System, method and device for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning, processor and medium

Abstract

The invention relates to a system for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning, and the system comprises a capacity index collection module which regularly collects the index values of an information system; the water level calculation module is used for comparing and calculating each index value and the upper limit value thereof to obtain the water level value of each index; the dynamic water level adjusting module is used for calculating the capacity bearing upper limit change condition under the condition of resource allocation of different technical components and calculating to obtain the maximum contractible capacity of the system; and the capacity expansion and shrinkage decision module is used for monitoring the capacity health degree in real time and judging whether capacity expansion or capacity shrinkage is needed. By adopting the system, the method, the device, the processor and the computer readable storage medium for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning, accurate control and efficient management of the information system capacity are realized. Due to the dynamic water level adjusting capacity, the system can flexibly adjust resource allocation according to real-time changes of services, and the operation cost of the system is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the fields of information system management technology and operation and maintenance system technology, and in particular to the field of operation and maintenance service capacity management technology. Specifically, it refers to a system, method, device, processor and computer-readable storage medium thereof for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning. Background Art

[0002] Against the backdrop of the rapid evolution of information technology, information systems have become the core support for operations and digital transformation across various industries. Whether it's the business operations systems of large enterprises, government administration platforms, online service applications of internet companies, or transaction processing systems of financial institutions, the scale of information systems is growing exponentially, their architectures becoming increasingly complex, and the requirements for refined system capacity management are constantly increasing. How to scientifically and efficiently manage information system capacity to ensure stable and efficient operation during peak business periods to meet unexpected business demands, while also avoiding idle resources and waste during low business periods, has become a key issue that needs to be addressed in the field of information systems management.

[0003] Traditional capacity management methods typically employ single-dimensional monitoring and analysis, focusing solely on server CPU utilization, storage usage, or network bandwidth consumption, for example. However, this isolated management model fails to comprehensively assess the overall system capacity, making it difficult to timely predict and respond to dynamic changes in business needs. When business volume surges, the system may face performance bottlenecks, resulting in delayed or even interrupted service responses, severely impacting user experience and business continuity. Conversely, when business volume decreases, the significant idleness of computing, storage, and other resources not only wastes resources but also significantly increases operating costs. Therefore, there is an urgent need to build an intelligent, refined, and dynamically adaptive capacity management system to achieve comprehensive capacity assessment, precise monitoring, and intelligent optimization, ensuring the efficient operation of information systems and optimal resource allocation. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a system, method, device, processor and computer-readable storage medium thereof that realizes dynamic information system capacity management based on multi-dimensional evaluation and machine learning, which meets the requirements of high performance, good stability and a wide range of applications.

[0005] To achieve the above objectives, the system, method, device, processor, and computer-readable storage medium thereof for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning are as follows:

[0006] The system for dynamic information system capacity management based on multi-dimensional evaluation and machine learning has the following main features:

[0007] Capacity indicator collection module, used to regularly collect indicator values of the information system's business capacity, service interface capacity, and technical component capacity;

[0008] A water level calculation module is connected to the capacity index acquisition module and is used to compare and calculate each index value with its upper limit value to obtain the water level value of each index;

[0009] The dynamic water level adjustment module is connected to the capacity index acquisition module and is used to calculate the change in the capacity carrying limit under different technical component resource configurations based on the historical data of each capacity index, and calculate the maximum scalable capacity of the system;

[0010] The capacity expansion and contraction decision module is connected to the water level calculation module and the dynamic water level adjustment module, and is used to monitor the capacity health in real time, compare the capacity health calculated by the water level calculation module with the preset capacity expansion threshold in real time, and combine the analysis results of the dynamic water level adjustment module to determine whether expansion or contraction is needed.

[0011] Preferably, the capacity indicator collection module collects high-precision data on business capacity, service interface capacity, and technical component capacity at preset time intervals, performs preliminary cleaning through outlier detection, noise filtering, and data format standardization, and stores the processed data in a dedicated data warehouse.

[0012] Preferably, the water level calculation module includes:

[0013] A water level calculation unit, connected to the capacity indicator collection module, is used to compare the collected key performance indicators with pre-configured thresholds to calculate the CPU water level value, storage occupancy water level value, and database query load water level value;

[0014] Capacity health calculation unit, used to calculate the overall capacity health based on the water level values of each indicator;

[0015] Visualization and monitoring unit, used to display water level values and capacity health data in a variety of visual ways on the management console;

[0016] The historical data storage and analysis unit is used to display the calculation results in real time on the system management platform and store them in the historical database.

[0017] Preferably, the water level calculation unit calculates the CPU water level value, the storage occupancy water level value and the database query load water level value, specifically as follows:

[0018] The CPU water level value is calculated according to the following formula:

[0019]

[0020] The storage utilization watermark is calculated using the following formula:

[0021]

[0022] The database query load water level value is calculated according to the following formula:

[0023]

[0024] Preferably, the capacity health calculation unit calculates the capacity health by arithmetic average method, weighted average method or exponentially weighted moving average method.

