Method and device for adjusting configuration of storage system, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a method and device for adjusting configuration of a storage system, a computer readable storage medium and electronic equipment. The method comprises the steps that the current configuration condition of the storage system is obtained; determining working performance parameters of the storage system when the storage system processes the target business model according to the current configuration condition; according to the current configuration condition of the storage system and the working performance parameters, determining risk indexes of the storage system when the storage system is in the working performance parameters; and when the risk index is greater than a preset threshold value, adjusting the current configuration condition of the storage system, so that the risk index corresponding to the adjusted current configuration condition is less than or equal to the preset threshold value. Through a closed-loop feedback mechanism and configuration adjustment, the system can dynamically and adaptively optimize self-configuration so as to cope with variable service models and potential fault risks. Therefore, in different scenes, the storage system can maintain stable performance and high availability, and the reliability and service continuity of the storage system are improved.

Description

Technical Field

[0001] Embodiments of the present application relate to the computer field, and more specifically, to a method and device for adjusting the configuration of a storage system, a computer-readable storage medium, and an electronic device. Background Art

[0002] With the development of new technologies such as big data, cloud computing, and artificial intelligence, there are increasingly higher requirements for data processing and storage capabilities. In modern data-intensive application environments, storage systems need to provide high-performance, high-stability storage services for enterprises or data centers to meet the business scenarios of enterprises or data centers.

[0003] In the related art, in order to ensure the performance of the storage system, the storage system is usually checked regularly, and the configuration of the storage system is adjusted regularly according to the requirements of data processing.

[0004] However, in the face of the increasingly diverse and complex data processing requirements, the use of related technologies for adjustments will result in insufficient stability and availability of the storage system. Summary of the invention

[0005] The embodiments of the present application provide a method, device, computer-readable storage medium and electronic device for adjusting the configuration of a storage system, which can monitor and dynamically adjust the configuration of the storage system in real time to ensure the stability and availability of the storage system.

[0006] According to an embodiment of the present application, a method for adjusting the configuration of a storage system is provided, the method comprising: obtaining a current configuration status of the storage system, wherein the configuration status comprises a configuration combination and configuration parameters; determining a working performance parameter of the storage system when processing a target business model according to the current configuration status; determining a risk indicator of the storage system when it is at the working performance parameter according to the current configuration status of the storage system and the working performance parameters; and when the risk indicator is greater than a preset threshold, adjusting the current configuration status of the storage system so that the risk indicator corresponding to the adjusted current configuration status is less than or equal to the preset threshold.

[0007] In an exemplary embodiment, based on the current configuration of the storage system and the working performance parameters, the risk indicators of the storage system when it is in the working performance parameters are determined, including: simulating the performance of the storage system in a variety of different fault scenarios and a variety of different load scenarios, respectively, wherein the configuration of the simulated storage system is the current configuration, and the performance parameters of the simulated storage system are the working performance parameters; determining the risk indicators based on the performance of the storage system in a variety of different fault scenarios and a variety of different load scenarios.

[0008] In an exemplary embodiment, the performance of a storage system in a variety of different fault scenarios and a variety of different load scenarios are simulated respectively, including: determining potential fault conditions of the storage system according to the configuration type of the storage system; constructing a variety of fault scenarios according to the potential fault conditions, wherein each fault scenario corresponds to a potential fault condition; inputting the storage system and the various fault scenarios into a preset analysis model, and simulating the performance of the storage system in the various fault scenarios respectively.

[0009] In an exemplary embodiment, the performance of a storage system in a variety of different fault scenarios and a variety of different load scenarios is simulated respectively, and also includes: determining the potential load condition of the storage system according to a variety of business models corresponding to the storage system; constructing a variety of load scenarios according to the potential load condition, wherein each load scenario corresponds to a potential load condition; inputting the storage system and the various load scenarios into a preset analysis model, and simulating the performance of the storage system in the various load scenarios respectively.

[0010] In an exemplary embodiment, risk indicators are determined based on the performance of a storage system in a variety of different fault scenarios and a variety of different load scenarios, including: determining a fault risk indicator based on the performance of the storage system in a variety of different fault scenarios; determining a pressure risk indicator based on the performance of the storage system in a variety of different load scenarios; determining a risk indicator based on the fault risk indicator and the pressure risk indicator.

[0011] In an exemplary embodiment, when a risk indicator is greater than a preset threshold, the current configuration of the storage system is adjusted so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold, including: when the risk indicator is greater than the preset threshold, the current configuration of the storage system and the target business model are input into the analysis model to determine a target configuration combination and target configuration parameters that make the risk indicator less than or equal to the preset threshold when the storage system processes the target business model; and according to the target configuration combination and the target configuration parameters, the current configuration of the storage system is adjusted.

[0012] In an exemplary embodiment, when the risk indicator is greater than a preset threshold, the current configuration status of the storage system and the target business model are input into the analysis model to determine a target configuration combination and target configuration parameters that make the risk indicator of the storage system less than or equal to the preset threshold when processing the target business model, including: inputting the current configuration status of the storage system and the target business model into the analysis model; obtaining multiple adjustment configuration statuses of the storage system fed back by the analysis model, wherein each adjustment configuration status includes an adjusted configuration combination and configuration parameters; determining the risk indicator of the storage system in each adjustment configuration status; and taking an adjustment configuration status whose corresponding risk indicator is less than or equal to the preset threshold as the target adjustment configuration status, wherein the adjustment configuration status includes the target configuration combination and the target configuration parameters.

[0013] In an exemplary embodiment, based on the current configuration, the working performance parameters of the storage system when processing the target business model are determined, including: inputting the current configuration of the storage system and the target business model into a preset prediction model, wherein the prediction model is trained using a training sample set, and the training sample set includes historical performance parameters of the storage system under different configurations and when processing different business models; based on the output results of the prediction model, determining the working performance parameters of the storage system when processing the target business model.

[0014] In an exemplary embodiment, the method also includes: obtaining actual working performance parameters of the storage system when processing the target business model under the current configuration; determining the error situation of the prediction model based on the actual working performance parameters and the predicted working performance parameters determined based on the output results of the prediction model; and adjusting the weight parameters of the prediction model according to the error situation.

[0015] In an exemplary embodiment, obtaining the current configuration status of a storage system includes: determining a target configuration indicator related to a performance parameter of the storage system; obtaining configuration parameters corresponding to the target configuration indicator of the storage system; and obtaining the current configuration status of the storage system based on the configuration parameters corresponding to the target configuration indicator.

[0016] In an exemplary embodiment, when the risk indicator is greater than a preset threshold, the current configuration of the storage system is adjusted so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold. The method also includes: acquiring actual working performance parameters of the storage system in real time; when the actual working performance parameters exceed a preset range, adjusting the current configuration of the storage system so that the actual working performance parameters are within the preset range.

[0017] Another aspect of the present application provides a device for adjusting the configuration of a storage system, the device comprising: a configuration status acquisition module, used to acquire the current configuration status of the storage system; a working performance parameter determination module, used to determine, based on the current configuration status, the working performance parameters of the storage system when processing a target business model; a risk indicator determination module, used to determine, based on the current configuration status and working performance parameters of the storage system, the risk indicator of the storage system when it is at the working performance parameters; an adjustment module, used to adjust the current configuration status of the storage system when the risk indicator is greater than a preset threshold, so that the risk indicator corresponding to the adjusted current configuration status is less than or equal to the preset threshold.

