Cloud computer startup method and device, electronic equipment and storage medium

Abstract

The invention provides a cloud computer starting method, which comprises the following steps of: before obtaining a starting request of a target user, constructing a target user portrait; predicting a pre-started cloud computer corresponding to the target user based on the target user portrait; and when a starting request of a target user is received, determining the pre-started cloud computer as a target cloud computer, and triggering the target cloud computer to respond to the starting request. Before the starting request of the target user is obtained, the target user portrait is constructed, the pre-started cloud computer corresponding to the target user is predicted according to the target user portrait, when the starting request of the target user is received, the pre-started cloud computer is determined as the target cloud computer, and the target cloud computer is triggered to respond to the starting request. The problem of poor user experience caused by incapability of loading personalized application and environment of a user in advance and incapability of quickly connecting to a cloud computer in an existing starting method is solved.

Description

Technical Field

[0001] This invention relates to the field of cloud computing technology, and in particular to a cloud computing power-on method, apparatus, electronic device, and storage medium. Background Technology

[0002] Cloud computing technology provides users with a flexible and elastic desktop experience by centralizing and cloudifying computing resources. However, the startup speed of cloud computers has always been a key bottleneck affecting user experience. Currently, existing startup methods cannot pre-load users' personalized applications and environments, and cannot quickly connect to the cloud computer, resulting in a poor user experience. Summary of the Invention

[0003] This invention provides a cloud computer startup method to address the problems of existing startup methods, such as the inability to pre-load user-specific applications and environments, and the inability to quickly connect to the cloud computer, resulting in a poor user experience. By constructing a target user profile before receiving the startup request, and predicting the corresponding pre-start cloud computer based on the profile, the method identifies the pre-start cloud computer as the target cloud computer upon receiving the startup request. This resolves the issues of existing startup methods, such as the inability to pre-load user-specific applications and environments, and the inability to quickly connect to the cloud computer, leading to a poor user experience.

[0004] In a first aspect, embodiments of the present invention provide a method for booting up a cloud computer, the method comprising the following steps:

[0005] Before obtaining the target user's enable request, construct the target user profile;

[0006] Based on the target user profile, the pre-launch cloud computer corresponding to the target user is predicted;

[0007] Upon receiving an activation request from a target user, the pre-start cloud computer is identified as the target cloud computer, and the target cloud computer is triggered to respond to the activation request.

[0008] Optionally, constructing the target user profile includes:

[0009] Acquire historical behavior data of the target user, including the time period for device startup, usage duration, and set of frequently used applications;

[0010] Based on the historical behavioral data, a target user profile is constructed.

[0011] Optionally, based on the target user profile, predicting the predicted number of pre-launched cloud computers corresponding to the target user includes:

[0012] Based on the target user profile, the cloud computer resource configuration and application set of the target user are predicted;

[0013] The cloud computer resource configuration and the application set are preloaded to obtain the pre-started cloud computer corresponding to the target user.

[0014] Optionally, predicting the cloud computer resource configuration and application set of the target user based on the target user profile includes:

[0015] The behavioral features of the target user are obtained by extracting behavioral features from the target user profile using a preset behavioral prediction model.

[0016] Behavioral prediction is performed on the behavioral characteristics of the target user to predict the cloud computer resource configuration and application set of the target user.

[0017] Optionally, before extracting behavioral features from the target user profile using a preset behavior prediction model to obtain the target user's behavioral features, the method further includes:

[0018] Acquire a training dataset and a pre-trained behavior prediction model. The training dataset includes sample user behavior profile data, cloud computer resource configuration annotation data and application set annotation data corresponding to the sample user behavior profile data. The pre-trained behavior prediction model outputs cloud computer prediction vectors and resource configuration data, application prediction vectors and set data of the application prediction vectors.

[0019] The pre-trained behavior prediction model is trained using the training dataset. Once training is complete, a well-trained behavior prediction model is obtained.

[0020] Optionally, before triggering the target cloud computer to respond to the activation request, the method further includes:

[0021] The network configuration of the target cloud computer is instantly switched to parameters that match the target user terminal in order to establish a network connection.

[0022] Optionally, after determining the pre-boot cloud computer as the target cloud computer, the method further includes:

[0023] When an abnormality is detected in the pre-boot cloud computer, a replacement machine is triggered to start, so that the replacement machine can replace the abnormal pre-boot cloud computer.