[0025] Preferably, the visualization and monitoring unit displays the current water level status of each key indicator through a real-time dashboard; displays the changing trend of water level value and capacity health over time through a trend analysis chart; and triggers an automatic alarm when the water level value exceeds the set threshold.

[0026] Preferably, the historical data storage and analysis unit stores time series data through a time series database and realizes big data storage.

[0027] Preferably, the dynamic water level regulation module regularly extracts various capacity index data from the historical database to construct a dynamic capacity prediction model. The various capacity index data include historical monitoring data, business load data, environmental configuration data and external factor data.

[0028] Preferably, if the expansion / contraction decision module detects that the capacity health H exceeds the expansion threshold T, the system load continues to grow with no downward trend in the short term, and the key resources continue to be highly occupied, the expansion decision is triggered; the early warning mechanism is triggered, the sub-indicators are analyzed in detail, and expansion strategies for computing expansion, storage expansion, network expansion or database optimization are formulated respectively, an expansion plan is generated, and automatic expansion or manual approval of expansion is performed.

[0029] Preferably, if the expansion and contraction decision module detects that the capacity health is lower than the contraction threshold L, the system load continues to decline and there is no peak expected in the short term, and the resource redundancy rate is high, the contraction decision is triggered; the resource utilization rate is analyzed, the dynamic water level adjustment is calculated, and a contraction strategy is formulated, that is, computing resources are reduced, storage and network are optimized, and automatic contraction or manual approval of contraction is performed.

[0030] The main feature of the method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning using the above system is that the method includes the following steps:

[0031] (1) Regularly collect the index values of the information system's business capacity, service interface capacity, and technical component capacity;

[0032] (2) Compare and calculate each indicator value with its upper limit value to obtain the water level value of each indicator;

[0033] (3) Calculate the overall capacity health of the information system based on the water level values of each indicator;

[0034] (4) Compare the capacity health with the preset capacity health expansion line to obtain the output result of whether expansion is needed, and determine the technical components that need to be expanded based on the situation of each sub-indicator;

[0035] (5) Based on the historical data of each capacity indicator, calculate the change in the capacity carrying limit under different technical component resource configurations, and obtain the maximum amount of resources that can be shrunk for different technical components while meeting the capacity threshold requirements;

[0036] (6) Calculate the overall capacity health of the information system based on the water level values of each indicator.

[0037] The main features of the device for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning are as follows:

[0038] a processor configured to execute computer-executable instructions;

[0039] A memory stores one or more computer-executable instructions, which, when executed by the processor, implement the various steps of the method for dynamic information system capacity management based on multi-dimensional evaluation and machine learning.

[0040] The main feature of the processor for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning is that the processor is configured to execute computer-executable instructions. When the computer-executable instructions are executed by the processor, the various steps of the above-mentioned method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning are implemented.

[0041] The computer-readable storage medium is characterized in that a computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for dynamic information system capacity management based on multi-dimensional evaluation and machine learning.

[0042] The system, method, device, processor and computer-readable storage medium thereof for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning of the present invention are adopted, and accurate control and efficient management of information system capacity are realized by virtue of comprehensive capacity evaluation methods and intelligent dynamic water level adjustment capabilities. Through multi-dimensional capacity evaluation, potential capacity bottlenecks and resource waste problems in the system can be discovered in a timely and accurate manner, providing a scientific basis for system optimization and upgrading. The dynamic water level adjustment capability enables the system to flexibly adjust resource allocation according to real-time changes in the business, ensure high performance and stability of the system during business peaks, and maximize resource utilization during business troughs, effectively reducing the operating costs of the system. At the same time, the method, device, platform and storage medium proposed in the present invention are highly versatile and scalable, and can be widely used in various information systems, whether they are traditional enterprise-level application systems or emerging cloud computing and big data platforms. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Schematic diagram of the information system architecture model of the system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning in the present invention.

[0044] Figure 2 This is a flow chart of capacity assessment for a system that implements dynamic information system capacity management based on multi-dimensional assessment and machine learning according to the present invention. DETAILED DESCRIPTION

[0045] In order to more clearly describe the technical content of the present invention, further description is given below in conjunction with specific embodiments.

[0046] The system of the present invention for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning includes:

[0047] Capacity indicator collection module, used to regularly collect indicator values of the information system's business capacity, service interface capacity, and technical component capacity;

[0048] A water level calculation module is connected to the capacity index acquisition module and is used to compare and calculate each index value with its upper limit value to obtain the water level value of each index;

[0049] The dynamic water level adjustment module is connected to the capacity index acquisition module and is used to calculate the change in the capacity carrying limit under different technical component resource configurations based on the historical data of each capacity index, and calculate the maximum scalable capacity of the system;

[0050] The capacity expansion and contraction decision module is connected to the water level calculation module and the dynamic water level adjustment module, and is used to monitor the capacity health in real time, compare the capacity health calculated by the water level calculation module with the preset capacity expansion threshold in real time, and combine the analysis results of the dynamic water level adjustment module to determine whether expansion or contraction is needed.