[0018] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0019] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0020] According to another embodiment of the present application, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0021] The above-mentioned method for adjusting the configuration of the storage system of the present application obtains the current configuration of the storage system, predicts its working performance parameters when processing the target business model, and analyzes risk indicators in combination with fault scenarios and load scenarios. The evaluation of risk indicators can help the system identify potential problems and performance bottlenecks, provide a quantitative basis for subsequent configuration adjustments, and ensure that the system can respond in time and maintain stable performance output under complex business and fault scenarios. In this way, the rationality and security of the configuration can be evaluated before the system is operated, avoiding performance bottlenecks or fault risks caused by improper configuration, and improving the stability and efficiency of the storage system. In summary, through the closed-loop feedback mechanism and configuration adjustment, the system can dynamically optimize its own configuration to cope with changing business models and potential fault risks. In this way, even under high load or fault scenarios, the storage system can maintain stable performance and high availability, improving the reliability and business continuity of the storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 is a hardware structure block diagram of a server device according to a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0024] Figure 2 is a flow chart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0025] Figure 3 is a second flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0026] Figure 4 is a third flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0027] Figure 5 is a fourth flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0028] Figure 6 is a fifth flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0029] Figure 7 is a sixth flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0030] Figure 8 is a seventh flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0031] Fig. 9 is a flowchart of an eighth method for adjusting configuration of a storage system according to an embodiment of the present application;

[0032] Fig.10 is a ninth flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0033] Fig.11 is a tenth flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0034] Fig.12 is a flowchart of a method for adjusting configuration of a storage system according to an embodiment of the present application;

[0035] Fig.13 is a structural block diagram of a storage system configuration adjustment device according to an embodiment of the present application;

[0036] Fig.14 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0039] The storage system configuration adjustment method embodiment provided in the embodiment of the present application can be executed in a server device or a similar computing device. Taking running on a server device as an example, Figure 1 1 is a hardware structure block diagram of a server device of a storage system configuration adjustment method according to an embodiment of the present application. Figure 1 As shown, the server device may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microcontroller unit (MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the server device may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above server device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0040] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for adjusting the configuration of the storage system in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the server device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0041] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the server device. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0042] As described in the background technology, the storage systems in the related art may have insufficient performance problems in the face of the increasingly diverse and complex data processing needs. The inventors have found that the management strategies of the storage systems in the related art are often based on preset configuration files. After these configurations are set at the initial stage of system deployment, they are usually not adjusted according to real-time business needs or system status. During business peak periods or when resources are tight, this static configuration may cause performance bottlenecks, affecting data processing speed and response time. In addition, although some storage systems allow administrators to manually adjust configurations to deal with performance issues, this manual optimization method is difficult to respond to rapidly changing business needs in real time, and has high requirements on the skills and experience of administrators. In complex or emergency situations, manual adjustments may not be timely or accurate, resulting in reduced stability of the storage system.

[0043] In view of the above problems, a method for adjusting the configuration of a storage system is provided in this embodiment, which can monitor and dynamically adjust the configuration of the storage system in real time to ensure the stability and availability of the storage system. Figure 2 As shown, the method includes the following steps S200-230:

[0044] Step S200, obtaining the current configuration of the storage system.

[0045] The configuration includes configuration combinations and configuration parameters. Configuration combinations refer to the combination of multiple configuration items in the storage system, which reflects the overall strategy of the system when handling different business scenarios. A configuration combination may include processor core binding strategy, memory allocation, disk array (Redundant Arrays of Independent Disks, RAID) settings, and the enabling status of advanced functions (such as data deduplication and compression). The combination of these configuration items determines how the storage system allocates hardware resources, processes data reading and writing, and manages advanced functions, thereby affecting the overall performance and system stability. For example, the processor core binding strategy determines how the central processing unit (CPU) core in the system is used, such as binding a specific process or task to a specific CPU core to optimize task processing speed and resource utilization. Memory configuration includes memory allocation strategy, cache size, data page replacement algorithm, etc., which affects data reading speed and system response time. RAID configuration, such as RAID O, RAID 5, RAID 6, etc., the selection and setting of different RAID levels affect data redundancy, read and write performance, and storage efficiency. The advanced function status affects whether the storage system enables data deduplication, compression, encryption and other functions, as well as the configuration parameters of these functions, which have a significant impact on the system resource usage and performance. Configuration parameters refer to the specific value or status setting of each configuration item in the configuration combination. For example, in the processor core binding strategy, the configuration parameters may specify which cores are used to process which process. In the memory configuration, the parameters may include the size of the cache, the priority of the replacement strategy, etc. In the RAID configuration, the parameters may involve the stripe size, the data block size, whether to enable the hot spare disk, etc.; in the advanced functions, the parameters may involve the ratio of deduplication and compression, the selection of encryption algorithms, etc. Configuration parameters are the key to optimizing the performance of the storage system. Reasonable parameters can significantly improve the system's operating efficiency and data processing capabilities, while incorrect or inappropriate parameter settings may lead to performance degradation, resource waste or data security issues.

[0046] Specifically, the current configuration of the storage system can be obtained in real time or periodically. The configuration includes but is not limited to processor core binding strategy, memory configuration, disk array (Redundant Arrays of Independent Disks, RAID) stripe width, the status of advanced functions of the volume type, etc., laying the foundation for subsequent performance prediction and analysis.

[0047] For example, the storage system can be monitored continuously, and the current configuration parameters can be accessed through system calls or monitoring APIs. These parameters may be stored in system configuration files or dynamically reflected in the operating status of the storage system. The acquisition module will periodically read this information and format it into the input data format required for subsequent processing (such as performance prediction).

[0048] Step S210: determining the working performance parameters of the storage system when processing the target business model according to the current configuration.

[0049] Specifically, the expected performance indicators of the storage system when processing the target business model under the current configuration, such as the number of read and write operations per second (Input / Output Operations Per Second, IOPS), bandwidth, latency, etc., are analyzed to provide a basis for optimizing the system configuration.

[0050] For example, the performance prediction module can be used to receive current configuration parameters and target business model data (such as concurrency, access mode, etc.), and predict the performance of the storage system under these conditions through a pre-trained recurrent neural network (RNN) model. The RNN model can learn the relationship between configuration parameters and performance indicators based on historical data, thereby providing a prediction of future operating performance for the storage system. Through the performance prediction of RNN, the system can understand in advance the expected performance of processing specific services under specific configurations, which helps to take preventive measures before the system reaches performance bottlenecks or fails, thereby improving the availability and efficiency of the system.

[0051] Step S220: determining the risk index of the storage system when it is at the working performance parameter according to the current configuration of the storage system and the working performance parameter.

[0052] Specifically, after predicting the working performance parameters of the storage system, further analyze the current configuration that has been stored, the reliability under the working performance parameters, and evaluate the risks that the storage system may encounter when processing the business, such as performance degradation, failure probability, etc.

[0053] For example, the deep Q learning algorithm can be used to simulate the performance of the storage system under different failure scenarios (such as controller failure, IO (Input / Output) card failure, etc.) and stress scenarios (such as business peak). By analyzing the predicted performance parameters and evaluating the responsiveness of the simulated scenarios, the risk indicators of the system when processing specific business are calculated. The evaluation of risk indicators can help the system identify potential problems and performance bottlenecks, provide a quantitative basis for subsequent configuration adjustments, and ensure that the system can respond in a timely manner and maintain stable performance output in complex business and failure scenarios.

[0054] Step S230: When the risk indicator is greater than a preset threshold, the current configuration of the storage system is adjusted so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold.

[0055] Specifically, if risk indicators exceed preset security or performance thresholds, the system can automatically adjust configurations to reduce risk and improve performance.

[0056] For example, when it is detected that the risk index is too high, the configuration parameters of the storage system can be automatically adjusted based on the optimization strategy provided by the deep Q network (DQN) model. The system can adjust the processor core binding strategy, optimize memory allocation, adjust the RAID stripe width, etc. In addition, the configuration action hierarchical list ensures that the adjustment process will not have too much impact on the business, and high-level adjustments can be performed through secondary authorization by the administrator when necessary. Through the closed-loop feedback mechanism and configuration adjustment, the system can dynamically optimize its own configuration to cope with changing business models and potential failure risks. In this way, even under high load or failure scenarios, the storage system can maintain stable performance and high availability, improving the reliability and business continuity of the entire system.

[0057] In this embodiment, by obtaining the current configuration of the storage system, predicting its working performance parameters when processing the target business model, and analyzing the risk indicators in combination with the fault scenario and the load scenario, the evaluation of the risk indicators can help the system identify potential problems and performance bottlenecks, provide a quantitative basis for subsequent configuration adjustments, and ensure that the system can respond in time and maintain stable performance output in complex business and fault scenarios. In this way, the rationality and security of the configuration can be evaluated before the system is operated, avoiding performance bottlenecks or fault risks caused by improper configuration, and improving the stability and efficiency of the storage system. In summary, through the closed-loop feedback mechanism and configuration adjustment, the system can dynamically optimize its own configuration to cope with changing business models and potential fault risks. In this way, even in high load or fault scenarios, the storage system can maintain stable performance and high availability, improving the reliability and business continuity of the storage system.