[0024] Secondly, embodiments of the present invention also provide a cloud computer power-on device, the cloud computer power-on device comprising:

[0025] The receiving module is used to build a target user profile before receiving the target user's enable request;

[0026] The prediction module is used to predict the pre-launched cloud computer corresponding to the target user based on the target user profile.

[0027] The determination module is used to determine the pre-start cloud computer as the target cloud computer when it receives the start request from the target user, and to trigger the target cloud computer to respond to the start request.

[0028] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the cloud computer boot-up method provided in the embodiments of the present invention.

[0029] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the cloud computer boot-up method provided in the embodiments of the present invention.

[0030] In this embodiment of the invention, a target user profile is constructed before obtaining the target user's activation request; based on the target user profile, the pre-launch cloud computer corresponding to the target user is predicted; upon receiving the target user's activation request, the pre-launch cloud computer is identified as the target cloud computer, and the target cloud computer is triggered to respond to the activation request. This invention solves the problem of existing activation methods failing to pre-load personalized applications and environments, resulting in poor user experience due to the inability to quickly connect to the cloud computer. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart of a cloud computer boot-up method provided in an embodiment of the present invention;

[0033] Figure 2 This is a schematic diagram of the structure of a cloud computer boot-up device provided in an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] like Figure 1 As shown, Figure 1 This is a flowchart of a cloud computer boot-up method provided by an embodiment of the present invention. The cloud computer boot-up method includes the following steps:

[0037] 101. Before obtaining the target user's activation request, construct the target user profile.

[0038] In this embodiment of the invention, the above-mentioned cloud computer boot-up method can be applied to cloud computers. The cloud computer is a virtual desktop service based on cloud computing technology, which migrates the computing, storage and operating environment of traditional personal computers to cloud servers. Users can access a fully functional cloud computer by connecting to the network through lightweight terminal devices (such as laptops, tablets, and mobile phones).

[0039] The target user mentioned above can be understood as the user who initiates the request to start the cloud computer. The start request can be understood as a user's request to start or activate the cloud computer.

[0040] The aforementioned target user profile can be understood as a tagged model built based on target user behavior data and preference analysis. This user profile records data such as user behavior paths and operation frequencies, including access paths and application usage frequency. Furthermore, data analysis can be used to generate user preference tags for applications.

[0041] Furthermore, before receiving a request from a target user to activate the cloud computer, a user profile of the target user can be constructed by collecting and analyzing the target user's historical behavioral data. The aforementioned historical behavioral data includes the cloud computer's power-on time period, usage duration, and frequently used application set.

[0042] 102. Based on the target user profile, predict the pre-launch cloud computer corresponding to the target user.

[0043] In this embodiment of the invention, the target user profile includes the cloud computer's boot time period, usage duration, and a set of commonly used applications.

[0044] Furthermore, by analyzing the user's cloud computer startup time, usage duration, and frequently used application set, the application and resource requirements of the target user can be predicted, thereby generating a pre-boot cloud computer corresponding to the target user.

[0045] The aforementioned pre-boot cloud computer can be understood as a cloud computer that performs computation and storage on a cloud server, and the user device only needs to connect to the network to operate it.

[0046] 103. Upon receiving a startup request from a target user, identify the pre-start cloud computer as the target cloud computer and trigger the target cloud computer to respond to the startup request.

[0047] In this embodiment of the invention, when a startup request is received from a target user, the pre-start cloud computer can be identified as the target cloud computer, and the target cloud computer can be triggered to respond to the startup request.

[0048] Understandably, when a target user sends an activation request, the pre-launched cloud computer can be used as the target cloud computer to respond to the target user's request and perform the corresponding operations.

[0049] In this embodiment of the invention, a target user profile is constructed before obtaining the target user's activation request; based on the target user profile, the pre-launch cloud computer corresponding to the target user is predicted; upon receiving the target user's activation request, the pre-launch cloud computer is identified as the target cloud computer, and the target cloud computer is triggered to respond to the activation request. This invention solves the problem of existing activation methods failing to pre-load personalized applications and environments, resulting in poor user experience due to the inability to quickly connect to the cloud computer.

