustrated cloud computing-based nuclear power plant PaaS platform resource quota monitoring system includes a visual platform management portal, a cloud database, a cloud resource scheduling center, and terminal tenants. The visual platform management portal monitors and displays the real-time operating status of key physical host resources and key cloud platform services.

[0039] Cloud database users collect operating data from end users and the platform's own application service programs, and push it to the data storage module and resource scheduling center according to certain processing rules. The data storage module stores the received data according to the pre-configured table structure.

[0040] The cloud resource scheduling center is used to manage resource quotas for initialization services and monitor, predict, and dynamically adjust resources for existing services. The resource monitoring module monitors node loads in real time and samples and analyzes the data obtained from monitoring according to pre-set rules. The resource policy control module and resource scheduling module decide whether dynamic resource adjustments are needed based on the analysis results transmitted by the monitoring module. Finally, the application is deployed to the service terminal and the data is transmitted to the cloud database.

[0041] The visual management platform portal includes: data display module, records and alarms, which realize real-time monitoring and display of the operating status of key resources of physical hosts and key services of cloud platforms, including: CPU utilization, memory utilization, network port I / O and disk utilization data collection and storage, log recording of abnormal situations and alarm functions for abnormal resource usage.

[0042] The data display module in the visualization platform management portal mainly displays the resource usage of each service terminal, including CPU utilization, memory utilization, network port I / O and disk utilization data collection and storage, log recording of abnormal situations and alarm information of abnormal resource utilization.

[0043] For exception alerts, the resource monitoring module analyzes and assesses acquired data, connecting to the alarm module via a message bus. This allows the module to promptly monitor abnormalities occurring during the operation of each node. The alarm module presets different alarm levels for different exception reports and sets different alarm notification methods for different alarm levels, enabling faster and more effective handling of platform failures. Abnormal severity levels are categorized as: Level I, Level II, Level III, and Level IV, represented on the display panel by red, orange, yellow, and blue, respectively, with Level I being the highest.

[0044] The cloud database includes data collection and storage, acquiring and storing data and resource quotas across different regions. The data collection module dynamically adjusts data transmitted from the resource scheduling module based on resources that do not require adjustment, based on resources transferred from the resource monitoring module. Using a custom algorithm, it predicts whether load quotas require dynamic adjustments, improving the efficiency of application load balancing.

[0045] The data collection module automatically obtains real-time data from the cloud resource scheduling center through the message bus. At the same time, it determines the data transmission path based on the judgment information of the resource policy control module and the resource scheduling module to ensure the dynamic balance of node resources.

[0046] The data storage module receives data transmitted by the monitoring module in a cluster manner, and uses a combination of MySQL and MongoDB to store structured data and unstructured data, making data processing and storage more flexible.

[0047] The cloud resource scheduling center includes: resource configuration module, resource monitoring module, resource policy control module, and resource scheduling module. The synergy of each module enables real-time monitoring of data and resource scheduling to ensure that service resources can operate efficiently and correctly.

[0048] The resource monitoring module mainly monitors the usage of various service resources in the pass cloud platform environment in real time, including the total resource usage and the resource usage of each tenant node.

[0049] The resource policy control module proposes an adaptive load balancing algorithm and adopts a dual-threshold elastic scaling method to achieve flexible scaling of the resource pool. It monitors the overload level of virtual node resources in real time and adaptively adopts resource policy control to dynamically adjust resources.

[0050] The resource scheduling module mainly performs flexible scaling of the load, and the strategies of the resource policy control module are further implemented through the resource scheduling module.

[0051] The resource configuration module configures the load balancing data for users in multiple locations. The quota can use the default quota or be manually adjusted.

[0052] Figure 2 The cloud computing-based nuclear power plant PaaS platform resource quota monitoring system shown includes a user registration service, a resource configuration module, a resource monitoring module, a resource policy control module, a resource scheduling module, data collection, and data storage.