[0058] In one embodiment, when adjusting the current configuration of the storage system, it is necessary to query a preset configuration action classification list. In the configuration action classification list, multiple configuration items are classified according to the degree of impact of the configuration on the business. High-level configuration actions require secondary authorization by the administrator, and low-level configuration actions can be directly run automatically. The list can be adjusted according to actual conditions.

[0059] Specifically, the configuration action classification list classifies the configuration adjustment operations in the system according to the size of the impact on the business, so as to guide the system on how to choose the appropriate configuration adjustment strategy in different situations. Specifically, a configuration action classification list can be embedded in the system, and each configuration in the list lists in detail the configuration action and its assessment level of impact on the business. For example, adjusting the processor core binding strategy may be listed as a high-level action because it directly involves the system's computing power allocation and has a greater impact on the immediate response of the business. Adjusting the RAID stripe width or memory cache strategy may be listed as a medium-level action because they affect the data read and write efficiency and latency, and the impact on the business is relatively small but still significant. Turning on or off data compression and deduplication functions can be listed as low-level actions because they mainly affect data storage efficiency and long-term space utilization, and have less impact on immediate business.

[0060] Specifically, in order to adapt to the needs of different business models and system states, an adjustment interface for the configuration action hierarchical list can be designed to allow system administrators to adjust the configuration action hierarchical list according to actual conditions. Administrators can adjust the classification of configuration actions. For example, during specific business peak periods, the processor core binding policy may be adjusted to a low-level action, allowing the closed-loop feedback correction module to automatically optimize under high load to ensure business continuity. During non-peak periods, this action can be adjusted to a high level, requiring secondary authorization to avoid unnecessary waste of resources and performance fluctuations.

[0061] In this embodiment, by configuring the action classification list, the system can make configuration adjustments more intelligently and cautiously. High-level actions can be designed to require secondary authorization from the administrator, ensuring that there is a decision-making process with manual intervention before making adjustments that may have a significant impact on the business, reducing the risk of business interruption caused by automatic adjustments. Low-level action modules can run automatically, which means that the system can automatically make configuration adjustments that slightly improve performance without affecting business stability, thereby improving the system's adaptability and operation and maintenance efficiency. By adjusting the configuration action classification list, the system can more flexibly respond to performance optimization requirements under different business models and system states. This mechanism not only improves the system's adaptability, but also ensures the controllability of business risks during performance optimization, reduces operation and maintenance costs, and improves business continuity and user experience. It provides system administrators with sufficient control rights, allowing them to accurately guide system optimization strategies while ensuring the normal operation of the business, thereby achieving optimal performance and resource utilization in their respective business scenarios.

[0062] In one embodiment, Figure 3As shown, step S220, according to the current configuration of the storage system and the working performance parameters, determines the risk index of the storage system when it is in the working performance parameters. It includes: steps S300-S310:

[0063] Step S300, simulating the performance of the storage system in a variety of different failure scenarios and a variety of different load scenarios.

[0064] Specifically, when the configuration of the simulated storage system is the current configuration and the performance parameters of the simulated storage system are working performance parameters, the working states of the storage system under different types of failures and high loads are simulated to evaluate the system performance.

[0065] For example, a deep Q learning algorithm can be used to simulate various failure scenarios (such as controller failure, IO card failure, etc.) and load scenarios (such as business peak, batch data processing, etc.). Through the DQN model, the system can explore various state and action combinations to find out the system's performance and possible performance bottlenecks in specific scenarios.

[0066] Step S310: determining a risk index according to the performance of the storage system in a variety of different failure scenarios and a variety of different load scenarios.

[0067] Specifically, based on the performance of the storage system in the simulation, the risks that the storage system may encounter under a specific business model are quantified, such as the extent of performance degradation and the probability of failure.

[0068] For example, the simulation results can be analyzed to use the performance degradation, failure frequency, and other performance index fluctuations of the storage system in different scenarios as risk indicators. The determination of risk indicators may involve statistical analysis and weighted average to comprehensively evaluate the risk level in different scenarios.

[0069] In this embodiment, by simulating the performance of the storage system in a variety of different fault scenarios and a variety of different load scenarios, the performance of the storage system under various conditions can be simulated and evaluated, the accuracy of the prediction and the efficiency of the fault scenario analysis are improved, and a solid foundation is provided for the subsequent determination of risk indicators. By simulating different scenarios, the system can comprehensively evaluate its response capabilities in the face of faults and high loads, which helps to identify potential performance problems and risk points in advance, provides data support for system optimization and fault prevention, and enhances the system's adaptability and stability. The quantification of risk indicators helps system administrators and automatic management systems to intuitively understand the vulnerability of storage systems when facing different challenges, provides a quantitative basis for configuration adjustment and optimization, ensures the high reliability and availability of the system in complex scenarios, reduces the risk of business interruption, and improves overall operation and maintenance efficiency.

[0070] In one embodiment, Figure 4 As shown, step S300 simulates the performance of the storage system in various fault scenarios and various load scenarios. It includes steps S400-S420:

[0071] Step S400: determining a potential failure situation of the storage system according to the configuration type of the storage system.

[0072] Specifically, the types of failures that may occur in the storage system can be identified in advance. These failures may originate from different configuration types, such as processor, memory, storage architecture, etc., and then a list of potential failure conditions may be listed.

[0073] Exemplarily, by analyzing the current configuration types, including but not limited to processor core binding strategy, memory configuration, RAID stripe width, advanced function activation status, etc., combined with fault history data and domain knowledge, a list of multiple potential fault conditions is constructed. For example, for the processor core binding strategy, possible fault conditions include processor overheating, uneven core load, etc.; for memory configuration, possible fault conditions include memory leak, insufficient memory, etc.

[0074] Step S410, constructing multiple fault scenarios according to potential fault conditions.

[0075] Each fault scenario corresponds to a potential fault condition.

[0076] Specifically, based on the identified potential fault conditions, a series of fault scenarios are constructed, each of which simulates the impact of a specific fault condition on system performance.

[0077] Exemplarily, based on a list of potential failure scenarios, a deep Q-learning algorithm is used to construct multiple failure scenarios. This may involve setting specific simulation parameters for each failure scenario, such as the frequency, duration, and impact of the failure, as well as changes in the business model under failure scenarios (such as an increase in concurrency, a change in access patterns, etc.). The constructed failure scenarios need to be as close to the actual situation as possible so that the simulation results can reflect the risks in the real environment.

[0078] Step S420: input the storage system and various fault scenarios into a preset analysis model, and simulate the performance of the storage system in various fault scenarios.

[0079] Specifically, after the storage system and fault scenarios are built, these are input into the preset analysis model, and the performance changes of the storage system under different fault scenarios are evaluated through simulation.

[0080] For example, the analysis model can be based on the DQN reinforcement learning algorithm, which receives the current configuration and working performance parameters of the storage system, as well as the corresponding fault scenario parameters. The model simulates the introduction of fault scenarios and analyzes the changes in system performance indicators under fault conditions, such as IOPS drop, latency increase, etc. The simulation process may involve multiple runs of the storage system, each time introducing different fault parameters to comprehensively evaluate the performance impact under different conditions.

[0081] In this embodiment, by systematically identifying potential failure situations, the risk points of the storage system can be more comprehensively evaluated, providing a clear direction for the subsequent construction of failure scenarios and simulations, and enhancing the system's failure warning and response capabilities. By constructing and simulating specific failure scenarios, the system can predict changes in its performance under different failure conditions, which helps to identify possible performance bottlenecks and failure points in advance, provides a basis for failure prevention and adaptive configuration adjustments, and enhances the stability and reliability of the storage system. By simulating the performance of the storage system under failure scenarios, the system can quantify the degree of impact of different failures on performance, provides data support for the subsequent determination of risk indicators, and helps the system adopt targeted configuration adjustment strategies to ensure that a stable and reliable operating state can be maintained when potential failures occur, thereby improving the adaptability and availability of the storage system in a dynamic environment.

[0082] In one embodiment, Figure 5 As shown, step S310 simulates the performance of the storage system in various fault scenarios and various load scenarios. It includes steps S500-S520:

[0083] Step S500: determining the potential load of the storage system according to multiple business models corresponding to the storage system.

[0084] Specifically, based on the various business models supported by the storage system, the possible load types and intensities are predicted to prepare for the subsequent load scenario construction.