[0051] As a preferred embodiment of the present invention, the capacity indicator collection module performs high-precision data collection on business capacity, service interface capacity, and technical component capacity at preset time intervals, and performs preliminary cleaning through outlier detection, noise filtering, and data format standardization, and stores the processed data in a dedicated data warehouse.

[0052] As a preferred embodiment of the present invention, the water level calculation module includes:

[0053] A water level calculation unit, connected to the capacity indicator collection module, is used to compare the collected key performance indicators with pre-configured thresholds to calculate the CPU water level value, storage occupancy water level value, and database query load water level value;

[0054] Capacity health calculation unit, used to calculate the overall capacity health based on the water level values of each indicator;

[0055] Visualization and monitoring unit, used to display water level values and capacity health data in a variety of visual ways on the management console;

[0056] The historical data storage and analysis unit is used to display the calculation results in real time on the system management platform and store them in the historical database.

[0057] As a preferred embodiment of the present invention, the water level calculation unit calculates the CPU water level value, the storage occupancy water level value and the database query load water level value, specifically as follows:

[0058] The CPU water level value is calculated according to the following formula:

[0059]

[0060] The storage utilization watermark is calculated using the following formula:

[0061]

[0062] The database query load water level value is calculated according to the following formula:

[0063]

[0064] As a preferred embodiment of the present invention, the capacity health calculation unit calculates the capacity health by arithmetic average method, weighted average method or exponentially weighted moving average method.

[0065] As a preferred embodiment of the present invention, the visualization and monitoring unit displays the current water level status of each key indicator through a real-time dashboard; displays the changing trend of water level value and capacity health over time through a trend analysis chart; and triggers an automatic alarm when the water level value exceeds the set threshold.

[0066] As a preferred embodiment of the present invention, the historical data storage and analysis unit stores time series data through a time series database and realizes big data storage.

[0067] As a preferred embodiment of the present invention, the dynamic water level regulation module regularly extracts various capacity indicator data from the historical database to build a dynamic capacity prediction model. The various capacity indicator data include historical monitoring data, business load data, environmental configuration data and external factor data.

[0068] As a preferred embodiment of the present invention, if the expansion / contraction decision module detects that the capacity health H exceeds the expansion threshold T, the system load continues to grow with no downward trend in the short term, and the key resources continue to be highly occupied, the expansion decision is triggered; the early warning mechanism is triggered, the sub-indicators are analyzed in detail, and expansion strategies for computing expansion, storage expansion, network expansion or database optimization are formulated respectively, an expansion plan is generated, and automatic expansion or manual approval of expansion is performed.

[0069] As a preferred embodiment of the present invention, if the expansion and contraction decision module detects that the capacity health is lower than the contraction threshold L, the system load continues to decline and there is no peak expected in the short term, and the resource redundancy rate is high, then the contraction decision is triggered; the resource utilization rate is analyzed, the dynamic water level adjustment is calculated, and a contraction strategy is formulated, that is, computing resources are reduced, storage and network are optimized, and automatic contraction or manual approval of contraction is performed.

[0070] The present invention provides a method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning using the above system, wherein the method comprises the following steps:

[0071] (1) Regularly collect the index values of the information system's business capacity, service interface capacity, and technical component capacity;

[0072] (2) Compare and calculate each indicator value with its upper limit value to obtain the water level value of each indicator;

[0073] (3) Calculate the overall capacity health of the information system based on the water level values of each indicator;

[0074] (4) Compare the capacity health with the preset capacity health expansion line to obtain the output result of whether expansion is needed, and determine the technical components that need to be expanded based on the situation of each sub-indicator;

[0075] (5) Based on the historical data of each capacity indicator, calculate the change in the capacity carrying limit under different technical component resource configurations, and obtain the maximum amount of resources that can be shrunk for different technical components while meeting the capacity threshold requirements;

[0076] (6) Calculate the overall capacity health of the information system based on the water level values of each indicator.

[0077] The device for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning of the present invention comprises:

[0078] a processor configured to execute computer-executable instructions;

[0079] A memory stores one or more computer-executable instructions, which, when executed by the processor, implement the various steps of the method for dynamic information system capacity management based on multi-dimensional evaluation and machine learning.

[0080] The processor of the present invention is used to implement dynamic information system capacity management based on multi-dimensional evaluation and machine learning, wherein the processor is configured to execute computer-executable instructions. When the computer-executable instructions are executed by the processor, the various steps of the above-mentioned method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning are implemented.

[0081] The computer-readable storage medium of the present invention stores a computer program thereon, which can be executed by a processor to implement the various steps of the above-mentioned method for dynamic information system capacity management based on multi-dimensional evaluation and machine learning.

[0082] This invention relates to the fields of information system management and operations and maintenance systems, and in particular to capacity management technology for operations and maintenance services. It aims to address the complex challenges faced in information system capacity management and build a comprehensive, intelligent, and efficient information system capacity management system. This system, encompassing methods, devices, platforms, and storage media, utilizes multi-dimensional assessment and machine learning techniques to achieve precise capacity monitoring, dynamic resource optimization, and intelligent capacity expansion decisions, meeting the digital era's demands for high-performance, high-reliability, and optimized resource utilization in information systems.