[0053] Its working process is as follows: the user registration service initiates a resource request, which carries the default resource quota information. The resource configuration module can manually change the resource configuration to improve resource utilization. The resource monitoring module monitors the resource information of each node through the message bus to facilitate timely feedback to the database and resource policy control module. The resource policy control module uses an adaptive load balancing algorithm and a dual-threshold elastic scaling method to achieve flexible scaling of the resource pool. The first threshold monitors the resource load and predicts whether resource adjustment is required. The second threshold determines whether the load information is within the normal threshold range based on the quota, and pre-expands and shrinks the resources. The data collection module determines the data transmission method based on the resources that do not need to be adjusted from the resource monitoring module, and the data transmitted by the resource scheduling module to dynamically adjust the transmission, to ensure that resources can be effectively deployed on the node.

[0054] The present invention provides a method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing, comprising the following steps:

[0055] Step 1: Configure the host environment and build a Kubernetes-based cluster architecture to run multiple tenants with different load balancing requirements on the same virtual node.

[0056] Step 2: Configure load balancing data for users in multiple locations. You can use the default quota or manually adjust the quota.

[0057] Step 3: Monitor the resource usage of each service in the Pass cloud platform environment in real time, including the total resource usage and the resource usage of each tenant node;

[0058] Step 4: Use a custom algorithm to predict whether resource adjustments are needed based on the data obtained from resource monitoring;

[0059] Step 5: Obtain the resource services that need to be adjusted and determine whether to dynamically expand or shrink the resources by setting the load threshold;

[0060] Step 6: The adaptively dynamically adjusted resource services will be deployed to the target node.

[0061] In order to make the above features and advantages of the present invention more clearly understood, embodiments are given and described in detail with reference to the accompanying drawings as follows:

[0062] Combine Figure 3 and Figure 4 , further explains the resource quota monitoring method of the nuclear power plant PaaS platform based on cloud computing:

[0063] Step 1: Configure the host environment and build a Kubernetes-based cluster architecture to run multiple tenants with different load balancing requirements on the same virtual node.

[0064] Step 11: Configure resource objects based on the Kubernetes cluster and provide a RESTful management interface to support operations such as adding, deleting, modifying, querying, and monitoring Kubernetes cluster resource objects.

[0065] Step 12: Create a visual user interface, configure the log module to record abnormal status, and configure the alarm module to send information about related resource abnormalities;

[0066] Step 13: Configure the functional plug-in for user authentication and authorization;

[0067] Step 14: Support HTTPS protocol for secure access.

[0068] Step 2: Configure the load balancing data for users in multiple locations. The quota can use the default quota or be manually adjusted. The specific steps include the following:

[0069] Step 21: Build a resource quota table and set default resource quotas based on the tenant's application service volume, including the maximum and minimum resource configuration values;

[0070] Step 22: For services with configured resource quotas, administrators can manually expand or shrink capacity and adjust resources based on actual resource requirements.

[0071] Step 3: Monitor resource usage across all services in the Pass cloud platform in real time, including total resource usage and resource usage by each tenant node. Specific monitoring data includes: CPU application quota, CPU usage, CPU utilization, memory application quota, memory usage, memory utilization, disk application quota, disk usage, disk utilization, cloud host IP, cloud host port, and subnet segment. This monitoring data is stored in a database table with the node ID as the primary key. This table is a separate table for storing monitoring data.

[0072] Step 4: The resource policy module makes a first judgment on the data obtained from resource monitoring based on a custom algorithm, and predicts whether the monitored resource node needs to make resource adjustments; obtains the CPU usage, memory usage, and disk usage of the monitored node, takes their weights according to the rules, and combines the total resource occupancy of the virtual machines running on the physical machine to obtain the judgment threshold T. Based on the classification tree algorithm, the resource status of the current node is obtained. The resource status includes: high load, low load, and load balance. The resource requests with high load and low load status are transmitted to the resource scheduling module. Specifically, it also includes the following steps:

[0073] Step 41: The judgment result of step 4 is a preliminary judgment. In order to ensure that the resource status can be adjusted effectively in real time, a second judgment is performed on the resource node obtained in the preliminary judgment to determine whether the obtained instance resource value is within the maximum and minimum value range of its quota configuration. If not, it is determined that the instance resource information is "abnormal" and needs to be dynamically adjusted.