[0085] Exemplarily, the performance data collection module analyzes historical data and business characteristics to determine potential business load conditions based on the business model actually running the storage system, such as database scenarios, virtualization scenarios, cloud platform scenarios, etc. For example, for a database business model, possible potential loads include high-frequency random reads and writes, sudden increases in data volume, etc.; for a virtualization scenario, potential loads may involve multi-tasking concurrency, virtual machine migration, etc.

[0086] Step S510, constructing multiple load scenarios according to potential load conditions.

[0087] Each load scenario corresponds to a potential load condition. Based on the determined potential load conditions, a series of load scenarios are constructed, each of which simulates a load condition under a specific business model for subsequent performance simulation.

[0088] For example, based on the list of potential load conditions, design and build multiple load scenarios. This involves setting specific business model parameters, such as the number of concurrent requests, data block size, read-write ratio, etc., as well as the duration and frequency of the simulation to ensure that the load scenario can truly reflect the potential business pressure. The construction of each scenario should simulate the actual business environment as much as possible to improve the reliability and accuracy of the simulation results.

[0089] Step S520: input the storage system and various load scenarios into a preset analysis model to simulate the performance of the storage system in various load scenarios.

[0090] Specifically, after the load scenario is constructed, the storage system and scenario are input into the preset analysis model, and the performance of the storage system when facing different business loads is evaluated through simulation.

[0091] Exemplarily, the preset analysis models, namely the RNN model embedded in the performance prediction module and the DQN model used in the fault and stress scenario traversal analysis module, are based on the current configuration and working performance parameters of the storage system, as well as the parameters of various load scenarios, to simulate the changes in the system's performance indicators in these scenarios, such as IOPS, bandwidth, latency, etc. The simulation may involve multiple runs of the storage system, each with a different load scenario as input, to comprehensively evaluate the performance changes of the system under different load conditions. Among them, the fault and stress scenario traversal analysis module simulates different fault / stress / functional scenarios (such as controller failure, IO card failure, high potential stress scenarios with increased concurrency, system upgrade / business migration and other functional scenarios, etc.) based on the predicted performance through a deep reinforcement learning algorithm, analyzes the reliability of the system under such high load or fault conditions, and identifies the performance bottleneck factors under the current configuration.

[0092] In this embodiment, by identifying and predicting potential business load conditions, the behavior of the storage system under these loads can be simulated more accurately, which helps to identify possible performance bottlenecks and optimization points in advance, and improves the system's resilience and resource allocation efficiency. By building and simulating load scenarios, the system can estimate changes in its performance under different business pressures, which helps to identify possible performance bottlenecks and optimization requirements, and provide strategic guidance for the stable operation of the storage system during business peaks or other high-load scenarios, thereby enhancing the system's adaptability. By simulating the performance of the storage system under load scenarios, the system can quantify the impact of different business loads on performance, and provide data support for subsequent risk indicator determination and configuration optimization. This method helps to ensure that the storage system can still maintain efficient and stable services when facing business peaks or other high-load situations, reduces performance bottlenecks and avoids the risk of service interruptions, and improves overall system availability and business continuity.

[0093] In one embodiment, Figure 6 As shown, step S320 determines the risk index according to the performance of the storage system in various failure scenarios and various load scenarios. It includes: steps S600-S620:

[0094] Step S600: determining a failure risk index according to the performance of the storage system in a variety of different failure scenarios.

[0095] Specifically, by analyzing the performance degradation of the storage system under simulated failure scenarios, the performance of the system when facing a failure, namely the failure risk index, is quantified.

[0096] For example, the output data of the storage system under different failure scenarios can be recorded, such as IOPS drop, latency increase, throughput reduction, etc. The module evaluates the degree of performance degradation of the system when encountering a specific failure through statistical analysis, such as calculating the average, maximum or percentage of performance degradation, so as to determine the failure risk index.

[0097] Step S610: determining a pressure risk index according to the performance of the storage system in a variety of different load scenarios.

[0098] Specifically, by analyzing the system's performance under simulated high-load scenarios, the risks it faces during business peaks, namely the stress risk index, are quantified.

[0099] For example, the predicted data of the storage system under different load scenarios can be recorded, including performance indicators such as IOPS, bandwidth, and latency. By comparing the performance prediction value with the performance baseline of the system under low load, the extent of performance degradation can be calculated as a stress risk indicator. It may also be necessary to consider the system's resource utilization, such as CPU and memory usage, to ensure the comprehensiveness of the risk indicator.

[0100] Step S620, determining a risk indicator according to the fault risk indicator and the pressure risk indicator.

[0101] Specifically, the failure risk index and the pressure risk index are combined to determine a comprehensive risk index to comprehensively evaluate the risks faced by the storage system.

[0102] For example, based on the quantitative results of the fault risk index and the stress risk index, a comprehensive risk index is determined through certain mathematical methods or algorithms (such as weighted average, risk matrix analysis, etc.). This comprehensive index can be a numerical value or a risk level list, which is used to represent the overall risk level of the system when it runs under a specific business model.

[0103] In this embodiment, by determining the fault risk index, the system can identify in which fault scenarios the performance may be seriously affected, which helps to take preventive measures, such as adjusting the configuration in advance and optimizing resource allocation, thereby reducing the performance loss when the actual fault occurs, and improving the stability and reliability of the system. The determination of the pressure risk index helps the system administrator understand to what extent the system performance may decline during business peaks or other high-load scenarios, so as to adopt corresponding resource optimization and load balancing strategies to ensure that the system can maintain good performance under any load, thereby improving the system's service quality and user experience. By comprehensively considering the fault risk and pressure risk, the system can more comprehensively identify potential performance bottlenecks and optimization requirements. The comprehensive risk index provides more comprehensive guidance for the formulation of configuration adjustments and fault prevention strategies, ensures the stable operation of the system in complex environments, and improves the overall performance and availability of the storage system. In addition, the quantification and comprehensive analysis of different risk indicators can also help system administrators or decision makers prioritize the most risky configurations or business models, optimize resource allocation and maintenance plans.

[0104] In one embodiment, Figure 7 As shown, in step S230, when the risk index is greater than the preset threshold, the current configuration of the storage system is adjusted so that the risk index corresponding to the adjusted current configuration is less than or equal to the preset threshold. It includes: steps S700-S710:

[0105] Step S700, when the risk indicator is greater than a preset threshold, the current configuration of the storage system and the target business model are input into the analysis model to determine a target configuration combination and target configuration parameters that make the risk indicator of the storage system less than or equal to the preset threshold when processing the target business model.

[0106] Specifically, when the assessed risk index exceeds the preset safety threshold, it indicates that the current configuration of the storage system cannot effectively cope with potential failures or high loads, and it is necessary to find a new configuration combination to reduce the risk.

[0107] For example, the current configuration of the storage system (such as processor binding strategy, memory configuration, RAID stripe width, etc.) and the target business model (such as database access mode, virtualization concurrency, etc.) are input into the analysis model (DQN-based reinforcement learning algorithm). The analysis model simulates and traverses different configuration combinations, evaluates the performance and risk of each configuration combination, and searches for configuration combinations and parameters that can reduce the risk index below the preset threshold.

[0108] Step S710: adjusting the current configuration of the storage system according to the target configuration combination and the target configuration parameters.

[0109] Specifically, after determining the target configuration combination and parameters, the system needs to perform configuration adjustment to update the current configuration to the target configuration to reduce risks and improve performance.

[0110] For example, specific configuration adjustment instructions can be generated based on the target configuration combination and parameters, involving the adjustment of processor core binding strategy, optimization of memory allocation, change of RAID configuration, etc. These instructions can be automatically executed (if the configuration action level is low, no administrator authorization is required) or submitted to the system administrator for secondary confirmation (if the configuration action level is high, it may affect the business). Once confirmed, the storage system will update its configuration to meet the requirements of the target configuration combination and parameters.

