Intelligent campus big data acquisition and management system based on cloud computing

Abstract

The invention discloses an intelligent campus big data acquisition and management system based on cloud computing, and relates to the technical field of intelligent campus construction and big data processing, and the system comprises the following components: a big data acquisition module, a cloud computing resource pool, a dynamic resource scheduling engine, an intelligent load balancer, and a visual monitoring and management platform. The system realizes real-time and comprehensive acquisition of multi-source heterogeneous data in the campus through the big data acquisition module by using the Internet of Things (IoT) technology, a sensor network and an API interface, can effectively remove repeated, wrong and missing data through the data cleaning and preprocessing process, improves the data quality, and in addition, improves the data processing efficiency. The visual monitoring and management platform provides a visual monitoring interface, and supports an administrator to perform remote configuration, task scheduling and resource optimization operation, so that the efficiency of data management and operation and maintenance is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of smart campus construction and big data processing technology, and specifically to a smart campus big data collection and management system based on cloud computing. Background Art

[0002] Traditional campus data management methods often have problems such as data silos, lack of information sharing, and low processing efficiency, which makes it difficult to meet the requirements of smart campuses for real-time, accurate, and comprehensive data. At the same time, with the expansion of campus scale and the complexity of business, the sharp increase in data volume has brought tremendous pressure to traditional data management systems, making it difficult to effectively support smart campus decision-making analysis, teaching evaluation, and security monitoring advanced applications.

[0003] With the deepening of smart campus construction, various types of teaching, management, and life data have exploded, and the demand for data processing capabilities and computing resources is increasing. Traditional data centers often face problems such as uneven resource allocation, poor scalability, and high operating costs, and it is difficult to support large-scale, high-concurrency data processing tasks. Therefore, it is particularly important to develop a smart campus big data collection and management system based on cloud computing. Summary of the invention

[0004] The present invention proposes a cloud computing-based smart campus big data collection and management system, which aims to achieve efficient, comprehensive and intelligent collection and management of various types of data in the smart campus by integrating advanced cloud computing technology and big data analysis methods, thereby optimizing campus resource allocation and improving campus management efficiency.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a cloud computing-based smart campus big data collection and management system, which includes a big data collection module, a cloud computing resource pool, a dynamic resource scheduling engine, an intelligent load balancer, and a visual monitoring and management platform;

[0006] The big data collection module: uses the Internet of Things (IoT) technology, sensor networks and API interfaces to achieve real-time and comprehensive collection of various teaching, management, safety and life multi-source heterogeneous data on campus, supports data cleaning and preprocessing, and ensures data quality;

[0007] The cloud computing resource pool: builds a resource pool based on cloud computing technology, including virtualized servers, storage resources and network resources. The resource pool can be flexibly expanded according to actual needs and supports dynamic allocation and recovery of computing resources;

[0008] The dynamic resource scheduling engine: by real-time monitoring of the load of data processing tasks, resource utilization and user request volume, it uses advanced prediction models and intelligent algorithms to automatically adjust the resource allocation of the cloud server cluster, automatically increase resource supply during peak periods, and reduce resource usage during off-peak periods, ensuring that the system can maintain high performance even in high-concurrency scenarios;

[0009] The intelligent load balancer integrates multiple load balancing algorithms and intelligently distributes processing tasks to different nodes based on task characteristics and system status, effectively avoiding overload of a single node and improving overall data processing capabilities and system scalability;

[0010] The visual monitoring and management platform provides an intuitive monitoring interface to display system operation status, resource usage, and task processing progress information, supports administrators to perform remote configuration, task scheduling, and resource optimization operations, simplifies the operation and maintenance process, and reduces operating costs.

