Device management method, system, device and storage medium based on cloud computing
By using the cloud computing platform's deep learning algorithm to identify key equipment information and resource constraints, combined with fault risk and safety risk assessment models, efficient, secure, and flexible equipment management is achieved, solving equipment resource constraints and safety issues.
Patent Information
- Application Number
- CN202411363957.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-28
- Publication Date
- 2025-09-30
- Estimated Expiration
- 2044-09-28
AI Technical Summary
The existing equipment management system has deficiencies in efficiency, security, flexibility and adaptability, especially in processing large amounts of data. The management of equipment information leakage and function expansion is not timely, and it is difficult to judge the tight status of equipment resources.
Use deep learning algorithms in the cloud computing platform to identify key equipment information, determine resource shortage status, screen equipment through fault risk assessment and safety risk level assessment models, and adopt adaptive resource scheduling strategies for equipment management.
It achieves efficient, safe, flexible and adaptable equipment management, and ensures the safety and flexibility of equipment by predicting resource shortages and making advance scheduling.
Smart Images
Figure CN119248496B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of cloud computing-based device management, and in particular to a cloud computing-based device management method, system, device, and storage medium. Background Art
[0002] During device use, a large amount of information is generated, which often needs to be processed and analyzed to adjust device management. Processing large amounts of data is time-consuming and labor-intensive, and it is also necessary to ensure data security and privacy, as well as provide high availability and reliability. These demands pose a huge challenge to traditional device management. Existing technologies for device management still lack efficiency, security, and flexibility. For example, when processing large amounts of data, performance bottlenecks often occur, resulting in low processing efficiency; device information involving user privacy and financial security is often leaked; and as business develops, device functions need to be continuously expanded and upgraded, making it impossible to manage newly added functional modules and corresponding data in a timely manner.
[0003] As cloud computing technology continues to mature and gain popularity, its elastic scalability and high availability features have revolutionized various industries. In the field of device management, leveraging the powerful computing capabilities of cloud computing can improve data processing efficiency and meet real-time processing needs. Technical measures such as data encryption, identity authentication, and access control can be used to ensure data security. Computing and storage resource usage can be flexibly adjusted based on business needs, achieving elastic system expansion. However, simply applying cloud computing technology to device management has limited adaptability to different device types and application scenarios. Summary of the Invention
[0004] In order to achieve efficient, secure, flexible and adaptable device management, the present application provides a device management method, system, device and storage medium based on cloud computing.
[0005] In a first aspect, the present application discloses a device management method based on cloud computing, comprising:
[0006] Identify and connect each device to the cloud computing platform, obtain device information of each device, and use the deep learning algorithm in the cloud computing platform to automatically identify the key information of each connected device; the key information includes: device identification, device type, device resource usage, and device operation status;
[0007] Using deep learning algorithms in the cloud computing platform to judge the operation status and resource usage of each device, determine whether each device has reached a resource-constrained state, and if so, determine the device type of the device that has reached a resource-constrained state and the devices of the same type as the device that has reached a resource-constrained state;
[0008] In the cloud computing platform, a preset fault risk assessment model corresponding to the determined device type is queried, and the queried fault risk assessment model is used to perform a fault risk assessment on devices of the same type as the device that has reached a resource-constrained state, and devices of the same type as the device that has reached a resource-constrained state that are assessed to have a fault risk are eliminated;
[0009] In the cloud computing platform, the corresponding preset resource scheduling strategy is queried according to the determined device type, and the resource scheduling of the device that has reached the resource-strained state and the device of the same type as the device that has reached the resource-strained state is completed according to the resources of the device that has completed the elimination process and the preset resource scheduling strategy.
[0010] By adopting the above solution, the intelligent algorithm in the cloud computing platform is used to identify the key information of each connected device to distinguish different types of devices. For different types of devices, adaptive intelligent algorithms are obtained to complete the identification of devices in a resource-constrained state and whether there is a risk of failure. After adaptively screening out devices that may have failures, the preset resource scheduling method is adaptively used to complete resource scheduling, thereby achieving efficient, safe, flexible and adaptable equipment management.