[0111] For example, in the fault and stress scenario traversal, the deep Q learning (DQN) model is used to simulate and analyze different scenarios. DQN continuously optimizes the system configuration to cope with fault or stress scenarios by learning the optimal strategy under different state and action combinations in reinforcement learning, ensuring that the system performs stably under business requirements. The core formula of the DQN algorithm is as follows:

[0112]

[0113] Among them, Q value = Q(s,a), Q value represents the expected total return of the system after taking action a in a certain state s. The state s here represents the current system configuration and business model, such as processor core binding, memory configuration, background advanced tasks, etc.; action a represents the adjustment of a configuration item or the introduction of a fault under the current configuration, such as the controller failure, IO card failure, high potential pressure scenario of increased concurrency, system upgrade / business migration and other functional scenarios mentioned in this application. k is the offset under hardware differences, which can be adjusted through the parameter module. maxQ(s',a') represents the maximum expected total return of the system after taking action a' in the next state s', that is, after taking the current action, when the system enters the next state s', in the next state s', all possible actions a' are executed, and then the maximum expected total return of the expected total returns obtained by executing all possible actions a' is taken as maxQ(s',a'). Other learning rates α, timely rewards r, and discount factors γ are preset values. Through the iterative update of DQN, the system can select the optimal configuration under different fault or pressure scenarios to ensure the high reliability of the storage system.

[0114] In this embodiment, by determining the target configuration combination and parameters, the system can actively adjust the configuration to avoid performance degradation and stability problems in high-risk situations, ensuring that the storage system can maintain the best performance state under different business models, and improving the system's availability and user experience. The automatic configuration adjustment process improves the system's response speed and adaptability, reduces the need for manual intervention, and reduces operation and maintenance costs. And by adjusting the configuration to the optimal state, the system can effectively respond to potential failures and high loads, avoid performance degradation and business interruption, and ensure the high availability and business continuity of the storage system.

[0115] In one embodiment, Figure 8 As shown, step S700, when the risk index is greater than the preset threshold, the current configuration of the storage system and the target business model are input into the analysis model to determine the target configuration combination and target configuration parameters that make the risk index of the storage system less than or equal to the preset threshold when processing the target business model. Including: Steps S800-S830:

[0116] Step S800: input the current configuration of the storage system and the target business model into the analysis model.

[0117] Specifically, in order to evaluate the performance and risks of a storage system when processing a specific business model, it is necessary to input the system's current configuration (such as processor core binding strategy, memory usage, RAID configuration, etc.) and the target business model (such as the concurrent access mode of the business, data access type, etc.) into the analysis model for further analysis and simulation.

[0118] Step S810, obtaining various adjustment configuration conditions of the storage system fed back by the analysis model.

[0119] Specifically, the analysis model will evaluate and feedback multiple possible configuration adjustment schemes, each of which contains specific configuration combinations and parameters for testing and optimizing system performance. The analysis model will evaluate the system performance and risks under each configuration combination by traversing and simulating different configuration combinations (such as adjusting processor core binding, changing RAID stripe width, optimizing memory allocation, etc.) and their parameters. Based on these configuration adjustments, further screening and decision-making will be carried out.

[0120] Each adjustment configuration situation includes an adjusted configuration combination and configuration parameters.

[0121] Step S820: determining the risk index of the storage system in each configuration adjustment situation.

[0122] Specifically, in order to select the most appropriate configuration adjustment plan, it is necessary to perform a risk assessment on each plan and determine the risk index of the storage system under each configuration adjustment situation.

[0123] Step S830: taking an adjustment configuration situation whose corresponding risk index is less than or equal to a preset threshold as a target adjustment configuration situation.

[0124] The adjustment configuration includes a target configuration combination and a target configuration parameter. From a variety of adjustment configurations, a configuration whose risk index is within a preset threshold is selected as a target adjustment configuration to ensure that the system can run safely and stably after adjustment. The risk indexes of different adjustment configurations can be compared, and a configuration with the lowest risk index that does not exceed the preset threshold is selected as the target adjustment configuration.

[0125] In this embodiment, by quantifying the risk indicators in each configuration adjustment situation, the system administrator or the automation system can scientifically compare and select, avoiding system instability or performance degradation that may be caused by blind adjustment, and ensuring the security and efficient operation of the storage system. By selecting the target configuration adjustment situation, the system can strategically adjust the configuration, avoiding resource waste or performance mismatch caused by over-optimization, and ensuring that the storage system can not only reduce risks but also maintain efficient and stable performance when processing the target business model.

[0126] In one embodiment, Fig. 9 As shown, step S210 determines the working performance parameters of the storage system when processing the target business model according to the current configuration. It includes: steps S900-S910:

[0127] Step S900: input the current configuration of the storage system and the target business model into a preset prediction model.

[0128] The prediction model is trained using a training sample set, which includes historical performance parameters of the storage system under different configurations and when processing different business models.

[0129] Exemplarily, the prediction model can be an RNN model, which uses the RNN model to analyze the input business model and configuration parameters, and predicts the system performance under the current configuration conditions based on the embedded theoretical performance database and RNN. The RNN application is as follows. Under the input conditions based on the business model and configuration parameters, the RNN long short-term memory network (Long Short-Term Memory, LSTM) calculation process is simplified as follows (forget gate, candidate memory state, update memory unit state, input and output gate, etc. are not the focus of this application and are simplified):

[0130] Yt=f(Business Model,Processor Binding,Memory Config,RAID Stripe,Advanced Features,PerformanceDB)

[0131] Among them, Yt is the expected performance output value of the system under the current configuration (such as IOPS, bandwidth, etc.). Business Model is the business model input, including the number of concurrency, access mode, and more specific combinations of large and small blocks / read and write / random order, which directly affect the performance prediction of the storage system. Processor Binding is the binding strategy of the processor core, which affects the computing power allocation of the system. Memory Config is the memory configuration that determines the response speed of the system in high-load or data-intensive tasks. RAID Stripe is the RAID stripe width that affects data distribution and reading speed, and is a key factor affecting storage performance. Advanced Features are advanced features such as deduplication and compression. Whether they are enabled or not has a significant impact on system load and performance. PerformanceDB is the theoretical performance data corresponding to the business model embedded in the module.

[0132] This simplified formula emphasizes the relationship between input conditions and system performance prediction, focusing on specific performance optimization of the storage system. Through these steps, RNN outputs the performance prediction value of the system under current conditions based on the input business model, configuration parameters, and theoretical performance figures, ensuring that the fault scenario traversal has reliable reference data.

[0133] Specifically, the current configuration data of the storage system (including processor core binding, memory configuration, RAID stripe width, etc.) and the target business model (describing the concurrent access mode of the business, data access type, etc.) are fed into the preset prediction model as input.

[0134] Among them, the prediction model is trained based on historical data and can predict the performance parameters of the system based on the input configuration and business model. The prediction model is trained based on historical data and can learn and understand the impact of different configurations and business models on system performance. The training process of the performance prediction model (using RNN) involves collecting and using a large amount of historical data of the storage system, including performance parameters such as IOPS, bandwidth, and latency when processing different business models under different configurations. Through training with these data, the model can learn the relationship between the configuration and business model and the performance parameters, and then predict future performance. The model trained based on historical data can more accurately predict the performance of the system when processing a specific business model, improve the accuracy and reliability of the prediction, and provide strong data support for performance optimization and fault prevention.

[0135] Step S910: Determine the working performance parameters of the storage system when processing the target business model according to the output result of the prediction model.

[0136] Specifically, the output of the prediction model is converted into specific performance parameters as the basis for performance evaluation and optimization. Based on the input current configuration and target business model, the prediction model outputs a series of performance parameter prediction values, such as predicted IOPS, bandwidth, latency, etc. The closed-loop feedback correction module receives these prediction values ​​and determines the expected working performance parameters of the storage system when processing the target business model by evaluating whether the indicators meet the business requirements.

[0137] In this embodiment, the output of the prediction model can quantify the performance of the storage system under a specific business model, which helps system administrators or automatic systems identify possible performance bottlenecks or optimization points in advance, ensure that the system can meet business needs, reduce the risk of performance mismatch or failure, and improve system stability and business continuity.

[0138] In one embodiment, Fig.10 As shown, the above storage system configuration adjustment method further includes: steps S1000-S1020:

[0139] Step S1000, obtaining actual working performance parameters of the storage system when processing a target business model under the current configuration.

[0140] Specifically, in order to verify and evaluate the accuracy of the prediction model, it is necessary to collect actual performance data of the storage system when processing the target business model under the current configuration.

[0141] For example, actual working performance parameters may be collected from the storage system periodically or when triggered by a specific event, including IOPS, throughput, latency, CPU utilization, etc. These data reflect the actual operating status of the system under the target business model.

[0142] Step S1010, determining the error of the prediction model according to the actual work performance parameters and the predicted work performance parameters determined based on the output result of the prediction model.