[0050] It is understood that in the specific implementation of this application, data such as profile data, behavioral data, and configuration data are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required. Furthermore, the collection, use, and processing of related data, as well as the training, deployment, and invocation of algorithm models, must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0051] Optionally, in the step of building a target user profile, historical behavioral data of the target user can be obtained; and the target user profile can be built based on the historical behavioral data.

[0052] In this embodiment of the invention, the aforementioned historical behavior data can be understood as records of a user's behavior in using products or services over a past period. This historical behavior data includes the time period for device startup, usage duration, and a set of frequently used applications. The time period for device startup is used to analyze the user's work habits and daily routine; the usage duration reflects the user's frequency of use and preferences for different applications; and the set of applications is used to analyze the user's interests and needs.

[0053] The aforementioned target user profile can be understood as a tagged model built based on the target user's historical behavioral data and preference analysis. This user profile records data such as user behavior paths and operation frequencies, including access paths and application usage frequency. Furthermore, data analysis can be used to generate user preference tags for applications.

[0054] It should be noted that user profiles of target users can be built by collecting their historical behavioral data. These user profiles are used to accurately describe the characteristics and needs of target users.

[0055] Optionally, in the step of predicting the predicted number of pre-launched cloud computers corresponding to the target user based on the target user profile, the cloud computer resource configuration and application set of the target user can be predicted according to the target user profile; the cloud computer resource configuration and application set can be pre-loaded to obtain the pre-launched cloud computers corresponding to the target user.

[0056] In this embodiment of the invention, the target user profile can be a labeled model constructed based on the target user's historical behavior data and preference analysis. The user profile is used to record data such as user behavior paths and operation frequencies, such as access paths and application usage frequencies.

[0057] The above-mentioned cloud computer resource configuration can be understood as dynamically adjusting computing, storage, and network resources according to user needs to achieve efficient utilization and cost control.

[0058] The aforementioned set of applications can be understood as a whole composed of multiple interconnected applications used to implement specific functions or services.

[0059] The aforementioned preloading can be understood as acquiring and caching data in advance so that it can be quickly provided to users when needed.

[0060] Understandably, behavioral prediction models can be used to analyze and predict target user profiles, thereby predicting the target user's cloud computer configuration and application set. These behavioral prediction models can be built based on machine learning or deep learning, such as SVM (Support Vector Machine) and GANs (Generative Adversarial Networks). SVM is a binary classification model; its core idea is to find a hyperplane that separates the training data and maximizes the margin between this hyperplane and the nearest training sample, thus improving generalization ability. GANs are deep learning models trained through adversarial interaction between two modules, used to generate data highly similar to real data.

[0061] It should be noted that, based on the target user profile, the cloud computer resource configuration and application set of the target user can be predicted, and the cloud computer resource configuration and application set can be pre-loaded to obtain the pre-started cloud computer corresponding to the target user. This can quickly obtain a cloud computer that meets the user's needs and improve the user experience.

[0062] Optionally, in the step of predicting the cloud computer resource configuration and application set of the target user based on the target user profile, behavioral features can be extracted from the target user profile using a preset behavioral prediction model to obtain the target user behavioral features; behavioral prediction can then be performed on the target user behavioral features to predict the target user's cloud computer resource configuration and application set.

[0063] In this embodiment of the invention, the aforementioned preset behavior prediction model can be a behavior prediction model built based on deep learning or machine learning, such as SVM (Support Vector Machine), GANs (Generative Adversarial Networks), etc. The aforementioned preset behavior prediction model can predict the user's cloud computer resource configuration and application set.

[0064] The aforementioned behavioral feature extraction can be understood as the process of analyzing target user profiles and identifying features that represent user behavior patterns. For example, by analyzing a user's access path and application usage frequency on a cloud computer, features such as the user's behavioral habits and preferences can be extracted.

[0065] The aforementioned target user behavior characteristics are identified by analyzing target user profiles and identifying features that represent the target user's behavior patterns, such as behavioral habits and preferences.

[0066] The aforementioned behavior prediction can be understood as the process of predicting the future behavior of target users by analyzing their behavioral characteristics.

[0067] The above-mentioned cloud computer resource configuration can be understood as dynamically adjusting computing, storage, and network resources according to user needs to achieve efficient utilization and cost control.