[0011] Furthermore, the big data collection module determines the specific data sources in the fields of teaching, management, security and life on campus that need to collect data. According to the data collection requirements, sensors and IoT devices are deployed at appropriate locations on campus, and various existing information systems on campus are connected. The API interface is configured. The data obtained through the API interface can be expressed as D api (t) Use the deployed sensors, IoT devices and API interfaces to collect data in real time, clean the collected data, and remove duplicate, erroneous and missing data. Assume that the original data set is D raw , the cleaned data set is D clean , the cleaning process can be expressed by the following expression: D clean =f(D raw ), where f is the cleaning function. The cleaned data is preprocessed and Z-score is used to detect outliers Z = (x-μ) / σ, where x is a single data point, μ is the mean, and σ is the standard deviation. If |Z| is greater than a certain threshold (such as 3), x may be considered an outlier and removed. The cleaned and preprocessed data needs to be quality assessed to ensure the accuracy, completeness, and consistency of the data.

[0012] Furthermore, the cloud computing resource pool determines the demand scale and performance requirements of the campus and organization for computing, storage and network resources, and selects appropriate servers, storage devices and network devices. The performance indicators of the server can be expressed in terms of the number of CPU cores. cpu , memory capacity M, storage capacity S, use virtualization software to virtualize physical server, storage and network resources into multiple logical resources. Assume that the computing capacity of the physical server is P compute, the number of virtual machines that can be allocated after virtualization is V, and the computing power of each virtual machine is V compute , then P c ompute=V×V c ompute, integrating virtualized server, storage, and network resources into a resource pool. The total computing capacity of the resource pool can be expressed as Where n is the number of virtual machines, and the strategy for dynamically allocating and reclaiming computing resources based on actual demand is determined according to the task priority P. priority and resource requirements R demand To allocate resources, the amount of resources allocated is R allocated The formula R allocated =f(P priority ,R demand ) calculation, where f is the allocation function, establish a monitoring and management system for resource pool usage, and monitor resource utilization U in real time, such as calculating resource utilization Storage resource utilization Flexibly expand the scale of the resource pool based on actual usage and changes in demand.

[0013] Furthermore, the dynamic resource scheduling engine collects load information of data processing tasks in real time, including CPU usage U cpu 、Memory usage U memory , Network bandwidth usage U network Record resource utilization, such as calculating resource utilization R compute_utilization , Storage resource utilization R s torage_utilization, count the number of user requests Q, use historical data, and establish a prediction model for load and user request volume through time series analysis or machine learning methods. Assume that the predicted load is L predicted (t), can be based on past load data L pust (t-Δt) establishes a model, such as L predicted (t) = f(L past (t-Δt)), where f is the prediction function.

[0014] Furthermore, the dynamic resource scheduling engine selects a suitable intelligent algorithm, such as the policy function π(s,a) in reinforcement learning. In reinforcement learning, action a is selected according to the current state s to optimize resource allocation. In genetic algorithms, selection, crossover and mutation operations are performed by calculating the fitness value of individuals. According to the prediction results and algorithm output, the adjustment plan of resource allocation is determined. Assuming that the current resource configuration is R current , the amount of resources that need to be adjusted is ΔR, then the new resource configuration is R new =R current+ΔR, when the load exceeds a certain threshold T high When the peak period is expected, the resource supply is automatically increased. The amount of increased resources can be calculated by the formula ΔR high =k high ×(L current -T high ) calculation, where k high is the peak resource adjustment factor.

[0015] Furthermore, the dynamic resource scheduling engine is configured to low When the off-peak period is predicted, the resource usage is automatically reduced. The amount of reduced resources can be expressed as ΔR low =k low ×(T low -L current ), where k low is the off-peak resource adjustment coefficient, and continuously monitors system performance indicators, such as response time T response , throughput T throughput , adjust the algorithm parameters and resource allocation strategies according to the performance evaluation results to form a feedback mechanism.

[0016] Furthermore, the intelligent load balancer integrates common load balancing algorithms, such as the round-robin algorithm, the weighted round-robin algorithm, and the least connection algorithm, to perform characteristic analysis on the processing tasks to be assigned, such as the computational complexity C of the task. task , Data size D t ask, real-time monitoring of the status information of each node, including the current load, real-time monitoring of the status information of each node, including the current load Processing capacity node , according to the task characteristics and system status, select the appropriate load balancing algorithm. For example, if the weighted polling algorithm is used, the weight W of the node node It can be calculated based on the processing capacity and current load of the node.