[0143] Specifically, the error of the prediction model is quantified by comparing the actual working performance parameters with the predicted working performance parameters output by the prediction model.

[0144] For example, the actual performance parameters collected are compared with the predicted performance parameters output by the prediction model to calculate the error value or error percentage. Statistical methods can be used to quantify the error, such as mean square error (MSE), mean absolute error (MAE) or more complex error evaluation algorithms to quantify the difference between the model prediction and the actual performance.

[0145] Step S1020, adjusting the weight parameters of the prediction model according to the error situation.

[0146] Specifically, based on the error situation, the prediction model is adjusted to reduce the prediction error and improve the accuracy and reliability of the prediction.

[0147] Exemplarily, the weight parameters of the prediction model (such as RNN) are adjusted by the back propagation algorithm according to the error situation. The prediction parameters with higher errors are identified, and the weight parameters of the model are updated by the back propagation algorithm to minimize the prediction error. This process may require multiple iterations until the prediction error of the model reaches a preset acceptable range.

[0148] In this embodiment, by collecting actual working performance parameters, a real data basis is provided for the subsequent evaluation of the accuracy of the prediction model, ensuring the scientificity and effectiveness of the model optimization process. The determination of error conditions provides a clear direction for the subsequent optimization of the model, helps to reduce prediction deviations, and improves the application effect and reliability of the model in different business scenarios. By dynamically adjusting the weight parameters of the model, the prediction ability of the model when processing specific business models can be improved in a targeted manner, the generalization ability and adaptability of the model are improved, and the accuracy and reliability of the storage system performance prediction are ensured, which plays a key guiding role in performance optimization and fault prevention.

[0149] In one embodiment, Fig.11 As shown, step S200, obtaining the current configuration of the storage system. It includes: steps S1100-S1120:

[0150] Step S1100, determining a target configuration index related to a performance parameter of a storage system.

[0151] Specifically, key configuration items that affect storage system performance are identified so that these configurations can be focused on and optimized later.

[0152] Exemplarily, the acquisition module collects the performance parameters of the storage system during operation, including processor core binding strategy, memory configuration, the activation status of advanced features of the volume type, RAID stripe width, business model (such as concurrency, access mode, more specific combinations of large and small blocks / read and write / random order), and current performance data (IOPS / bandwidth / latency / CPU utilization, etc.). Based on these parameters, the adaptive feedback management module uses analysis algorithms (such as multivariate regression analysis, decision tree analysis, etc.) to determine which configuration indicators have a significant impact on performance parameters. These configuration indicators may include processor core binding strategy, memory configuration, RAID stripe width, the activation status of advanced features of the volume type, etc.

[0153] Step S1110, obtaining configuration parameters corresponding to the target configuration index of the storage system.

[0154] Specifically, after identifying the key configuration indicators that affect performance, it is necessary to obtain the current configuration parameters of these indicators for subsequent comparison and adjustment.

[0155] Exemplarily, configuration parameters related to the target configuration indicator are read from the configuration database of the storage system. For example, if the processor core binding policy is determined as a key indicator, the module will obtain the current processor core binding policy configuration; if the memory configuration is determined as a key indicator, the current memory usage and allocation strategy will be obtained.

[0156] Step S1120: obtaining the current configuration status of the storage system according to the configuration parameters corresponding to the target configuration indicator.

[0157] Specifically, the acquired target configuration indicators and their configuration parameters are integrated into a comprehensive description of the current configuration of the storage system for subsequent analysis and comparison.

[0158] For example, the configuration parameters corresponding to all target configuration indicators are integrated to form a data packet describing the current configuration of the storage system. This data packet may include processor core binding strategies, memory configuration parameters, RAID stripe width settings, current status of volume type advanced features, etc., providing a comprehensive view to evaluate the configuration of the storage system.

[0159] In this embodiment, by determining the target configuration indicators related to the performance parameters, it is possible to focus on the optimization of key configurations, avoid blind adjustments to non-key configurations, improve optimization efficiency and effectiveness, and reduce resource waste and interference with normal system operation. Obtaining the configuration parameters corresponding to the target configuration indicators provides specific data support for subsequent performance analysis and configuration optimization, helps system administrators or automation systems to more accurately identify configuration problems and optimization directions, and ensures the scientific nature and pertinence of performance optimization. By integrating the current configuration status, the system can more comprehensively evaluate the configuration status of the storage system under a specific business model, provides a comprehensive perspective for performance analysis and optimization, and helps identify potential problems and optimization potential in configuration combinations. This step also provides a basis for subsequent performance prediction and configuration adjustments, ensuring the consistency and systematic nature of the performance optimization process.

[0160] In one embodiment, Fig.12 As shown, in step S230, when the risk indicator is greater than the preset threshold, the current configuration of the storage system is adjusted so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold. The method for adjusting the configuration of the storage system further includes steps S1200-S1210:

[0161] Step S1200, obtaining actual working performance parameters of the storage system in real time.

[0162] Specifically, the operating status of the storage system can be monitored in real time to ensure timely response to performance fluctuations and prevent potential system failures or performance degradation.

[0163] For example, the operation of the storage system is continuously monitored to collect actual performance parameters, including but not limited to key performance indicators such as IOPS (input and output operations per second), throughput, latency, CPU utilization, etc. These data are recorded periodically or when an event (such as performance degradation) is triggered through real-time stream processing technology (such as Apache Kafka) or polling.

[0164] Step S1210: When the actual working performance parameter exceeds the preset range, adjust the current configuration of the storage system to make the actual working performance parameter within the preset range.

[0165] Specifically, when the monitored performance parameters exceed the preset safety or performance range, the system needs to take measures to adjust the configuration to restore performance to normal levels to avoid business interruption or system failure.

[0166] Exemplarily, actual working performance parameters are obtained and compared with preset performance ranges. If it is found that the performance parameters exceed the preset range, the configuration adjustment process will be started. The method in the above embodiment is used to automatically adjust the processor core binding strategy, optimize memory allocation, adjust RAID configuration, or manage the opening and closing of advanced functions. The adjustment process follows the configuration action classification list to ensure that the adjustment action is performed within the range with the least impact on the business, avoiding additional interference with system stability.

[0167] In this embodiment, real-time acquisition of actual working performance parameters can not only ensure that the system administrator or automation system has a comprehensive understanding of the operating status of the storage system, but also promptly detect signs of performance degradation, provide timely data support for subsequent performance adjustment and optimization, improve the system's response speed and adaptability, reduce operation and maintenance costs, and enhance business continuity and user experience. By dynamically adjusting the configuration of the storage system, it is possible to respond to performance fluctuations in a timely manner and restore performance parameters to the preset normal range, ensuring the stable operation and efficient performance of the system under various business models. This method reduces the impact of performance degradation or failures on the business and improves system reliability and user satisfaction.

[0168] Finally, examples are given for application scenarios of the method of the present application.

[0169] Application scenario 1: Dynamic performance optimization during business peak hours:

[0170] During periods of high business pressure, the storage system in the data center needs to cope with significantly increased concurrent loads. The system predicts performance by inputting the business model (including the number of concurrent requests and access frequency) and existing configuration parameters (processor core binding, memory configuration, RAID stripe width, etc.). The RNN model predicts the response delay and IOPS that the current configuration may produce based on these input conditions.

[0171] 1. Performance prediction: The prediction results from the RNN model show that the current configuration may experience latency close to the upper limit during peak load periods.

[0172] 2. Configuration adjustment: After the fault and stress scenario traversal analysis module (based on DQN) simulates the configuration adjustment, it is recommended to adjust the processor core binding relationship and optimize the memory access configuration, and suspend non-critical background deduplication and compression operations to release more resources for the front-end business.

[0173] 3. Feedback Adjustment: The adaptive feedback management module implements adjustments and monitors system performance in real time to ensure that latency remains within an acceptable range and the system operates stably.

[0174] Application scenario 2: Fault warning and fault tolerance improvement:

[0175] In daily operations, a controller of the storage system may fail, and the failure of the controller may threaten the continuity of the overall business. The system first predicts the performance of normal operation through business models and configuration conditions, and inputs the failure scenario into the deep reinforcement learning module to simulate the impact of controller failure.

[0176] 1. Fault scenario simulation: The deep Q learning module simulates the system performance after the controller fails, analyzes the possible performance degradation, and finds that the system's IOPS is at risk of being significantly reduced.