[0011] Preferably, it also includes:
[0012] Use deep learning algorithms in the cloud computing platform to automatically identify the application scenarios of connected devices;
[0013] Determine the device application scenarios corresponding to the devices that have reached a resource-constrained state, match the corresponding preset security risk level requirements based on the determined device application scenarios, and set corresponding preset security risk level requirements for different device application scenarios;
[0014] In the cloud computing platform, a preset security risk level assessment model is queried according to the determined device type, and the security risk level of devices of the same type as the device that has reached a resource-constrained state is assessed using the queried security risk level assessment model, and devices of the same type as the device that has reached a resource-constrained state whose assessment results are lower than the preset security risk level are eliminated.
[0015] By adopting the above solution, considering that different application scenarios of each device have different security risk requirements for the device, it is necessary to schedule devices that meet the security risk level requirements during the resource scheduling process to further ensure the security management of the device.
[0016] Preferably, it also includes:
[0017] In the cloud computing platform, the preset disaster recovery performance level requirements are matched based on the determined device application scenarios, and the corresponding disaster recovery performance level requirements are set for different device application scenarios;
[0018] For devices of the same type as the device that has reached a resource-constrained state, the disaster recovery performance level of each device is determined based on the historical disaster recovery drill test results of each device, and devices of the same type as the device that has reached a resource-constrained state whose disaster recovery performance level is lower than the preset disaster recovery performance level requirement are eliminated.
[0019] By adopting the above solution, taking into account the different application scenarios of each device and the different requirements for the disaster recovery performance level of the device, the device with high processing rate requirements has high disaster recovery performance requirements. Therefore, during the resource scheduling process, it is necessary to schedule devices that meet the disaster recovery performance requirements to further ensure the safe management of the equipment.
[0020] Preferably, in the process of completing resource scheduling of devices that have reached a resource-constrained state and devices of the same type as the devices that have reached a resource-constrained state, when the resource gap between some devices that follow the resource scheduling strategy is less than the preset resource gap, devices with a high device security risk assessment level or a high disaster recovery performance level are preferentially selected for resource scheduling.
[0021] By adopting the above solution, during the resource scheduling process according to the preset scheduling strategy, some devices are added or reduced based on the resource usage of each device to cope with the situation where some devices have the same resource availability. In order to further ensure the security and flexibility of the equipment, devices with high security risk assessment levels or high disaster recovery performance levels are given priority for resource scheduling.
[0022] Preferably, in the process of completing resource scheduling of devices that have reached a resource-constrained state and devices of the same type as the devices that have reached a resource-constrained state, when the resource gap between some devices that follow the resource scheduling strategy is less than the preset resource gap, the historical scheduling times of these devices are counted, and the devices with the lowest historical scheduling times are preferentially selected for resource scheduling.
[0023] By adopting the above solution, during the resource scheduling process according to the preset scheduling strategy, some devices are added or reduced based on the resource usage of each device to deal with the situation where some devices have the same resource availability. In order to further ensure the service life of each device and balance the number of times each device is used, the device with the lowest historical scheduling number is given priority for resource scheduling.
[0024] Preferably, it also includes:
[0025] The deep learning algorithm in the cloud computing platform is used to predict the operation status and resource usage of each device, and determine whether each device will reach a resource-constrained state after the preset time. If there is a device that reaches a resource-constrained state after the preset time, the device that reaches a resource-constrained state after the preset time is added to the device type determined to reach a resource-constrained state.
[0026] By adopting the above solution, deep learning in the cloud computing platform can be used to predict the devices that may reach resource-constrained status in the current adjacent period, and resource scheduling adjustments can be made for these device types in advance to further ensure efficient device management.
[0027] Preferably, it also includes:
[0028] When it is determined that the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is lower than a preset ratio, some devices are reselected for supplementary restoration from the devices whose disaster recovery performance levels are lower than the preset disaster recovery performance level requirements, in descending order of the disaster recovery performance levels of each device, until the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is not lower than the preset ratio, and the supplementary restoration operation is stopped.
[0029] By adopting the above solution, in order to ensure the number of devices that can be resource-scheduled as much as possible, the disaster recovery level requirement is correspondingly reduced, so that the devices eliminated according to the disaster recovery performance level are replenished as device objects for schedulable resources.