[0083] Information systems rely on a variety of technical components to provide basic capabilities such as computing and storage to support the operation of application services, and provide various business functions to the outside world through service interfaces. Service interfaces are arranged through business logic to form complete business service scenarios. The present invention comprehensively evaluates system capacity based on three core dimensions: business capacity, service interface capacity, and technical component capacity. By collecting key system indicators and comparing them with preset upper limits, the water level value is calculated to measure the overall capacity health, and based on the comparison with the expansion threshold, it is determined whether the system or technical component needs to be expanded.

[0084] Furthermore, the present invention incorporates machine learning technology to analyze historical operational data to predict changes in capacity limits under different technical component resource configurations, enabling dynamic water level control. This invention can accurately monitor system capacity status, promptly identify capacity bottlenecks and resource redundancies, and automatically optimize resource allocation based on changing business needs, effectively reducing operating costs. Its high versatility and scalability provide strong support for the development of information system management technology.

[0085] The present invention provides a dynamic information system capacity management device based on multi-dimensional evaluation and machine learning, comprising:

[0086] (1) Capacity index collection module, used to regularly collect the index values of the information system's business capacity, service interface capacity, and technical component capacity;

[0087] (2) Water level calculation module, which is used to compare each indicator value with its upper limit value to obtain the water level value of each indicator. The specific calculation method is: where c i is the water level value of the i-th indicator, m i is the index value collected, M i is the maximum value of the corresponding indicator item, which is obtained from stress testing;

[0088] (3) Capacity health calculation unit, used to calculate the overall capacity health based on the water level values of each indicator. The calculation formula is:

[0089]

[0090] C is the overall health of the information system, c i The health of each capacity indicator item;

[0091] (4) The capacity expansion / contraction decision module is used to compare the capacity health with the preset capacity health expansion line, obtain the output result of whether capacity expansion is needed, and determine the technical components that need to be expanded based on the situation of each sub-indicator;

[0092] (5) Dynamic water level adjustment module, which is used to calculate the change of capacity carrying limit under different technical component resource configurations based on the historical data of each capacity indicator, and to obtain the maximum amount of resources that can be shrunk under the premise of meeting the capacity threshold requirements of different technical components.

[0093] The present invention provides a dynamic information system capacity management method based on multi-dimensional evaluation and machine learning, the method comprising the following steps:

[0094] (1) Regularly collect the index values of the information system's business capacity, service interface capacity, and technical component capacity;

[0095] (2) Compare and calculate each indicator value with its upper limit value to obtain the water level value of each indicator;

[0096] (3) Calculate the overall capacity health of the information system based on the water level values of each indicator;

[0097] (4) Compare the capacity health with the preset capacity health expansion line to obtain the output result of whether expansion is needed, and determine the technical components that need to be expanded based on the situation of each sub-indicator;

[0098] (5) Based on the historical data of each capacity indicator, calculate the change in the capacity carrying limit under different technical component resource configurations, and obtain the maximum amount of resources that can be shrunk for different technical components while meeting the capacity threshold requirements;

[0099] (6) Calculate the overall capacity health of the information system based on the water level values of each indicator.

[0100] The present invention aims to overcome the limitations of existing technologies in information system capacity management and provide an intelligent and efficient information system capacity management method, device, platform and storage medium to achieve comprehensive assessment, accurate monitoring and dynamic optimization of system capacity, ensure reasonable allocation of resources, and improve system stability and reliability.

[0101] To achieve the above objectives, the present invention proposes a method, device, platform, and storage medium for dynamic information system capacity management based on multi-dimensional evaluation and machine learning. The specific contents are as follows:

[0102] The information system managed by this invention is a highly integrated and hierarchical architecture. At the bottom layer, multiple technical components, including various server hardware, storage devices, and the basic service modules of the PaaS platform, work together to provide the entire information system with robust computing and storage capabilities, as well as rich PaaS capabilities. Above these technical components, the application service component is responsible for encapsulating and transforming the underlying resources, providing a variety of external functions in the form of interface services. These multiple interface services are cleverly orchestrated based on complex business logic, ultimately forming a rich and diverse business service scenario that directly serves users and meets business needs.

[0103] like Figure 1 As shown in the figure, the information system architecture diagram shows the hierarchical relationship and data flow between technical components, application service components, and business service scenarios. The underlying server hardware, storage devices, and PaaS platform basic service modules work closely together to provide support for the upper layer; application service components interact with technical components through interface services to implement resource calls; business service scenarios are composed of multiple interface services combined according to business logic to provide services to users;