[0068] The aforementioned set of applications can be understood as a whole composed of multiple interconnected applications used to implement specific functions or services.

[0069] It should be noted that behavioral features can be extracted from the target user profile using a preset behavior prediction model to obtain the target user's behavioral characteristics. Then, behavior prediction can be performed on the target user's behavioral characteristics to predict the target user's cloud computer resource configuration and application set, which can improve user experience and work efficiency.

[0070] Optionally, before the step of extracting behavioral features from the target user profile using a preset behavior prediction model to obtain the target user behavior features, a training dataset and a pre-trained behavior prediction model can be obtained; the pre-trained behavior prediction model can be trained using the training dataset, and after training is completed, a trained behavior prediction model can be obtained.

[0071] In this embodiment of the invention, the training dataset includes sample user behavior profile data, cloud computer resource configuration annotation data corresponding to the sample user behavior profile data, and application set annotation data. The annotation data represents the process of transforming raw data into a form understandable by machine learning models. By adding semantic labels or structured information to the data, the machine can learn and perform tasks such as classification and detection.

[0072] The pre-trained behavior prediction model mentioned above can be a behavior prediction model built based on machine learning or deep learning, such as SVM (Support Vector Machine) or GANs (Generative Adversarial Networks). SVM is a binary classification model. The core idea of ​​SVM is to find a hyperplane that can separate the training data and maximize the margin between this hyperplane and the nearest training sample, thereby improving generalization ability. GANs are deep learning models trained through adversarial interaction between two modules, used to generate data highly similar to real data.

[0073] The pre-trained behavior prediction model outputs cloud computer prediction vectors, resource configuration data related to the computer prediction vectors, application prediction vectors, and set data of application prediction vectors.

[0074] The training described above can be supervised training. Supervised training uses a set of data with known labels to train the model. By optimizing the model parameters, the model can predict the labels of new data or make decisions based on the characteristics of existing data. During training, a minimum loss function can be used to adjust the model parameters to minimize the difference between the model's output label and the input data. The loss function mentioned above measures the difference between the model's prediction and the true result. Its purpose is to improve prediction accuracy by minimizing the loss function value by adjusting the model parameters. The loss function can be the mean squared error loss function, cross-entropy loss function, etc. Parameter tuning refers to the process of optimizing model performance by adjusting parameters such as weights and biases within the model. During training, the model parameters are optimized using labeled data to achieve better prediction or decision-making capabilities.

[0075] Specifically, during training, the model parameters can be adjusted using the backpropagation algorithm with the goal of minimizing the loss function. This adjustment process is iterated until the error loss is less than a preset value or the number of iterations reaches a preset number, at which point the training process ends, resulting in a well-trained behavior prediction model. The backpropagation algorithm described above is a supervised learning algorithm that updates weights by calculating the gradient of the loss function to minimize the error between the predicted output and the true value.

[0076] The trained behavior prediction model described above can predict the cloud computer resource configuration and application set of the target user.

[0077] Optionally, before triggering the target cloud computer to respond to the start request, the network configuration of the target cloud computer can be switched to parameters that match the target user terminal in real time to establish a network connection.

[0078] In this embodiment of the invention, the above network configuration can be understood as the process of managing and setting up network devices, software and services, with the aim of ensuring efficient network operation and secure data transmission.

[0079] The aforementioned instant switching can be understood as achieving seamless connection between the network configuration of the target cloud computer and the parameters of the target user terminal through fast switching technology, ensuring a continuous switching process.

[0080] It should be noted that the network configuration of the target cloud computer can be switched to parameters that match the target user terminal to establish a network connection. This can improve the speed and stability of the network connection and reduce problems caused by network configuration mismatch.

[0081] In one possible implementation, for example, the gateway and DNS server of the target cloud computer can be set according to the IP address and subnet mask of the target user terminal, so that the network configuration of the target cloud computer matches the parameters of the target user terminal, which can improve the speed and stability of network connection and reduce problems caused by network configuration mismatch.

[0082] Optionally, before determining the pre-boot cloud computer as the target cloud computer, a replacement machine can be triggered to start when an abnormality is detected in the pre-boot cloud computer, so that the replacement machine can replace the abnormal pre-boot cloud computer.