[0030] In a second aspect, the present application discloses a cloud computing-based device management system, comprising:
[0031] The device key information identification module is used to identify and connect each device to the cloud computing platform, obtain the device information of each device, and automatically identify the key information of each connected device using the deep learning algorithm in the cloud computing platform; the key information includes: device identification, device type, device resource usage, and device operation status;
[0032] The device resource status judgment module is used to use the deep learning algorithm in the cloud computing platform to judge the operation status and resource usage of each device, determine whether each device has reached a resource-constrained state, and if so, determine the device type of the device that has reached a resource-constrained state and the devices of the same type as the device that has reached a resource-constrained state;
[0033] The device execution scheduling screening module is used to query the corresponding preset fault risk assessment model based on the determined device type in the cloud computing platform, use the queried fault risk assessment model to perform fault risk assessment on devices of the same type as the device that has reached the resource-constrained state, and eliminate devices of the same type as the device that has reached the resource-constrained state that are assessed to have fault risks;
[0034] The device resource scheduling execution module is used to query the corresponding preset resource scheduling strategy according to the determined device type in the cloud computing platform, and complete the resource scheduling of the device that has reached the resource-strained state and the device of the same type as the device that has reached the resource-strained state based on the resources of the device that has completed the elimination process and the preset resource scheduling strategy.
[0035] By adopting the above solution, the cloud computing platform is used to perform adaptive resource scheduling for different types of equipment, achieving efficient, secure, flexible and adaptive equipment management.
[0036] In a third aspect, the present application provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method as described above.
[0037] In a fourth aspect, the present application provides a computer device, which includes a memory, a processor, and a program stored and executable on the memory, and the program implements the steps of the above method when executed by the processor.
[0038] In summary, this application has the following beneficial effects:
[0039] 1. Utilize multiple intelligent algorithms in the cloud computing platform to identify key information of connected devices to distinguish different types of devices. Adaptive intelligent algorithms are then used to identify devices in resource-constrained states and determine whether they are at risk of failure. This eliminates potentially faulty devices and adaptively implements pre-set resource scheduling strategies to achieve efficient, secure, flexible, and adaptable device management.
[0040] 2. Consider the security risks and disaster recovery performance requirements of devices in different application scenarios, and select devices that meet these requirements to complete resource scheduling, further ensuring safe and flexible device management. In the event that some devices have similar resources during resource scheduling, devices with higher security risk levels and higher disaster recovery performance levels will be prioritized for scheduling, achieving safer and more flexible device management.
[0041] 3. Predict the devices that may reach resource shortage in the current adjacent period, adjust resource scheduling for these device types in advance, prevent resource shortages in advance, and further ensure efficient equipment management. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 is a flow chart of a device management method based on cloud computing described in a specific embodiment;
[0043] Figure 2 It is a structural diagram of the cloud computing-based device management system described in a specific embodiment. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.
[0045] like Figure 1 As shown, the embodiment of the present application discloses a device management method based on cloud computing. This embodiment is only illustrated by taking ticketing equipment as an example. The specific steps include:
[0046] S1. Identify and connect each device to the cloud computing platform.
[0047] Specifically, the cloud computing platform's automated discovery mechanism is used to identify and access various device resources, such as self-service ticket machines, window ticketing equipment, online devices (websites, apps, etc.), and special equipment (3D printing ticket machines). Through standardized interface protocols, interconnection between different devices is achieved to ensure smooth access to resources.
[0048] In addition, in order to facilitate subsequent resource calls and device data management, the connected device resources are abstracted, converted into unified resource units, and resource standardization specifications are formulated; a dynamic device resource pool is built using the cloud computing platform, and the abstracted device resources are included; dynamic allocation, release, monitoring and optimization of resources are realized, and security policies and access control mechanisms are set for each connected device to ensure the security and isolation of resources.
[0049] S2. Obtain the device information of each device and use the deep learning algorithm in the cloud computing platform to automatically identify the key information of each connected device.
[0050] Specifically, the key information includes: device identification, device type, device resource usage, device operation status, device application scenario and other device information.