[0017] Furthermore, the intelligent load balancer distributes tasks to different nodes according to the selected algorithm. Assuming there are n nodes in total, the number of tasks distributed to the i-th node is T i , then the total number of tasks T total Meet T total =∑ i＝1 T, continuously monitor the node load, if it is found that a node load exceeds the threshold L threshoid , then the tasks of the node are reallocated, and the number of reallocated tasks can be expressed as T reallocate =T ovenloaded -L tlireshold , where T overloadedis the number of tasks that overload the node.

[0018] Furthermore, the intelligent load balancer collects relevant data from various components and modules of the system, including system operation status parameters, resource usage, storage resource usage and task processing progress, processes and calculates the collected data, and obtains relevant indicators and statistical information. The system resource utilization rate can be calculated as: resource utilization rate = resource usage / total resource amount, that is, U resource =R used / R total , the task completion rate can be expressed as: task completion rate = number of completed tasks / total number of tasks, that is, R taskcompletion =T completed / T total , designed an intuitive monitoring interface, using charts, tables and real-time data display methods to present system operation status, resource usage and task processing progress information.

[0019] Furthermore, the intelligent load balancer provides administrators with an operation interface for remote configuration, task scheduling, and resource optimization, ensuring that the data in the monitoring interface can be updated in real time or regularly to reflect the latest status of the system. Assuming that the update interval is t update When the system operation status and resource usage exceed the set threshold, an alarm is triggered to notify the administrator. The alarm trigger condition can be expressed as: resource >T threshoid When the system is in use, an alarm is triggered. Based on the administrator's feedback and the actual operation effect of the system, the performance of the visual monitoring and management platform is evaluated, and corresponding improvements and optimizations are made.

[0020] Compared with the existing technology, a cloud computing-based smart campus big data collection and management system has the following beneficial effects:

[0021] 1. The system realizes the virtualization and dynamic allocation of computing resources, storage resources and network resources by building a cloud computing resource pool. This design enables the system to flexibly expand the scale of the resource pool according to actual needs, ensuring stable operation in high-concurrency scenarios. At the same time, the dynamic resource scheduling engine can automatically adjust resource configuration according to real-time monitoring of load conditions, further enhancing the flexibility and responsiveness of the system.

[0022] 2. The system uses the Internet of Things (IoT) technology, sensor networks and API interfaces through the big data collection module to achieve real-time and comprehensive collection of multi-source heterogeneous data on campus. Through data cleaning and preprocessing, the system can effectively remove duplicate, erroneous and missing data and improve data quality. In addition, the visual monitoring and management platform provides an intuitive monitoring interface that supports administrators to perform remote configuration, task scheduling and resource optimization operations, thereby greatly improving the efficiency of data management and operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 An operation flow chart of a smart campus big data collection and management system based on cloud computing. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.

[0026] Embodiment 1

[0027] (1) Deploy high-precision, low-energy sensors at key locations in the classroom. These sensors can monitor the environmental parameters in the classroom in real time to ensure a comfortable and healthy teaching environment. Introduce IoT devices such as smart blackboards, smart desks, and student behavior analysis cameras to complement traditional teaching equipment in the classroom. The smart blackboard supports touch writing and wireless screen projection. The student behavior analysis camera uses AI algorithms to identify students' listening status and interaction status, providing data support for teaching evaluation. Through the API interface, it is deeply connected with the existing teaching management system to achieve automatic synchronization and sharing of teaching data. At the same time, it also provides a data access interface for third-party teaching applications to enrich teaching resources and methods.