[0083] In this embodiment of the invention, the above-mentioned abnormality can be understood as an error or malfunction that occurs in the pre-start cloud computer, causing the pre-start cloud computer to fail to work properly.

[0084] The aforementioned replacement machine can be understood as a backup cloud computer used to temporarily replace the malfunctioning pre-start cloud computer when the pre-start cloud computer fails. The replacement machine can synchronize the status of the pre-start cloud computer.

[0085] Furthermore, when an anomaly is detected in the pre-start cloud computer, the alternative machine is immediately triggered to start. The alternative machine synchronizes the state of the pre-start cloud computer, which can avoid problems caused by the anomaly of the pre-start cloud computer and improve the stability and reliability of the system.

[0086] like Figure 2 As shown, this embodiment of the invention provides a cloud computer boot-up device, which includes:

[0087] Module 201 is used to build a target user profile before obtaining the target user's enable request;

[0088] Prediction module 202 is used to predict the pre-launched cloud computer corresponding to the target user based on the target user profile;

[0089] The determination module 203 is used to determine the pre-start cloud computer as the target cloud computer when it receives the start request from the target user, and to trigger the target cloud computer to respond to the start request.

[0090] Optionally, the construction module 201 is further configured to acquire historical behavior data of the target user, including the power-on time period, usage duration, and set of commonly used applications; and to construct a target user profile based on the historical behavior data.

[0091] Optionally, the prediction module 202 is further configured to predict the cloud computer resource configuration and application set of the target user based on the target user profile; preload the cloud computer resource configuration and application set to obtain the pre-start cloud computer corresponding to the target user.

[0092] Optionally, the prediction module 202 is further configured to extract behavioral features from the target user profile using a preset behavioral prediction model to obtain target user behavioral features; and to predict the target user's cloud computer resource configuration and application set based on the target user behavioral features.

[0093] Optionally, the device is further configured to acquire a training dataset and a pre-trained behavior prediction model. The training dataset includes sample user behavior profile data, cloud computer resource configuration annotation data corresponding to the sample user behavior profile data, and application set annotation data. The pre-trained behavior prediction model outputs cloud computer prediction vectors and resource configuration data corresponding to the computer prediction vectors, application prediction vectors, and set data of the application prediction vectors. The pre-trained behavior prediction model is trained using the training dataset. After training is completed, a trained behavior prediction model is obtained.

[0094] Optionally, the device is also configured to instantly switch the network configuration of the target cloud computer to parameters that match the target user terminal in order to establish a network connection.

[0095] Optionally, the device is further configured to trigger the startup of a replacement machine when an abnormality is detected in the pre-start cloud computer, so that the replacement machine replaces the abnormal pre-start cloud computer.

[0096] like Figure 3 As shown, this embodiment of the invention also provides an electronic device, including a processor, which can execute any of the above-described cloud computer boot-up methods.

[0097] Specifically, it includes a processor 301 and a memory 302, as well as a computer program stored on the memory 302 and capable of running on the processor 301 to execute the cloud computer boot method, wherein:

[0098] The processor 301 executes the calculator program containing the cloud computer boot method stored in the memory 302, and performs the following steps:

[0099] Before obtaining the target user's enable request, construct the target user profile;

[0100] Based on the target user profile, the pre-launch cloud computer corresponding to the target user is predicted;

[0101] Upon receiving an activation request from a target user, the pre-start cloud computer is identified as the target cloud computer, and the target cloud computer is triggered to respond to the activation request.

[0102] Optionally, the process of building the target user profile executed by processor 301 includes:

[0103] Acquire historical behavior data of the target user, including the time period for device startup, usage duration, and set of frequently used applications;

[0104] Based on the historical behavioral data, a target user profile is constructed.

[0105] Optionally, the processor 301 executes a process to predict the predicted number of pre-launched cloud computers corresponding to the target user based on the target user profile, including:

[0106] Based on the target user profile, the cloud computer resource configuration and application set of the target user are predicted;

[0107] The cloud computer resource configuration and the application set are preloaded to obtain the pre-started cloud computer corresponding to the target user.

[0108] Optionally, the step of processor 301 predicting the cloud computer resource configuration and application set of the target user based on the target user profile includes:

[0109] The behavioral features of the target user are obtained by extracting behavioral features from the target user profile using a preset behavioral prediction model.