[0177] 2. Optimize configuration: The system provides feedback and automatically reallocates processor cores, switches the load from the failed controller to other normally functioning controllers, and adjusts the RAID configuration to increase read and write speeds.

[0178] 3. Performance verification and feedback: The feedback management module confirms that the new configuration can meet business requirements in the event of a failure, ultimately improving the system's fault tolerance under controller failures.

[0179] Application scenario three: Resource conservation under dynamic business model:

[0180] The storage system of a financial enterprise needs to process batch data analysis tasks at night, while mainly processing user access and transaction requests during the day. This business model requires the system to be able to dynamically adjust performance according to time periods to save resources. The system inputs different requirements during the day and night according to the business model, and performs performance prediction and optimization configuration respectively.

[0181] 1. Business model prediction: During the day when concurrent requests are high, the RNN prediction model outputs results with higher response time and IOPS requirements, while in batch tasks at night the system focuses on data throughput.

[0182] 2. Automatic configuration switching: The adaptive feedback management module adjusts the configuration according to the daytime and nighttime demands predicted by the RNN. For example, it enables data compression and deduplication functions at night to improve data storage efficiency.

[0183] 3. Performance monitoring and feedback: The system verifies the configuration effect when the business switches, and dynamically adjusts the performance prediction model through the feedback management module to achieve a dynamic balance between resource conservation and business needs.

[0184] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0185] In this embodiment, a device for adjusting the configuration of a storage system is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the modules described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0186] Fig.13 : is a structural block diagram of a storage system configuration adjustment device according to an embodiment of the present application, the device comprising:

[0187] The configuration status acquisition module 1301 is used to acquire the current configuration status of the storage system, wherein the configuration status includes configuration combinations and configuration parameters.

[0188] The working performance parameter determination module 1302 is used to determine the working performance parameters of the storage system when processing the target business model according to the current configuration.

[0189] The risk index determination module 1303 is used to determine the risk index of the storage system when it is at the working performance parameter according to the current configuration of the storage system and the working performance parameter.

[0190] The adjustment module 1304 is used to adjust the current configuration of the storage system when the risk index is greater than a preset threshold, so that the risk index corresponding to the adjusted current configuration is less than or equal to the preset threshold.

[0191] The above-mentioned device obtains the current configuration of the storage system, predicts its working performance parameters when processing the target business model, and analyzes risk indicators in combination with fault scenarios and load scenarios. The evaluation of risk indicators can help the system identify potential problems and performance bottlenecks, provide a quantitative basis for subsequent configuration adjustments, and ensure that the system can respond in time and maintain stable performance output in complex business and fault scenarios. In this way, the rationality and security of the configuration can be evaluated before the system is operated, avoiding performance bottlenecks or fault risks caused by improper configuration, and improving the stability and efficiency of the storage system. In summary, through the closed-loop feedback mechanism and configuration adjustment, the system can dynamically optimize its own configuration to cope with changing business models and potential fault risks. In this way, even in high-load or fault scenarios, the storage system can maintain stable performance and high availability, improving the reliability and business continuity of the storage system.

[0192] In an exemplary embodiment, the risk indicator determination module 1303 further includes:

[0193] The test unit is used to simulate the performance of the storage system in a variety of different fault scenarios and a variety of different load scenarios, wherein the configuration of the simulated storage system is the current configuration, and the performance parameters of the simulated storage system are working performance parameters.

[0194] The risk indicator determination unit is used to determine the risk indicator according to the performance of the storage system in a variety of different failure scenarios and a variety of different load scenarios.

[0195] In an exemplary embodiment, the test unit is further used to determine a potential failure condition of the storage system according to a configuration type of the storage system. Multiple failure scenarios are constructed according to the potential failure conditions, wherein each failure scenario corresponds to a potential failure condition. The storage system and the multiple failure scenarios are input into a preset analysis model, and the performance of the storage system in the multiple failure scenarios is simulated respectively.

[0196] In an exemplary embodiment, the test unit is further used to determine the potential load conditions of the storage system according to the multiple business models corresponding to the storage system. Multiple load scenarios are constructed according to the potential load conditions, wherein each load scenario corresponds to a potential load condition. The storage system and the multiple load scenarios are input into a preset analysis model to simulate the performance of the storage system in the multiple load scenarios.

[0197] In an exemplary embodiment, the risk indicator determination unit is further configured to determine a fault risk indicator according to the performance of the storage system in a plurality of different fault scenarios, determine a pressure risk indicator according to the performance of the storage system in a plurality of different load scenarios, and determine a risk indicator according to the fault risk indicator and the pressure risk indicator.

[0198] In an exemplary embodiment, the adjustment module 1304 includes:

[0199] The target configuration determination unit is used to input the current configuration of the storage system and the target business model into the analysis model when the risk indicator is greater than a preset threshold, so as to determine the target configuration combination and target configuration parameters that make the risk indicator of the storage system less than or equal to the preset threshold when processing the target business model.

[0200] The adjustment unit is used to adjust the current configuration of the storage system according to the target configuration combination and the target configuration parameters.

[0201] In an exemplary embodiment, the above-mentioned device further includes: the target configuration determination unit is also used to input the current configuration of the storage system and the target business model into the analysis model. A plurality of adjustment configuration conditions of the storage system fed back by the analysis model are obtained, wherein each adjustment configuration condition includes an adjusted configuration combination and configuration parameters. The risk index of the storage system under each adjustment configuration condition is determined. An adjustment configuration condition whose corresponding risk index is less than or equal to a preset threshold is used as a target adjustment configuration condition, wherein the adjustment configuration condition includes a target configuration combination and a target configuration parameter.

[0202] In an exemplary embodiment, the working performance parameter determination module 1302 is further used to input the current configuration of the storage system and the target business model into a preset prediction model, wherein the prediction model is trained using a training sample set, and the training sample set includes historical performance parameters of the storage system under different configurations and when processing different business models. Based on the output results of the prediction model, the working performance parameters of the storage system when processing the target business model are determined.

[0203] In an exemplary embodiment, the above device further includes:

[0204] The actual parameter acquisition module is used to obtain the actual working performance parameters of the storage system when processing the target business model under the current configuration.

[0205] The error determination module is used to determine the error of the prediction model according to the actual working performance parameters and the predicted working performance parameters determined based on the output results of the prediction model.

[0206] The weight adjustment module is used to adjust the weight parameters of the prediction model according to the error situation.

[0207] In an exemplary embodiment, the configuration status acquisition module 1301 includes:

[0208] The target indicator determination unit is used to determine a target configuration indicator related to a performance parameter of the storage system.

[0209] The configuration parameter acquisition unit is used to acquire the configuration parameters corresponding to the target configuration index of the storage system.

[0210] The current configuration status determination unit is used to obtain the current configuration status of the storage system according to the configuration parameters corresponding to the target configuration indicator.

[0211] In an exemplary embodiment, the above device further includes:

[0212] The real-time parameter acquisition module is used to obtain the actual working performance parameters of the storage system in real time.

[0213] The feedback adjustment module is used to adjust the current configuration of the storage system when the actual working performance parameter exceeds the preset range, so as to make the actual working performance parameter within the preset range.

[0214] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to this: the above modules are all located in the same processor. Alternatively, the above modules are located in different processors in any combination.

[0215] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0216] Optionally, in this embodiment, the computer program may be configured to perform the following steps by means of the computer program:

[0217] S1, obtaining the current configuration of the storage system, wherein the configuration includes configuration combinations and configuration parameters.

[0218] S2, determining the working performance parameters of the storage system when processing the target business model according to the current configuration.

[0219] S3, determining a risk index of the storage system when it is at the working performance parameter according to the current configuration of the storage system and the working performance parameter.

[0220] S4, when the risk index is greater than a preset threshold, adjusting the current configuration of the storage system so that the risk index corresponding to the adjusted current configuration is less than or equal to the preset threshold.

[0221] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0222] The embodiment of the present application also provides an electronic device, such as Fig.14 As shown, the electronic device includes a memory 1402 and a processor 1404. The memory 1402 stores a computer program, and the processor 1404 is configured to execute the steps in any of the above method embodiments through the computer program.

[0223] Optionally, in this embodiment, the processor 1404 may be configured to perform the following steps through a computer program:

[0224] S1, obtaining the current configuration of the storage system, wherein the configuration includes configuration combinations and configuration parameters.