[0051] Among them, the device identifier is a unique device ID that distinguishes the device; the device type can be divided according to the device function, such as: according to the ticket transaction function, it can be divided into ticket checking equipment, ticket vending equipment, and ticket printing equipment; it can also be divided according to the device usage method, such as: the ticket device transaction method is divided into online transaction equipment and offline transaction equipment; or the device type can be divided according to other classification methods (such as a comprehensive device function and usage method); the device resource usage includes: CPU, memory, storage, network and other resources; the device operation status includes: device usage (such as: transaction content, transaction object, transaction duration, etc. in ticket device transaction status) and device usage requirements (ticket transaction requirements), the device's own operating data (such as: power consumption, service life, operation status of device submodules or components), etc.; the device application scenarios include: application scenarios of different usage scales or different usage time periods, such as: concerts of different scales, sports events and smart travel, or scenic spot tours at different time periods, etc.; other device information includes: the device's IP address or physical address, etc.
[0052] Among them, a neural network can be selected using the deep learning algorithm in the cloud computing platform, which is recorded as the first neural network; the input of the first neural network is the data information of the device, and the output is the key information of the device, which is trained and generated through the data information of historical devices with historical annotations of the key information of the device.
[0053] S3. Use the deep learning algorithm in the cloud computing platform to analyze the operation status and resource usage of each device to determine whether each device has reached a resource-constrained state.
[0054] Specifically, a neural network can be selected using the deep learning algorithm in the cloud computing platform, which is denoted as the second neural network; the input of the second neural network is the operating status and resource usage of each device, and the output is whether each device has reached a resource-constrained state. It is trained and generated through the operating status and resource usage of historical devices that are historically marked with whether the devices have reached a resource-constrained state.
[0055] Inputting the real-time or periodically acquired operation status and resource usage of each device into a second neural network to determine whether there is a device that has reached a resource-constrained state, and determining the device type of the device that has reached a resource-constrained state (referred to as a first device) and a device of the same type as the device that has reached a resource-constrained state (referred to as a second device);
[0056] S4. In the cloud computing platform, query the corresponding preset fault risk assessment model according to the determined device type, use the fault risk assessment model to determine whether the device has a fault risk, and eliminate the device with a fault risk.
[0057] Considering the resource constraints of various devices, corresponding resource scheduling is required to adjust the resources of each device, such as increasing the computing resources of online devices or increasing the number of offline devices in use. However, during resource scheduling, attention must be paid to whether the adjusted devices have failure risks. If a faulty offline device or online device is added, the required resources cannot be scheduled. Therefore, devices with failure risks are preemptively eliminated. At the same time, considering that different types of devices have different resource scheduling methods and failure modes, the failure risk assessment of the same type of devices is adaptively completed for a certain type of device.
[0058] Specifically, the corresponding preset fault risk assessment model is queried in the cloud computing platform according to the determined equipment type; wherein, the cloud computing platform stores preset fault risk models corresponding to different types of equipment; each preset fault risk model adopts a neural network model to determine whether a specific type of equipment has a fault risk. The input of the model is the operating status information of the specific type of equipment, and the output of the model is the inference result of whether the equipment has a fault risk or not, which is generated through training on the operating status of the specific type of equipment that has been historically marked as having a fault risk or not.
[0059] In the cloud computing platform, the corresponding preset fault risk assessment model is queried according to the determined device type, such as the fault risk assessment model of online devices and the fault risk assessment model of offline devices; the fault risk assessment of the second device is performed using the queried fault risk assessment model, and the devices in the second device that are assessed to have fault risks are eliminated.
[0060] S5. In the cloud computing platform, the preset resource scheduling strategy is searched according to the determined device type, and the resource scheduling of the device is completed.
[0061] Specifically, considering that different types of equipment have different focuses on resource scheduling, for example, the resource scheduling strategy for online equipment focuses more on responding to sudden traffic peaks and network security threats. The main strategy is to use the elastic expansion capabilities of cloud computing to automatically increase the server resources of equipment that are resource-constrained during traffic peaks. When it is difficult to increase the server resources of equipment that are resource-constrained, consider adding online equipment, such as opening a new APP as a new ticketing transaction channel to alleviate high transaction demand; the intelligent scheduling strategy for offline equipment pays more attention to the physical location and frequency of use of the equipment. If the resources of the current equipment are tight, choose to open a device that is less than the preset distance from the physical location of the current device and has not reached a resource-constrained state to alleviate high transaction demand.