[0110] Behavioral prediction is performed on the behavioral characteristics of the target user to predict the cloud computer resource configuration and application set of the target user.

[0111] Optionally, before extracting behavioral features from the target user profile using a preset behavior prediction model to obtain target user behavior features, the method executed by the processor 301 further includes:

[0112] Acquire a training dataset and a pre-trained behavior prediction model. The training dataset includes sample user behavior profile data, cloud computer resource configuration annotation data and application set annotation data corresponding to the sample user behavior profile data. The pre-trained behavior prediction model outputs cloud computer prediction vectors and resource configuration data, application prediction vectors and set data of the application prediction vectors.

[0113] The pre-trained behavior prediction model is trained using the training dataset. Once training is complete, a well-trained behavior prediction model is obtained.

[0114] Optionally, before triggering the target cloud computer to respond to the activation request, the method executed by the processor 301 further includes:

[0115] The network configuration of the target cloud computer is instantly switched to parameters that match the target user terminal in order to establish a network connection.

[0116] Optionally, after determining the pre-boot cloud computer as the target cloud computer, the method executed by the processor 301 further includes:

[0117] When an abnormality is detected in the pre-boot cloud computer, a replacement machine is triggered to start, so that the replacement machine can replace the abnormal pre-boot cloud computer.

[0118] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the cloud computer booting method provided in this invention and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0119] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0120] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A cloud computer booting method, characterized by, The method comprises: Before obtaining the starting request of the target user, a target user portrait is constructed; Based on the target user portrait, a pre-start cloud computer corresponding to the target user is predicted; When the starting request of the target user is received, the pre-start cloud computer is determined as the target cloud computer, and the target cloud computer is triggered to respond to the starting request.

2. The method of claim 1, wherein, The target user portrait comprises: Obtain the historical behavior data of the target user, including the boot time period, the use time length, and the set of commonly used application programs; Based on the historical behavior data, a target user portrait is constructed.

3. The method of claim 2, wherein, Based on the target user portrait, a predicted number of pre-start cloud computers corresponding to the target user are predicted, comprising: According to the target user portrait, the cloud computer resource configuration and the application program set of the target user are predicted; The cloud computer resource configuration and the application program set are preloaded to obtain the pre-start cloud computer corresponding to the target user.

4. The method of claim 3, wherein, According to the target user portrait, the cloud computer resource configuration and the application program set of the target user are predicted, comprising: The target user behavior features are extracted from the target user portrait by a preset behavior prediction model to obtain target user behavior features; The target user behavior features are predicted to predict the cloud computer resource configuration and the application program set of the target user.

5. The method of claim 4, wherein, Before the target user behavior features are extracted from the target user portrait by the preset behavior prediction model, the method further comprises: Obtain a training data set and a pre-trained behavior prediction model, the training data set comprising sample user behavior portrait data, resource configuration annotation data and application program set annotation data of a cloud computer corresponding to the sample user behavior portrait data, and the pre-trained behavior prediction model outputting a cloud computer prediction vector and resource configuration data of the computer prediction vector, an application program prediction vector and set data of the application program prediction vector; The pre-trained behavior prediction model is trained by the training data set, and a trained behavior prediction model is obtained after training.

6. The method of claim 1, wherein, Before the target cloud computer responds to the starting request, the method further comprises: The network configuration of the target cloud computer is instantaneously switched to parameters matching the target user terminal to establish a network connection.

7. The method of claim 1, wherein, After the pre-start cloud computer is determined as the target cloud computer, the method further comprises: When the pre-start cloud computer is detected to be abnormal, a substitute machine is triggered to start to replace the abnormal pre-start cloud computer.

8. A cloud computer booting apparatus, comprising: The cloud computer boot device comprises: A construction module for constructing a target user portrait before obtaining the starting request of the target user; A prediction module for predicting a pre-start cloud computer corresponding to the target user based on the target user portrait; A determination module for determining the pre-start cloud computer as the target cloud computer when the starting request of the target user is received, and triggering the target cloud computer to respond to the starting request.

9. An electronic device, comprising: Comprise: The memory, the processor and the computer program stored on the memory and capable of running on the processor, the processor implementing the steps in the cloud computer booting method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium and capable of being executed by the processor to implement the steps in the cloud computer booting method according to any one of claims 1 to 7.

Images