[0028] (2) Using the data cleaning function in the big data acquisition module, the collected multi-source heterogeneous data is deduplicated, corrected, and completed to ensure the accuracy and integrity of the data. Assume that the original data set is D = {d1, d2, …, d n}, the data set after deduplication is D unique ={du1 ,d u2 ,…,d um}, where m≤n, preprocess the cleaned data, including data format unification, data compression, and data encryption steps, to lay a solid foundation for subsequent data analysis and storage. Use Z-score and IQR statistical methods to detect and remove outliers in the data to prevent abnormal data from interfering with subsequent analysis results. Z-sec detects outliers: For data point x, the Z-score calculation formula is: Where μ is the mean of the data and σ is the standard deviation of the data.

[0029] (3) The dynamic resource scheduling engine collects key information such as equipment usage, network bandwidth usage, and user request volume in the classroom in real time, builds a comprehensive system load model, and uses historical data and advanced prediction models to predict the changing trend of resource demand in the classroom and make preparations in advance. Based on the prediction results and the current system status, the dynamic resource scheduling engine automatically adjusts the resource configuration of the cloud server cluster, such as increasing or decreasing the number of virtual machines and adjusting CPU and memory allocation, to ensure that stable and efficient services can be provided even during peak teaching periods.

[0030] (4) Through the visual monitoring and management platform, the real-time status of the classroom, equipment usage, and student behavior analysis information are intuitively displayed in the form of charts, tables, and real-time data. Administrators can understand the overall operation of the classroom at a glance. It supports administrators to remotely configure teaching equipment and adjust teaching strategies through mobile applications. Whether it is adjusting the brightness and temperature of the classroom lights or replacing teaching courseware, it can be easily achieved. Using the built-in data analysis tools of the platform, the collected data can be deeply mined and analyzed to discover potential problems and improvement points in the teaching process. For example, through student behavior analysis data, students with learning difficulties can be identified and targeted tutoring can be provided. Through environmental data analysis, classroom layout and equipment configuration can be optimized to improve teaching effectiveness. Alarm thresholds and rules can be set. When the system detects an abnormal situation, it automatically triggers an alarm to notify the administrator to handle it in time. At the same time, it supports multiple alarm methods such as SMS and email to ensure timely delivery of information.

[0031] Embodiment 2

[0032] (1) Student attendance sensors, teaching equipment status monitoring sensors, and environmental monitoring sensors are deployed in smart classrooms to comprehensively collect various data in the teaching process. Through the API interface, they are seamlessly connected to the existing teaching management system to obtain course schedules, student information, and teacher information data in real time, providing a basis for teaching data analysis. The collected data is cleaned to remove duplicate, erroneous, and missing data. The Z-score method is used to detect and eliminate outliers. For a data point x, its Z-score calculation formula is: Where μ is the mean of the data and σ is the standard deviation of the data to ensure data quality.

[0033] (2) Build a resource pool that includes high-performance virtualized servers, high-speed storage resources, and stable network resources. The resource pool supports dynamic expansion and can flexibly adjust its scale according to actual needs. It automatically adjusts the supply of computing resources according to the course schedule and changes in the number of students. For example, during exam weeks and peak course selection periods, the system automatically increases server resources to ensure that the teaching system can continue to run stably in high-concurrency scenarios. Establish a monitoring and management system for resource pool usage, monitor key indicators such as resource utilization, response time, and throughput in real time, and ensure efficient use of resources.

[0034] (3) Using time series analysis or machine learning models, based on historical data and current system status, predict the usage trend of classroom resources in the future. According to the prediction results and the current system load, automatically adjust the resource allocation. Use time series analysis and machine learning models to predict the usage trend of classroom resources in the future. The prediction result can be expressed as P(t) = f(H, S), where P(t) represents the predicted usage at time t, H represents historical data, S represents the current system status, and f is the prediction function. During high-load periods and low-load periods, recycle idle resources to achieve dynamic balance of resources. Use reinforcement learning and continuously optimize resource allocation strategies. The strategy function of reinforcement learning is π(s, a), where s represents the system status and a represents the resource allocation action. Through continuous learning and optimization, the reward function R(s, a) is maximized to improve the overall performance of the system.