[0062] In the cloud computing platform, query the corresponding preset resource scheduling policy according to the determined device type, such as the preset resource scheduling policy for online devices and the resource scheduling policy for offline devices;
[0063] Based on the resources of the second device that has been eliminated (such as device A with a CPU usage of 45% and device B with a CPU usage of 50%) and the preset resource scheduling policy (such as the resource scheduling policy for offline devices), the resource scheduling of the first and second devices is completed (for example, based on the location information of the second device, it is determined that device A is closest to device C physically, and the resources of device A are in a tight state. Therefore, device A is opened and device B is not operated to complete the resource scheduling).
[0064] By adopting the method described in the above embodiment, efficient, safe, flexible and adaptable device management can be achieved.
[0065] In a specific embodiment, considering that different devices require different information to be acquired in different application scenarios, each device is required to ensure that the information is within a specific security risk level in order to protect the security of the collected information. Therefore, during the resource scheduling process, devices that meet the security risk level should be called as much as possible. The method further includes:
[0066] Determine the device application scenario corresponding to the device that has reached a resource-constrained state; in this embodiment, the device application scenario is an application scenario of different usage scales; such as: (concert) application scenarios of different usage scales, including: a small-scale (concert) application scenario, a medium-scale (concert) application scenario, and a large-scale (concert) application scenario.
[0067] Based on the determined device application scenario matching corresponding preset security risk level requirements, different device application scenarios are set with corresponding preset security risk level requirements, such as: the preset general security risk level requirements corresponding to the small-scale (concert) application scenario setting, the preset lower security risk level requirements corresponding to the medium-scale (concert) application scenario setting, and the low risk level requirements corresponding to the large-scale (concert) application scenario setting.
[0068] In the cloud computing platform, the corresponding preset security risk level assessment model is queried according to the determined device type, such as: the security risk level assessment model for online devices and the security risk level assessment model for offline devices; among them, the cloud computing platform stores preset security risk level assessment models corresponding to different types of devices; each security risk level assessment model adopts a neural network model to determine the security risk level of a specific type of device. The input of the model is the operating status information of the specific type of device, and the output of the model is the inferred result of the security risk level of the device, which is generated through training on the operating status of the specific type of device with a specific security risk level historically marked.
[0069] Perform security risk level assessment on the second device using the retrieved security risk level assessment model. For example, perform security risk level assessment on the second device using the security risk level assessment model for the currently retrieved offline device to obtain assessment results (e.g., the security risk level of device A, which has a CPU usage rate of 45% and is not in use, is a relatively low risk level, and the security risk level of device B, which has a CPU usage rate of 50%, is a low risk level).
[0070] Eliminate devices of the same type as the device that has reached resource-constrained status and whose assessment results are lower than the preset security risk level; for example, the application scenario of device C is a medium-sized concert application scenario with a corresponding preset lower security risk level requirement. The security risk level of device A meets the requirement and can be called without being eliminated.
[0071] In a specific embodiment, considering that different application scenarios of various devices have different requirements for the disaster recovery performance level of the devices, larger application scenarios have higher requirements for the disaster recovery performance level, during the resource scheduling process, it is necessary to schedule devices that meet the disaster recovery performance requirements to further ensure the security management of the devices. The method further includes:
[0072] In the cloud computing platform, the preset disaster recovery performance level requirements are matched based on the determined device application scenarios, and the corresponding disaster recovery performance level requirements are set for different device application scenarios; for example: the preset general disaster recovery performance level is set for the small-scale (concert) application scenario, the preset higher disaster recovery performance level is set for the medium-scale (concert) application scenario, and the preset high disaster recovery performance level is set for the large-scale (concert) application scenario.
[0073] For devices of the same type as the device experiencing resource strain, the disaster recovery performance level of each device is determined based on the historical disaster recovery test results of each second device. Specifically, the scores of each device's disaster recovery test results closest to the current time, stored in the cloud computing platform, are used to determine the different score threshold ranges, and then determine the disaster recovery performance levels that match the different score threshold ranges. For example, if X1 is in the range of [60-75], the matching disaster recovery performance level is the preset general disaster recovery performance level; if X1 is in the range of [75-85], the matching disaster recovery performance level is the preset higher disaster recovery performance level; if X1 is in the range of [80-100], the matching disaster recovery performance level is the preset high disaster recovery performance level.