[0074] Step 42: The resource monitoring module transmits data to the resource policy control module and the data collection module in real time;

[0075] Step 43: The optimal scaling strategy in the resource strategy module includes: optimal expansion strategy and optimal reduction strategy. The optimal expansion strategy or optimal reduction strategy is selected based on the resource status of the node.

[0076] Step 5: The resource scheduling module obtains the resource services that need to be dynamically adjusted and dynamically expands and shrinks the node resources;

[0077] Step 51: In the judgment result of step 43, if the current value of the instance resource is less than the minimum value configured in the instance resource information, it indicates that the resource situation is good and flexible resource scaling can be performed. If the current value of the instance resource is greater than the maximum value configured in the instance resource information, the service instance is pre-scaled or scaled in real time according to the optimal scaling strategy rules. If the resource monitoring module's prediction results indicate that the resource instance is still insufficient at the next moment after pre-scaling, vertical real-time scaling is required to maximize resource utilization.

[0078] Step 6: The adaptively dynamically adjusted resource services will be deployed to the target node.

Claims

1. A method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing, characterized in that: The following steps are involved: Step 1: Configure the host environment and build a Kubernetes-based cluster architecture to run multiple tenants with different load balancing requirements on the same virtual node. Step 2: Configure load balancing data for users in multiple locations. You can use the default quota or manually adjust the quota. Step 3: Monitor the resource usage of each service in the Pass cloud platform environment in real time, including the total resource usage and the resource usage of each tenant node; Step 4: Use a custom algorithm to predict whether resource adjustments are needed based on the data obtained from resource monitoring; Step 5: Obtain the resource services that need to be adjusted and determine whether to dynamically expand or shrink the resources by setting the load threshold; Step 6: The adaptively dynamically adjusted resource services will be deployed to the target node.

2. The method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing according to claim 1, characterized in that: The step 1 includes: Step 11: Configure resource objects based on the Kubernetes cluster and provide a RESTful management interface to support operations such as adding, deleting, modifying, querying, and monitoring Kubernetes cluster resource objects. Step 12: Create a visual user interface, configure the log module to record abnormal status, and configure the alarm module to send information about related resource abnormalities; Step 13: Configure the functional plug-in for user authentication and authorization; Step 14: Support HTTPS protocol for secure access.

3. The method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing according to claim 1, characterized in that: Said step 2 comprises, Step 21: Build a resource quota table and set default resource quotas based on the tenant's application service volume, including the maximum and minimum resource configuration values; Step 22: For services with configured resource quotas, administrators can manually expand or shrink capacity and adjust resources based on actual resource requirements.

4. The method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing according to claim 1, characterized in that: Said step 4 comprises, Step 41: The judgment result of step 4 is a preliminary judgment. To ensure that the resource status can be adjusted effectively in real time, a second judgment is performed on the resource node obtained in the preliminary judgment to determine whether the obtained instance resource value falls within the maximum and minimum values ​​of its quota configuration. If not, it is determined that the instance resource information is "abnormal" and needs to be dynamically adjusted. Step 42: The resource monitoring module transmits data to the resource policy control module and the data collection module in real time; Step 43: The optimal scaling strategy in the resource strategy module includes: an optimal expansion strategy and an optimal contraction strategy. The optimal expansion strategy or the optimal contraction strategy is selected based on the resource status of the node.

5. The method for monitoring resource quotas of a nuclear power plant PaaS platform based on cloud computing according to claim 4, characterized in that: Said step 5 comprises, Step 51: In the judgment result of step 43, when the current value of the instance resource is less than the minimum value configured in the instance resource information, it indicates that the resource situation is good and flexible resource scaling is performed. When the current value of the instance resource is greater than the maximum value configured in the instance resource information, the service instance is pre-expanded or expanded in real time according to the optimal expansion strategy rules. If the prediction result of the resource monitoring module shows that there is still insufficient resource at the next moment after the instance resource is pre-expanded, real-time expansion is required to maximize resource utilization.

Images