[0035] (4) Analyze the characteristics of teaching tasks, including computational complexity and data size, to provide a basis for task allocation. Monitor the load of each node in real time. Select a suitable load balancing algorithm based on task characteristics and system status. Intelligently allocate tasks to different nodes. Select a suitable load balancing algorithm based on task characteristics and system status. In the weighted polling algorithm, the weight of node i can be expressed as W i =f(C task ,D task ,L i ), where f is a function that calculates weights based on task characteristics and node load. If a node is found to be overloaded, the load balancing mechanism is automatically triggered to reallocate tasks to the node to ensure overall load balance of the system.

[0036] (5) Provides an intuitive monitoring interface that displays key information such as classroom resource usage, teaching task processing progress, and student attendance. It supports remote viewing and management by administrators and provides administrators with an operation interface for remote configuration, task scheduling, and resource optimization. It simplifies the operation and maintenance process and reduces operating costs. When the system operating status or resource usage exceeds the set threshold, it automatically triggers an alarm to notify the administrator to ensure that the problem is handled in a timely manner.

[0037] Embodiment 3

[0038] (1) Introduce AI-assisted data capture algorithms to automatically identify and adapt to the data formats and update frequencies of different service providers, reduce manual intervention, improve the automation and intelligence level of data collection, integrate stream processing frameworks to quickly process real-time data, achieve monitoring and early warning with a response time of seconds, ensure instant feedback on service quality, conduct sentiment analysis on user reviews, build a user satisfaction model, and build a refined user profile based on user behavior data to provide a basis for personalized recommendations and service optimization.

[0039] (2) Use containerization technology and orchestration tools to achieve dynamic scaling of computing resources, ensuring that resources are automatically added during peak demand and released during low demand, reducing costs while ensuring performance. Assume that the current resource configuration is R current , the amount of resources that need to be adjusted is ΔR, then the new resource configuration is R new =R current +ΔR, when the load exceeds a certain threshold T high When the peak period is expected, the resource supply is automatically increased. The amount of increased resources can be calculated by the formula ΔR high =k high ×(L current -T high ) calculation, where k high It is the peak resource adjustment coefficient, deploys multi-layer load balancers, effectively disperses access pressure, and realizes decoupling and independent deployment of services through microservice architecture, improves the fault tolerance and maintainability of the system, implements strict data isolation strategies, and ensures the security of data between different services.

[0040] (3) In addition to basic operating status and resource usage, we have added advanced views such as service dependency diagrams and abnormal event tracking diagrams to help administrators quickly locate the root cause of problems. We have combined machine learning algorithms to conduct predictive analysis of the system operating status, discover potential problems in advance and trigger warnings, reduce the probability of failures, and develop supporting mobile apps and mini-programs so that administrators can check the platform status anytime and anywhere and perform emergency operations, thereby improving the flexibility and response speed of operation and maintenance. We have integrated user feedback channels into the management platform to collect user opinions and suggestions in real time, and quickly transform them into product improvement and service optimization measures through data analysis.

[0041] (4) Based on user portraits and behavioral data, collaborative filtering, deep learning and other methods are used to provide users with personalized dining recommendations, shopping suggestions, etc., to improve user experience, and to provide customized service interfaces for special groups on campus, such as barrier-free access and multi-language support, to enhance the inclusiveness and convenience of the platform.

[0042] (5) Establish a data-driven decision-making mechanism, regularly analyze platform operation data, evaluate service effectiveness, guide product iteration and function optimization, encourage teachers and students to participate in the construction and improvement of the platform, collect ideas and suggestions through user communities and suggestion boxes, and form an ecological environment for co-creation and sharing.