[0074] Eliminate devices of the same type as the resource-constrained device whose disaster recovery performance level is lower than the preset disaster recovery performance level requirement. For example, if device C is used for a medium-sized concert scenario and has a corresponding preset higher disaster recovery performance level, and device A has a high disaster recovery performance level that meets the requirements, it can be used without being eliminated.
[0075] In addition, in order to ensure the number of devices that can perform resource scheduling as much as possible, the method further includes:
[0076] When it is determined that the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is lower than a preset ratio, the preset ratio can be set manually; indicating that there are fewer devices currently retained and available for resource call, then among the devices whose disaster recovery performance level is lower than the preset disaster recovery performance level requirement, some devices are reselected for supplementary restoration in order of the disaster recovery performance level of each device from high to low, and the supplementary restoration operation is stopped when the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is not lower than the preset ratio.
[0077] In a specific embodiment, considering that during resource scheduling according to a preset scheduling policy, some devices may be added or removed based on the resource usage of each device to cope with the situation where some devices have the same available resource status, in order to further ensure the security and flexibility of the devices, the method further includes:
[0078] During the resource scheduling process for devices experiencing resource strain and devices of the same type as those experiencing resource strain, if the resource gap between some devices complying with the resource scheduling policy is less than a preset resource gap, then, according to the resource scheduling policy, some devices that meet the resource scheduling policy requirements are selected from the second device. If the resource gap (which can be a numerical gap in resource scores obtained by weighted calculation of various resource scores) between any two devices in this portion of devices is less than the preset resource gap (which can be a preset resource gap score), devices with a high security risk assessment level or a high disaster recovery performance level are preferentially selected for resource scheduling. Furthermore, devices can be selected for resource scheduling by comprehensively considering the security risk assessment level and the disaster recovery performance level according to a preset weight ratio.
[0079] In a specific embodiment, during resource scheduling according to a preset scheduling policy, some devices are added or reduced based on the resource usage of each device to cope with the situation where some devices have the same available resource status. In order to further ensure the service life of each device, the method further includes:
[0080] During the resource scheduling process for devices that have reached a resource-constrained state and devices of the same type as those that have reached a resource-constrained state, when the resource gap between some devices that follow the resource scheduling strategy is less than the preset resource gap, the historical scheduling times of these devices are counted. In order to further balance the scheduling times of each device, the device with the lowest historical scheduling times is preferentially selected for resource scheduling.
[0081] Alternatively, the historical service life and scheduling times of the equipment are counted, and the equipment with the shortest historical service life is given priority. When the historical service life is the same, the equipment with the lowest scheduling times is given priority for resource scheduling.
[0082] In a specific embodiment, to further ensure efficient device management, the method further includes:
[0083] Using deep learning algorithms within the cloud computing platform, we predict the operational status and resource usage of each device and determine whether each device will reach resource-constrained status after a preset time. The preset time is manually set. If a device reaches resource-constrained status after the preset time, the device will be added to the device category determined to reach resource-constrained status.
[0084] like Figure 2 As shown, the embodiment of the present application further discloses a device management system based on cloud computing, which is applied to a cloud computing platform and includes:
[0085] The device key information identification module 101 is used to identify and connect each device to the cloud computing platform, obtain device information of each device, and automatically identify the key information of each connected device using the deep learning algorithm in the cloud computing platform; the key information includes: device identification, device type, device resource usage, and device operation status;
[0086] The device resource status determination module 102 is configured to use a deep learning algorithm in the cloud computing platform to determine the operating status and resource usage of each device, determine whether each device has reached a resource-constrained state, and if so, determine the device type of the device reaching the resource-constrained state and any devices of the same type as the device reaching the resource-constrained state;
[0087] The device execution scheduling screening module 103 is configured to query a preset fault risk assessment model corresponding to the determined device type in the cloud computing platform, use the queried fault risk assessment model to perform fault risk assessment on devices of the same type as the device that has reached a resource-constrained state, and eliminate devices of the same type as the device that has reached a resource-constrained state that are assessed to have a fault risk;
[0088] The device resource scheduling execution module 104 is used to query the corresponding preset resource scheduling strategy according to the determined device type in the cloud computing platform, and complete the resource scheduling of the device that has reached the resource-strained state and the device of the same type as the device that has reached the resource-strained state according to the resources of the device that has completed the elimination processing and the preset resource scheduling strategy.