[0104] Capacity assessment uses a comprehensive approach based on three dimensions: business capacity, service interface capacity, and technical component capacity. At the business capacity level, a wide variety of metrics are considered, including not only business request volume and business data processing volume, but also concurrent processing capabilities and business data growth trends. Comprehensive analysis of these metrics allows for a precise understanding of the actual business capacity requirements for the system. Service interface capacity also considers factors such as interface response time and interface throughput. Technical component capacity metrics, in addition to conventional CPU utilization, memory utilization, and storage utilization, also focus on network bandwidth usage and storage read / write bandwidth utilization. In practice, sensors and monitoring tools deployed at key system nodes regularly and accurately collect these three capacity metrics. The frequency of collection can be flexibly configured based on the system's business characteristics and real-time requirements. After collecting the metrics, they are compared with pre-set upper limits. These upper limits are not fixed but are determined through regular system performance and capacity testing and dynamically adjusted. This comparison results in a watermark for each metric, which intuitively reflects the current level of the metric's proximity to the upper capacity limit. Then, using a scientific algorithm, the overall capacity health is calculated based on the watermark values of each indicator, using methods such as the maximum, average, or weighted average. The weighted average algorithm assigns weights to different indicators based on their impact on overall system performance. For example, for a real-time trading system, the weight of business request volume might be set to 0.4, interface response time to 0.3, and CPU utilization to 0.2. Finally, the calculated capacity health is compared with the preset capacity health expansion threshold. This expansion threshold is also dynamically adjusted based on historical system data, business development expectations, and industry standards. This comparison results in an output indicating whether capacity expansion is necessary. Furthermore, when determining the need for capacity expansion, an in-depth analysis of each sub-indicator is conducted to precisely identify the technical components requiring expansion, thereby determining the location of resource expansion. For example, if the business capacity watermark value is consistently high while the service interface capacity and technical component capacity watermark values are relatively normal, further analysis may reveal insufficient computing resources in the business processing module. In this case, the servers responsible for business processing can be targeted for capacity expansion, either by increasing the number of CPU cores or expanding memory capacity.

[0105] Combine Figure 2The capacity assessment flowchart details the specific operations and data interactions for each step, including indicator collection, water level calculation, dynamic water level adjustment calculation, and expansion decision-making. It begins with regular indicator value collection, calculates the water level by comparing it with the upper limit, calculates capacity health based on the water level, and finally compares it with the expansion line to determine whether to expand capacity and which components to expand. The dynamic water level adjustment module, on the other hand, calculates the maximum scalable capacity of the system based on the accumulated historical values of capacity indicators and outputs this value to the expansion decision module. The expansion decision module outputs the expansion and contraction recommendations derived from the previous steps to the administrator for actual expansion and contraction operations.

[0106] In the specific implementation of the present invention, the following structure is specifically included:

[0107] (1.1) Capacity indicator collection module: This module consists of a set of highly specialized monitoring tools distributed across key nodes of the information system, including computing resources (servers, virtual machines, containers, etc.), storage devices (SAN, NAS, distributed storage, etc.), network infrastructure (switches, load balancers, firewalls, etc.), and application layers (microservices, API gateways, databases, etc.).

[0108] This module supports multiple monitoring methods, including the system's native monitoring APIs (such as the Linux / proc file system and Windows Performance Counters), which can directly collect information about the usage of underlying resources such as CPU, memory, disk I / O, and network throughput. It can also integrate with mature third-party monitoring solutions in the industry, such as Metricbeat, Prometheus, Zabbix, and Nagios, to achieve refined collection and monitoring of a wider range of system operation indicators (such as container resource utilization, microservice call links, transaction response time, database query load, etc.).

[0109] The data collection process uses a distributed architecture that supports on-demand deployment and elastic expansion. The collection module collects high-precision data on business capacity, service interface capacity, and technical component capacity at preset intervals (the collection frequency can be adjusted based on static policies or dynamic adaptive algorithms). To improve data quality, the module has built-in anomaly detection and data preprocessing mechanisms. Before data storage, these include:

[0110] Outlier detection: Isolation forest machine learning models remove abnormal data points;

[0111] Noise filtering: Use the time series smoothing algorithm of weighted sliding average to reduce data noise and improve data stability;

[0112] Data format standardization: Unify the monitoring data formats from different sources to ensure cross-system compatibility.

[0113] The processed data is stored in a dedicated data warehouse, and a time series database can be used to provide reliable data support for subsequent capacity assessment, trend analysis, and intelligent prediction.

[0114] (1.2) Water Level Calculation Module: This module is responsible for reading cleaned and organized core indicator data from the data warehouse and performing efficient calculations and analysis to assess the capacity health of the information system in real time. Its core functions include water level calculation, capacity health assessment, visual monitoring, and historical data storage and analysis.

[0115] (1.2.1) Water level calculation mechanism:

[0116] This module compares collected key performance indicators (such as CPU utilization, memory usage, network bandwidth usage, and database query response time) with pre-configured thresholds and calculates the watermark for each indicator, reflecting the current system capacity utilization relative to the set upper limit. This calculation utilizes efficient data processing algorithms to ensure that even in environments with massive amounts of data, computations can be completed in record time.

[0117] a.CPU water level calculation formula:

[0118]

[0119] b. Storage occupancy rate watermark:

[0120]

[0121] c. Database query load water level value:

[0122]

[0123] The calculated water level value will be updated in real time to the system monitoring interface and pushed to the operation and maintenance management platform so that administrators can intuitively grasp the operating status of each key indicator and promptly identify potential performance bottlenecks or resource shortages.