[0043] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A cloud computing-based smart campus big data collection and management system, characterized in that: The system includes a big data acquisition module, a cloud computing resource pool, a dynamic resource scheduling engine, an intelligent load balancer, and a visual monitoring and management platform; The big data collection module: uses the Internet of Things (IoT) technology, sensor networks and API interfaces to achieve real-time and comprehensive collection of various teaching, management, safety and life multi-source heterogeneous data on campus, supports data cleaning and preprocessing, and ensures data quality; The cloud computing resource pool: builds a resource pool based on cloud computing technology, including virtualized servers, storage resources and network resources. The resource pool can be flexibly expanded according to actual needs and supports dynamic allocation and recovery of computing resources; The dynamic resource scheduling engine: by real-time monitoring of the load of data processing tasks, resource utilization and user request volume, it uses advanced prediction models and intelligent algorithms to automatically adjust the resource allocation of the cloud server cluster, automatically increase resource supply during peak periods, and reduce resource usage during off-peak periods, ensuring that the system can maintain high performance even in high-concurrency scenarios; The intelligent load balancer integrates multiple load balancing algorithms and intelligently distributes processing tasks to different nodes based on task characteristics and system status, effectively avoiding overload of a single node and improving overall data processing capabilities and system scalability; The visual monitoring and management platform provides an intuitive monitoring interface to display system operation status, resource usage, and task processing progress information, supports administrators to perform remote configuration, task scheduling, and resource optimization operations, simplifies the operation and maintenance process, and reduces operating costs.

2. According to the cloud computing-based smart campus big data collection and management system described in claim 1, it is characterized in that: The big data collection module determines the specific data sources in the fields of teaching, management, security and life on campus that need to collect data. According to the data collection requirements, sensors and IoT devices are deployed at appropriate locations on campus, and various existing information systems on campus are connected and configured with API interfaces. The data obtained through the API interface can be expressed as D api (t) Use the deployed sensors, IoT devices and API interfaces to collect data in real time, clean the collected data, and remove duplicate, erroneous and missing data. Assume that the original data set is D raw , the cleaned data set is D clean , the cleaning process can be expressed by the following expression: D clean =f(D raw ), where f is the cleaning function. The cleaned data is preprocessed and Z-score is used to detect outliers Z = (x-μ) / σ, where x is a single data point, μ is the mean, and σ is the standard deviation. If |Z| is greater than a certain threshold (such as 3), x may be considered an outlier and removed. The cleaned and preprocessed data needs to be quality assessed to ensure the accuracy, completeness, and consistency of the data.

3. According to the cloud computing-based smart campus big data collection and management system of claim 1, it is characterized in that: The cloud computing resource pool determines the scale and performance requirements of the campus and organization for computing, storage and network resources, and selects appropriate servers, storage devices and network devices. The performance indicators of the server can be expressed as the number of CPU cores. cpu , memory capacity M, storage capacity S, use virtualization software to virtualize physical server, storage and network resources into multiple logical resources. Assume that the computing capacity of the physical server is P compute , the number of virtual machines that can be allocated after virtualization is V, and the computing power of each virtual machine is V compute , then P c ompute=V×V c ompute, integrating virtualized server, storage, and network resources into a resource pool. The total computing capacity of the resource pool can be expressed as Where n is the number of virtual machines, and the strategy for dynamically allocating and reclaiming computing resources based on actual demand is determined according to the task priority P. priority and resource requirements R demand To allocate resources, the amount of resources allocated is R allocated The formula R allocated =f(P priority ,R demand ) calculation, where f is the allocation function, establish a monitoring and management system for resource pool usage, and monitor resource utilization U in real time, such as calculating resource utilization Storage resource utilization Flexibly expand the scale of the resource pool based on actual usage and changes in demand.

4. According to the cloud computing-based smart campus big data collection and management system of claim 1, it is characterized in that: The dynamic resource scheduling engine collects load information of data processing tasks in real time, including CPU usage U cpu 、Memory usage U memory , Network bandwidth usage U network Record resource utilization, such as calculating resource utilization R compute_utilization , Storage resource utilization R s torage_utilization, count the number of user requests Q, use historical data, and establish a prediction model for load and user request volume through time series analysis and machine learning methods. Assume that the predicted load is L predicted (t), can be based on past load data L pust (t-Δt) establishes a model, such as L predicted (t) = f(L past (t-Δt)), where f is the prediction function.