[0089] In a specific embodiment, the device resource status judgment module 102 in the system is also used to use the deep learning algorithm in the cloud computing platform to predict the operation status and resource usage of each device, and judge whether each device reaches a resource-constrained state after a preset time. If there is a device that reaches a resource-constrained state after a preset time, the device that reaches a resource-constrained state after the preset time is added to the device type determined to have reached a resource-constrained state.
[0090] In a specific embodiment, the device execution scheduling screening module 103 in the system is also used to use the deep learning algorithm in the cloud computing platform to automatically identify the application scenarios of each device connected to the device; determine the device application scenario corresponding to the device that has reached a resource-constrained state, match the corresponding preset security risk level requirements based on the determined device application scenario, and set corresponding preset security risk level requirements for different device application scenarios; query the corresponding preset security risk level assessment model in the cloud computing platform according to the determined device type, and use the queried security risk level assessment model to assess the security risk level of devices of the same type as the device that has reached a resource-constrained state, and eliminate devices of the same type as the device that has reached a resource-constrained state whose assessment results are lower than the preset security risk level.
[0091] In a specific embodiment, the device execution scheduling screening module 103 in the system is further configured to match a preset disaster recovery performance level requirement based on a determined device application scenario in the cloud computing platform, and set corresponding disaster recovery performance level requirements for different device application scenarios; for devices of the same type as the device that has reached a resource-constrained state, determine the disaster recovery performance level of each device based on the historical disaster recovery drill test results of each device, and eliminate devices of the same type as the device that has reached a resource-constrained state whose disaster recovery performance level is lower than the preset disaster recovery performance level requirement;
[0092] It is also used to determine that when the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is lower than a preset ratio, some devices are reselected from the eliminated devices whose disaster recovery performance levels are lower than the preset disaster recovery performance level requirements in order of the disaster recovery performance levels of each device from high to low for supplementary restoration, and the supplementary restoration operation is stopped when the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is not lower than the preset ratio.
[0093] The embodiment of the present application also discloses a computer-readable storage medium.
[0094] Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and execute the above-mentioned cloud computing-based device management method. The computer-readable storage medium includes, for example: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.
[0095] The embodiment of the present application also discloses a computer device.
[0096] Specifically, the computer device includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute the above-mentioned cloud computing-based device management method.
[0097] The above are all preferred embodiments of the present application and are not intended to limit the scope of protection of this application. Unless otherwise stated, any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features. In other words, unless otherwise stated, each feature is merely an example of a series of equivalent or similar features.
Claims
1. A device management method based on cloud computing, characterized in that: include: Identify and connect each device to the cloud computing platform, obtain device information of each device, and use the deep learning algorithm in the cloud computing platform to automatically identify the key information of each device connected to the device; The key information includes: device identification, device type, device resource usage, and device operation status; Using deep learning algorithms in the cloud computing platform to judge the operation status and resource usage of each device, determine whether each device has reached a resource-constrained state, and if so, determine the device type of the device that has reached a resource-constrained state and the devices of the same type as the device that has reached a resource-constrained state; In the cloud computing platform, a preset fault risk assessment model corresponding to the determined device type is queried, and the queried fault risk assessment model is used to perform a fault risk assessment on devices of the same type as the device that has reached a resource-constrained state, and devices of the same type as the device that has reached a resource-constrained state that are assessed to have a fault risk are eliminated; In the cloud computing platform, a preset resource scheduling policy is queried according to the determined device type, and resource scheduling of the device that has reached the resource-strained state and ... resource scheduling policy is completed according to the resources of the device that has been eliminated and is of the same type as the device that has reached the resource-strained state; and further comprising: Use deep learning algorithms in the cloud computing platform to automatically identify the application scenarios of connected devices; Determine the device application scenarios corresponding to the devices that have reached a resource-constrained state, match the corresponding preset security risk level requirements based on the determined device application scenarios, and set corresponding preset security risk level requirements for different device application scenarios; In the cloud computing platform, a preset security risk level assessment model is queried according to the determined device type, and the security risk level of devices of the same type as the device that has reached a resource-constrained state is assessed using the queried security risk level assessment model, and devices of the same type as the device that has reached a resource-constrained state whose assessment results are lower than the preset security risk level are eliminated.