[0124] (1.2.2) Capacity Health Assessment:

[0125] After obtaining the water level values of each indicator, the module further calculates the overall capacity health of the system to comprehensively evaluate the resource utilization of the system. The calculation method of capacity health can be flexibly adjusted and supports multiple calculation models:

[0126] a. Arithmetic average method (applicable to systems with similar weights):

[0127]

[0128] Among them, W iRepresents the water level value of the i-th monitoring system, and N is the total number of indicators.

[0129] b. Weighted average method (applicable to systems where different indicators have different influence weights):

[0130]

[0131] Among them, w i is the water level value of the i-th monitoring indicator, ensuring that reasonable impact factors are allocated to different types of resources.

[0132] c. Exponentially Weighted Moving Average (EWMA) method (suitable for smoothing short-term fluctuations and highlighting long-term trends):

[0133] H t =αW t +(1-α)H t-1 ...(1.6)

[0134] Among them, α is a smoothing factor, which is used to adjust the influence weight of new data and historical data.

[0135] The calculated capacity health can serve as an important reference for system expansion and resource adjustment.

[0136] (1.2.3) Visualization and monitoring of calculation results:

[0137] After the calculation is complete, the module displays the water level and capacity health data in a variety of visualization methods on the management console, including:

[0138] Real-time dashboard: displays the current status of key indicators and uses a color-coded warning mechanism to help administrators quickly identify abnormal situations.

[0139] Trend analysis chart: Displays the changing trends of water level values and capacity health over time. It can be combined with time window analysis (Sliding Window) to provide short-term and long-term trend insights.

[0140] Abnormal alarm system: When the water level exceeds the set threshold, an automatic alarm is triggered and the operation and maintenance team is notified via email, SMS, etc.

[0141] (1.2.4) Historical data storage and analysis:

[0142] The calculation results are not only displayed in real time on the system management platform, but also stored in the historical database for subsequent trend analysis, capacity forecasting, and comparative research. Supported data storage methods include:

[0143] a. Time series database: Efficiently stores time series data and facilitates retrospective queries.

[0144] b. Big data storage: Suitable for large-scale data analysis, supporting multi-dimensional queries and complex analysis tasks.

[0145] At the same time, this module supports machine learning model training, which can predict future capacity trends based on historical data, provide early warning of possible resource bottlenecks, and optimize system expansion and scheduling strategies.

[0146] (1.3) Dynamic Water Level Adjustment Module: This module leverages a high-performance data analysis engine and advanced machine learning algorithms to regularly extract various capacity indicator data from historical databases, conduct in-depth mining and analysis, and construct a dynamic capacity prediction model. This model accurately predicts the capacity requirements of information systems based on factors such as business models, time periods, and resource allocation, and intelligently adjusts resource allocation to ensure system stability, efficiency, and cost optimization.

[0147] The data sources required for model training include:

[0148] a. Historical monitoring data (CPU, memory, storage, bandwidth, database load, etc.);

[0149] b. Business load data (user request volume, transaction volume, peak concurrent number, etc.);

[0150] c. Environmental configuration data (deployment architecture of servers, containers, and microservices);

[0151] d. External factor data (holiday traffic patterns, seasonal changes, marketing activities, etc.).

[0152] The modeling method uses the LSTM (Long Short Time Memory) deep learning model to predict future capacity trends based on historical data; linear regression is used to analyze the relationship between different business characteristics and resource requirements; K-means is used to classify business traffic patterns to formulate optimization strategies for different scenarios; and the Deep Q Network (DQN) algorithm is used to simulate the effects of different resource adjustment strategies to achieve dynamic optimization.

[0153] (1.4) Capacity Expansion / Scaling Decision Module: This module monitors the capacity health of the information system in real time and, based on the analysis results of the Water Level Calculation Module and the Dynamic Water Level Adjustment Module, accurately determines whether capacity expansion or reduction is necessary. This module utilizes an intelligent decision-making algorithm that combines business load characteristics, system resource utilization, and historical capacity trends to automatically generate resource optimization plans, ensuring the optimal balance between system performance stability and resource utilization.

[0154] The core function of this module is to compare the capacity health calculated by the water level calculation module with the preset capacity expansion threshold in real time, and combine the analysis results of the dynamic water level adjustment module to make the following types of expansion and contraction decisions:

[0155] (1.4.1) Expansion decision logic:

[0156] (1.4.1.1) Trigger conditions for expansion:

[0157] a. The capacity health H exceeds the expansion threshold T (H>T);

[0158] b. The system load continues to grow with no sign of decreasing in the short term;

[0159] c. Continuous high usage of key resources (CPU, memory, database connections, etc.).

[0160] (1.4.1.2) Decision-making process:

[0161] a. Early warning mechanism triggering: Once the system detects that the capacity health exceeds the expansion limit, an alarm is immediately triggered and a detailed data analysis process is initiated.

[0162] b. Detailed analysis of sub-indicators: Conduct in-depth analysis of the specific load conditions of each technical component (computing, storage, network, database, etc.) to determine the necessity and priority of capacity expansion.

[0163] (1.4.1.3) Expansion Strategy Development:

[0164] a. Computational capacity expansion: Increase the number of CPU cores and expand computing instances (such as Kubernetes Pod scaling and virtual machine expansion).