5. According to the cloud computing-based smart campus big data collection and management system of claim 4, it is characterized in that: The dynamic resource scheduling engine selects a suitable intelligent algorithm, such as the policy function π(s,a) in reinforcement learning. In reinforcement learning, action a is selected according to the current state s to optimize resource allocation. In genetic algorithms, selection, crossover and mutation operations are performed by calculating the fitness value of individuals. According to the prediction results and algorithm output, the adjustment plan of resource allocation is determined. Assuming that the current resource configuration is R current , the amount of resources that need to be adjusted is ΔR, then the new resource configuration is R new =R current +ΔR, when the load exceeds a certain threshold T high When the peak period is expected, the resource supply is automatically increased. The amount of increased resources can be calculated by the formula ΔR high =k high ×(L current -T high ) calculation, where k high is the peak resource adjustment factor.

6. According to the cloud computing-based smart campus big data collection and management system of claim 5, it is characterized in that: The dynamic resource scheduling engine is activated when the load is lower than a certain threshold T low When the off-peak period is predicted, the resource usage is automatically reduced. The amount of reduced resources can be expressed as ΔR low =k low ×(T low -L current ), where k low is the off-peak resource adjustment coefficient, and continuously monitors system performance indicators, such as response time T response , throughput T throughput , adjust the algorithm parameters and resource allocation strategies according to the performance evaluation results to form a feedback mechanism.

7. According to the cloud computing-based smart campus big data collection and management system of claim 1, it is characterized in that: The intelligent load balancer integrates common load balancing algorithms, such as the round-robin algorithm, the weighted round-robin algorithm, and the least connection algorithm, and performs characteristic analysis on the processing tasks to be assigned, such as the computational complexity C of the task. task , Data size D t ask, real-time monitoring of the status information of each node, including the current load, real-time monitoring of the status information of each node, including the current load Processing capacity node , according to the task characteristics and system status, select the appropriate load balancing algorithm. For example, if the weighted polling algorithm is used, the weight W of the node node It can be calculated based on the processing capacity and current load of the node.

8. According to the cloud computing-based smart campus big data collection and management system of claim 7, it is characterized in that: The intelligent load balancer distributes tasks to different nodes according to the selected algorithm. Assuming there are n nodes in total, the number of tasks allocated to the i-th node is T i , then the total number of tasks T total Meet T total =∑ i＝1 T, continuously monitor the node load, if it is found that a node load exceeds the threshold L threshoid , then the tasks of the node are reallocated, and the number of reallocated tasks can be expressed as T reallocate =T ovenloaded -L tlireshold , where T overloaded is the number of tasks that overload the node.

9. According to the cloud computing-based smart campus big data collection and management system of claim 1, it is characterized in that: The visual monitoring and management platform collects relevant data from various components and modules of the system, including system operation status parameters, resource usage, storage resource usage and task processing progress, processes and calculates the collected data, and obtains relevant indicators and statistical information. The system resource utilization rate can be calculated as: resource utilization rate = resource usage / total resource amount, that is, U resource =R used / R total , the task completion rate can be expressed as: task completion rate = number of completed tasks / total number of tasks, that is, R taskcompletion =T completed / T total , designed an intuitive monitoring interface, using charts, tables and real-time data display methods to present system operation status, resource usage and task processing progress information.

10. According to the cloud computing-based smart campus big data collection and management system of claim 9, it is characterized in that: The visual monitoring and management platform provides administrators with an operation interface for remote configuration, task scheduling, and resource optimization, ensuring that the data in the monitoring interface can be updated in real time or regularly to reflect the latest status of the system. Assume that the update interval is t update When the system operation status and resource usage exceed the set threshold, an alarm is triggered to notify the administrator. The alarm trigger condition can be expressed as: resource >T threshoid When the system is in use, an alarm is triggered. Based on the administrator's feedback and the actual operation effect of the system, the performance of the visual monitoring and management platform is evaluated, and corresponding improvements and optimizations are made.

Images