2. The device management method based on cloud computing according to claim 1, characterized in that: Also includes: In the cloud computing platform, the preset disaster recovery performance level requirements are matched based on the determined device application scenarios, and the corresponding disaster recovery performance level requirements are set for different device application scenarios; For devices of the same type as the device that has reached a resource-constrained state, the disaster recovery performance level of each device is determined based on the historical disaster recovery drill test results of each device, and devices of the same type as the device that has reached a resource-constrained state whose disaster recovery performance level is lower than the preset disaster recovery performance level requirement are eliminated.
3. The device management method based on cloud computing according to claim 2, characterized in that: During the resource scheduling process for devices that have reached a resource-constrained state and devices of the same type as those that have reached a resource-constrained state, when the resource gap between some devices that follow the resource scheduling strategy is less than the preset resource gap, devices with a high device security risk assessment level or a high disaster recovery performance level are given priority for resource scheduling.
4. The device management method based on cloud computing according to claim 2, characterized in that: During the resource scheduling process for devices that have reached a resource-constrained state and devices of the same type as those that have reached a resource-constrained state, if the resource gap between some devices that follow the resource scheduling strategy is less than the preset resource gap, the historical scheduling times of these devices are counted, and the devices with the lowest historical scheduling times are prioritized for resource scheduling.
5. The device management method based on cloud computing according to claim 1, characterized in that: Also includes: The deep learning algorithm in the cloud computing platform is used to predict the operation status and resource usage of each device, and determine whether each device will reach a resource-constrained state after the preset time. If there is a device that reaches a resource-constrained state after the preset time, the device that reaches a resource-constrained state after the preset time is added to the device type determined to reach a resource-constrained state.
6. The device management method based on cloud computing according to claim 1, characterized in that: Also includes: When it is determined that the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is lower than a preset ratio, some devices are reselected for supplementary restoration from the devices whose disaster recovery performance levels are lower than the preset disaster recovery performance level requirements, in descending order of the disaster recovery performance levels of each device, until the ratio of devices of the same type as the devices that have reached a resource-constrained state before and after the elimination process is completed is not lower than the preset ratio, and the supplementary restoration operation is stopped.
7. A cloud computing-based equipment management system, characterized in that: include: The device key information identification module is used to identify and connect each device to the cloud computing platform, obtain the device information of each device, and use the deep learning algorithm in the cloud computing platform to automatically identify the key information of each connected device; The key information includes: device identification, device type, device resource usage, and device operation status; The device resource status judgment module is used to use the deep learning algorithm in the cloud computing platform to judge the operation status and resource usage of each device, determine whether each device has reached a resource-constrained state, and if so, determine the device type of the device that has reached a resource-constrained state and the devices of the same type as the device that has reached a resource-constrained state; The device execution scheduling screening module is used to query the corresponding preset fault risk assessment model according to the determined device type in the cloud computing platform, and use the queried fault risk assessment model to perform fault risk assessment on devices of the same type as the device that has reached the resource-constrained state, and eliminate devices of the same type as the device that has reached the resource-constrained state whose assessment results show that there is a fault risk; it is also used to use the deep learning algorithm in the cloud computing platform to automatically identify the application scenarios of each device connected to the device; determine the device application scenario corresponding to the device that has reached the resource-constrained state, match the corresponding preset security risk level requirements based on the determined device application scenario, and set corresponding preset security risk level requirements for different device application scenarios; query the corresponding preset security risk level assessment model according to the determined device type in the cloud computing platform, and use the queried security risk level assessment model to perform security risk level assessment on devices of the same type as the device that has reached the resource-constrained state, and eliminate devices of the same type as the device that has reached the resource-constrained state whose assessment results are lower than the preset security risk level; The device resource scheduling execution module is used in the cloud computing platform to query the corresponding preset resource scheduling strategy according to the determined device type, and complete the resource scheduling of the device that has reached the resource-strained state and the device of the same type as the device that has reached the resource-strained state according to the resources of the devices that have completed the elimination processing and are of the same type as the device that has reached the resource-strained state and the preset resource scheduling strategy.
8. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 6.
9. A computer device, characterized in that: The computer device includes a memory, a processor, and a program stored and executable on the memory, and when the program is executed by the processor, the steps of the method according to any one of claims 1 to 6 are implemented.