[0165] b. Storage expansion: Expand database instances, increase disk IOPS, and adjust storage layer architecture.

[0166] c. Network expansion: adjust load balancing strategies and increase bandwidth.

[0167] d. Database optimization: optimize SQL queries, add a cache layer, and expand the database connection pool.

[0168] (1.4.1.4) Expansion plan generation: Based on the characteristics of the system architecture, a resource-balanced expansion strategy is recommended to ensure performance improvement while avoiding single-point resource bottlenecks.

[0169] (1.4.1.5) Implementation method:

[0170] a. Auto-Scaling: You can directly call the cloud platform API to complete automatic expansion.

[0171] b. Manual approval of capacity expansion: For critical business systems, capacity expansion proposals are generated and submitted to the administrator for approval before execution.

[0172] (1.4.2) Scaling decision logic:

[0173] (1.4.2.1) Triggering conditions for scaling down:

[0174] a. Capacity health is lower than the shrinking threshold L(H <L);

[0175] b. The system load continues to decrease, and no peak is expected in the short term;

[0176] c. The resource redundancy rate is high, resulting in increased costs.

[0177] (1.4.2.2) Decision-making process:

[0178] a. Resource utilization analysis: Evaluate whether current computing, storage, and network resources are idle for a long time.

[0179] b. Dynamic water level adjustment calculation: Calculate the maximum amount of shrinkable resources to ensure that business peak demand can still be met after shrinkage; combine with the business SLA (Service Level Agreement) to avoid the impact of shrinkage on system availability.

[0180] (1.4.2.3) Scaling strategy formulation:

[0181] a. Computing resource reduction: Reduce the number of cloud instances and lower the specifications of virtual machines.

[0182] b. Storage optimization: archive cold data and adjust storage tier strategies.

[0183] c. Network optimization: Adjust CDN load and reduce unused public IP resources.

[0184] (1.4.2.4) Implementation method:

[0185] a. Automatic scaling: For non-critical components (such as cache layers and temporary computing instances), resources can be automatically released.

[0186] Manual approval for scaling down: For core business systems, administrator confirmation is required before execution.

[0187] The technical solution of this invention utilizes a multi-dimensional assessment approach, encompassing dimensions such as business capacity, service interface capacity, and technical component capacity, to comprehensively evaluate the resource status of information systems. This multi-dimensional assessment comprehensively considers a wider range of system performance indicators, providing more comprehensive capacity management capabilities.

[0188] The technical solution of this invention introduces machine learning technology, which uses historical data to perform dynamic capacity forecasting and optimize resource allocation. By learning from historical monitoring data, business load data, and external factors, machine learning can predict future resource needs and system bottlenecks, thereby making resource adjustments at a more detailed level.

[0189] The technical solution of this invention features a dynamic water level adjustment module. This module calculates the upper capacity limit for different technical component resource configurations based on historical data of various capacity indicators and provides calculations for the system's maximum scalable capacity. This module allows for flexible resource allocation adjustments to adapt to real-time business changes.

[0190] The technical solution of the present invention, in combination with historical data analysis in capacity management, can predict the maximum capacity bearing under different resource configurations and flexibly adjust resources according to changes in business load.

[0191] The technical solution of the present invention calculates capacity health and adjusts capacity in combination with a dynamic water level regulation module. These data are not only presented through a real-time dashboard, but also through trend analysis charts to show the changing trends of water level and capacity health.

[0192] The technical solution of this invention takes a broader, multi-dimensional approach, encompassing business capacity, service interface capacity, and technology component capacity. It also incorporates machine learning for predictive adjustments and real-time monitoring, thus providing a more sophisticated, data-driven, and flexible approach that leverages machine learning for prediction and dynamic adjustments.

[0193] The specific implementation scheme of this embodiment can be found in the relevant descriptions in the above embodiments and will not be repeated here.

[0194] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0195] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" is at least two.

[0196] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0197] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0198] Those skilled in the art will understand that all or part of the steps in the method for implementing the above-mentioned embodiment can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0199] Furthermore, the functional units in the various embodiments of the present invention may be integrated into a single processing module, each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0200] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0201] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0202] The system, method, device, processor and computer-readable storage medium thereof for realizing dynamic information system capacity management based on multi-dimensional evaluation and machine learning of the present invention are adopted, and accurate control and efficient management of information system capacity are realized by virtue of comprehensive capacity evaluation methods and intelligent dynamic water level adjustment capabilities. Through multi-dimensional capacity evaluation, potential capacity bottlenecks and resource waste problems in the system can be discovered in a timely and accurate manner, providing a scientific basis for system optimization and upgrading. The dynamic water level adjustment capability enables the system to flexibly adjust resource allocation according to real-time changes in the business, ensure high performance and stability of the system during business peaks, and maximize resource utilization during business troughs, effectively reducing the operating costs of the system. At the same time, the method, device, platform and storage medium proposed in the present invention are highly versatile and scalable, and can be widely used in various information systems, whether they are traditional enterprise-level application systems or emerging cloud computing and big data platforms.