[0225] S2, determining the working performance parameters of the storage system when processing the target business model according to the current configuration.

[0226] S3, determining a risk index of the storage system when it is at the working performance parameter according to the current configuration of the storage system and the working performance parameter.

[0227] S4, when the risk index is greater than a preset threshold, adjusting the current configuration of the storage system so that the risk index corresponding to the adjusted current configuration is less than or equal to the preset threshold.

[0228] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0229] Alternatively, a person skilled in the art may understand that: Fig.14 The structure shown is for illustration only. Fig.14 The structure of the electronic device is not limited. For example, the electronic device may also include Fig.14 More or fewer components (such as network interfaces, etc.) as shown in, or with Fig.14 Different configurations are shown.

[0230] Among them, the memory 1402 can be used to store software programs and modules, such as the program instructions / modules corresponding to the method for adjusting the configuration of the storage system and the device for adjusting the configuration of the storage system in the embodiment of the present application. The processor 1404 executes various functional applications and data processing by running the software programs and modules stored in the memory 1402, that is, realizing the above-mentioned method for adjusting the configuration of the storage system. The memory 1402 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1402 may further include a memory remotely located relative to the processor 1404, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 1402 may be specifically, but not limited to, used to store information such as system configuration files. As an example, such as Fig.14 As shown, the memory 1402 may include but is not limited to the modules in the storage system configuration adjustment device. In addition, it may also include but is not limited to other module units in the storage system configuration adjustment device, which will not be repeated in this example.

[0231] Optionally, the transmission device 1406 is used to receive or send data via a network. Specific examples of the network may include a wired network and a wireless network. In one example, the transmission device 1406 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 1406 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0232] In addition, the electronic device further includes: a display 1408; and a connection bus 1410, which is used to connect various module components in the electronic device.

[0233] In other embodiments, the electronic device may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. The nodes may form a peer-to-peer (P2P) network, and any form of computing device, such as a server, terminal, or other electronic device, may become a node in the blockchain system by joining the peer-to-peer network.

[0234] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0235] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0236] The embodiments of the present application also provide a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any one of the above method embodiments.

[0237] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0238] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for adjusting the configuration of a storage system, characterized in that: The method comprises: Acquire the current configuration of the storage system, wherein the configuration includes configuration combinations and configuration parameters; Determining, according to the current configuration, working performance parameters of the storage system when processing a target business model; Determining, according to the current configuration of the storage system and the operating performance parameter, a risk indicator of the storage system when the storage system is at the operating performance parameter; When the risk indicator is greater than a preset threshold, the current configuration of the storage system is adjusted so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold.

2. The method for adjusting the configuration of a storage system according to claim 1, characterized in that: The step of determining, according to the current configuration of the storage system and the working performance parameter, a risk indicator of the storage system when the storage system is at the working performance parameter, includes: Simulating the performance of the storage system in a plurality of different fault scenarios and a plurality of different load scenarios respectively, wherein the configuration of the storage system being simulated is the current configuration, and the performance parameters of the storage system being simulated are the working performance parameters; The risk index is determined according to the performance of the storage system in a variety of different failure scenarios and a variety of different load scenarios.

3. The method for adjusting the configuration of a storage system according to claim 2, characterized in that: The respectively simulating the performance of the storage system in a plurality of different fault scenarios and a plurality of different load scenarios includes: Determining a potential failure condition of the storage system according to a configuration type of the storage system; Constructing multiple fault scenarios according to the potential fault conditions, wherein each of the fault scenarios corresponds to a potential fault condition; The storage system and the multiple fault scenarios are input into a preset analysis model to simulate the performance of the storage system in the multiple fault scenarios respectively.

4. The method for adjusting the configuration of a storage system according to claim 3, characterized in that: The simulating the performance of the storage system in a plurality of different fault scenarios and a plurality of different load scenarios respectively further includes: Determining a potential load condition of the storage system according to multiple business models corresponding to the storage system; Constructing a plurality of load scenarios according to the potential load conditions, wherein each of the load scenarios corresponds to a potential load condition; The storage system and the multiple load scenarios are input into the preset analysis model to simulate the performance of the storage system in the multiple load scenarios respectively.

5. The method for adjusting the configuration of a storage system according to claim 4, characterized in that: Determining the risk indicator according to the performance of the storage system in a plurality of different failure scenarios and a plurality of different load scenarios includes: Determining a failure risk indicator according to the performance of the storage system in a variety of different failure scenarios; Determining a pressure risk indicator based on the performance of the storage system in a variety of different load scenarios; The risk indicator is determined according to the failure risk indicator and the pressure risk indicator.

6. The method for adjusting the configuration of a storage system according to any one of claims 3 to 5, characterized in that: When the risk indicator is greater than a preset threshold, adjusting the current configuration of the storage system so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold includes: When the risk index is greater than the preset threshold, the current configuration of the storage system and the target business model are input into the analysis model to determine a target configuration combination and target configuration parameters that make the risk index of the storage system less than or equal to the preset threshold when processing the target business model; According to the target configuration combination and the target configuration parameters, the current configuration of the storage system is adjusted.

7. The method for adjusting the configuration of a storage system according to claim 6, characterized in that: When the risk index is greater than the preset threshold, the current configuration of the storage system and the target business model are input into the analysis model to determine a target configuration combination and target configuration parameters that make the risk index of the storage system less than or equal to the preset threshold when processing the target business model, including: Inputting the current configuration of the storage system and the target business model into the analysis model; Acquire multiple adjustment configuration conditions of the storage system fed back by the analysis model, wherein each of the adjustment configuration conditions includes an adjusted configuration combination and configuration parameters; Determining a risk index of the storage system under each of the adjusted configurations; An adjustment configuration situation whose corresponding risk index is less than or equal to the preset threshold is used as the target adjustment configuration situation, wherein the adjustment configuration situation includes the target configuration combination and the target configuration parameters.

8. The method for adjusting the configuration of a storage system according to any one of claims 1 to 5, characterized in that: The determining, according to the current configuration, a working performance parameter of the storage system when processing the target business model includes: Inputting the current configuration of the storage system and the target business model into a preset prediction model, wherein the prediction model is trained using a training sample set, and the training sample set includes historical performance parameters of the storage system under different configurations and when processing different business models; Based on the output result of the prediction model, the working performance parameters of the storage system when processing the target business model are determined.

9. The method for adjusting the configuration of a storage system according to claim 8, characterized in that: The method further comprises: Acquire actual working performance parameters of the storage system when processing the target business model under the current configuration; Determining an error condition of the prediction model according to the actual working performance parameter and a predicted working performance parameter determined based on an output result of the prediction model; The weight parameters of the prediction model are adjusted according to the error situation.

10. The method for adjusting the configuration of a storage system according to claim 8, characterized in that: The obtaining of the current configuration of the storage system includes: Determining target configuration indicators related to performance parameters of the storage system; Obtaining configuration parameters corresponding to the target configuration indicator of the storage system; The current configuration status of the storage system is obtained according to the configuration parameters corresponding to the target configuration indicator.

11. The method for adjusting the configuration of a storage system according to any one of claims 1 to 5, characterized in that: After adjusting the current configuration of the storage system when the risk indicator is greater than the preset threshold so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold, the method further includes: Acquire actual working performance parameters of the storage system in real time; When the actual working performance parameter exceeds a preset range, the current configuration of the storage system is adjusted to make the actual working performance parameter within the preset range.

12. A device for adjusting the configuration of a storage system, characterized in that: The device comprises: A configuration status acquisition module, used to acquire the current configuration status of the storage system, wherein the configuration status includes a configuration combination and configuration parameters; A working performance parameter determination module, used to determine, according to the current configuration, working performance parameters of the storage system when processing a target business model; A risk indicator determination module, used to determine the risk indicator of the storage system when it is at the working performance parameter according to the current configuration of the storage system and the working performance parameter; The adjustment module is used to adjust the current configuration of the storage system when the risk indicator is greater than a preset threshold, so that the risk indicator corresponding to the adjusted current configuration is less than or equal to the preset threshold.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 11 when executed by a processor.

14. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 11 are implemented.

15. A computer program product comprising a computer program, characterized in that The computer program is executed by the processor Carry out the steps of the method described in any one of claims 1 to 11.

Images