[0203] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it will be apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Accordingly, the specification and drawings are to be regarded as illustrative rather than restrictive.

Claims

1. A system for dynamic information system capacity management based on multi-dimensional evaluation and machine learning, characterized by: The system comprises: Capacity indicator collection module, used to regularly collect indicator values of the information system's business capacity, service interface capacity, and technical component capacity; A water level calculation module is connected to the capacity index acquisition module and is used to compare and calculate each index value with its upper limit value to obtain the water level value of each index; The dynamic water level adjustment module is connected to the capacity index acquisition module and is used to calculate the change in the capacity carrying limit under different technical component resource configurations based on the historical data of each capacity index, and calculate the maximum scalable capacity of the system; The capacity expansion and contraction decision module is connected to the water level calculation module and the dynamic water level adjustment module, and is used to monitor the capacity health in real time, compare the capacity health calculated by the water level calculation module with the preset capacity expansion threshold in real time, and combine the analysis results of the dynamic water level adjustment module to determine whether expansion or contraction is needed.

2. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 1, characterized in that: The capacity indicator collection module collects high-precision data on business capacity, service interface capacity, and technical component capacity at preset time intervals, performs preliminary cleaning through outlier detection, noise filtering, and data format standardization, and stores the processed data in a dedicated data warehouse.

3. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 1 is characterized in that: The water level calculation module includes: A water level calculation unit, connected to the capacity indicator collection module, is used to compare the collected key performance indicators with pre-configured thresholds to calculate the CPU water level value, storage occupancy water level value, and database query load water level value; Capacity health calculation unit, used to calculate the overall capacity health based on the water level values of each indicator; Visualization and monitoring unit, used to display water level values and capacity health data in a variety of visual ways on the management console; The historical data storage and analysis unit is used to display the calculation results in real time on the system management platform and store them in the historical database.

4. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 3, characterized in that: The water level calculation unit calculates the CPU water level value, storage occupancy water level value and database query load water level value, specifically: The CPU water level value is calculated according to the following formula: The storage utilization watermark is calculated using the following formula: The database query load water level value is calculated according to the following formula:

5. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 3 is characterized in that: The capacity health calculation unit calculates the capacity health by arithmetic average method, weighted average method or exponentially weighted moving average method.

6. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 3, characterized in that: The visualization and monitoring unit displays the current water level status of each key indicator through a real-time dashboard; displays the changing trend of water level value and capacity health over time through a trend analysis chart; and triggers an automatic alarm when the water level value exceeds the set threshold.

7. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 3, characterized in that: The historical data storage and analysis unit stores time series data through a time series database and realizes big data storage.

8. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 1, characterized in that: The dynamic water level regulation module regularly extracts various capacity index data from the historical database to build a dynamic capacity prediction model. The various capacity index data include historical monitoring data, business load data, environmental configuration data and external factor data.

9. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 1, characterized in that: The expansion decision module triggers an expansion decision if it detects that the capacity health H exceeds the expansion threshold T, the system load continues to increase with no downward trend in the short term, and the key resources continue to be highly occupied. Trigger the early warning mechanism, analyze the sub-indicators in detail, formulate expansion strategies for computing expansion, storage expansion, network expansion, or database optimization, generate expansion plans, and perform automatic expansion or manual approval of expansion.

10. The system for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning according to claim 1, characterized in that: If the expansion / scaling decision module detects that the capacity health is lower than the shrinking threshold L, the system load continues to decline and there is no peak expected in the short term, and the resource redundancy rate is high, it triggers a shrinking decision; analyzes resource utilization, calculates dynamic water level adjustment, and formulates a shrinking strategy, that is, reducing computing resources, optimizing storage and network, and performing automatic shrinking or manually approved shrinking.

11. A method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning using the system of claim 1, characterized in that: The method comprises the following steps: (1) Regularly collect the index values of the information system's business capacity, service interface capacity, and technical component capacity; (2) Compare and calculate each indicator value with its upper limit value to obtain the water level value of each indicator; (3) Calculate the overall capacity health of the information system based on the water level values of each indicator; (4) Compare the capacity health with the preset capacity health expansion line to obtain the output result of whether expansion is needed, and determine the technical components that need to be expanded based on the situation of each sub-indicator; (5) Based on the historical data of each capacity indicator, calculate the change in the capacity carrying limit under different technical component resource configurations, and obtain the maximum amount of resources that can be shrunk for different technical components while meeting the capacity threshold requirements; (6) Calculate the overall capacity health of the information system based on the water level values of each indicator.

12. A device for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning, characterized in that: The device comprises: a processor configured to execute computer-executable instructions; A memory storing one or more computer-executable instructions, wherein when the computer-executable instructions are executed by the processor, the steps of the method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning as described in claim 11 are implemented.

13. A processor for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning, characterized in that: The processor is configured to execute computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method for implementing dynamic information system capacity management based on multi-dimensional evaluation and machine learning as described in claim 11 are implemented.

14. A computer-readable storage medium, characterized in that A computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the method for dynamic information system capacity management based on multi-dimensional evaluation and machine learning as described in claim 11.

Images