CN111598390B - Server high availability evaluation methods, devices, equipment and readable storage media - Google Patents

Abstract

This application relates to a server high availability evaluation method, device, computer-readable storage medium and computer equipment. The method includes: obtaining a high-availability service catalog; selecting a target feature vector from the high-availability service catalog as an evaluation index; according to the The evaluation index allocates corresponding resources in the server to the target user; obtains the historical usage value of the resource in the server used by the target user within the historical period; and predicts the use by the target user within the preset period based on the historical usage value. Predicted values of resources in the server; obtaining actual usage values of resources in the server by the target user within the preset period; calculating user satisfaction according to the actual usage values and the predicted values; according to the User satisfaction determines the availability of the server. The solution provided by this application can evaluate the high availability of servers.

Description

Server high availability evaluation methods, devices, equipment and readable storage media

This application claims priority to the Chinese patent application submitted to the China Patent Office on October 16, 2019, with the application number 2019109841043 and the invention title "A method of high availability assessment using SLA", the entire content of which is incorporated by reference in in this application.

Technical field

This application relates to the field of computer technology, and in particular to a server high availability evaluation method, device, equipment and readable storage medium.

Background technique

With the development of computer technology, the computerization requirements of daily operations of commercial and social institutions have reached an unprecedented level. In order to solve the serious value loss caused by server shutdown, high-availability server clusters emerged. High-availability server clusters mainly provide users with as continuous services as possible. Traditional high-availability evaluation is mainly based on the functions and performance of the high-availability service cluster itself. The evaluation results are often unable to effectively meet users' requirements for high-availability service quality.

Contents of the invention

Based on this, it is necessary to provide a server high availability evaluation method, device, equipment and readable storage medium to address the technical problem that traditional high availability evaluation cannot meet users' requirements for high availability service quality.

A server high availability assessment method, including:

Obtain the high availability service catalog;

Select target feature vectors from the high-availability service catalog as evaluation indicators;

Allocate corresponding resources in the server to target users according to the evaluation indicators;

Obtain the historical usage value of the resources in the server used by the target user within the historical period;

According to the historical usage value, predict the predicted value of the resource in the server used by the target user within a preset period;

Obtain the actual usage value of the resources in the server used by the target user within the preset period;

Calculate user satisfaction based on the actual usage value and the predicted value;

The availability of the server is determined based on the user satisfaction level.

In one embodiment, selecting a target feature vector from the high-availability service catalog as an evaluation index includes:

Based on the decision tree algorithm, the target feature vector is selected from the high-availability service catalog as an evaluation index through a machine learning algorithm.

In one embodiment, allocating corresponding resources in the server to target users based on the evaluation indicators includes:

Use the consistency analysis of α algorithm to analyze user data to obtain the matching degree of user needs and the actual process and the adaptability of user needs;

According to the evaluation index, the matching degree and the fitness degree, corresponding resources in the server are allocated to the target user.

In one embodiment, predicting the predicted value of the target user's use of resources in the server within a preset period based on the historical usage value includes:

Get the default time period;

Enter the preset period into a prediction model; the prediction model is a time series model obtained based on the historical usage value;

The predicted value of the target user's use of resources in the server within the preset period is predicted through the prediction model.

In one embodiment, the method further includes:

Determine whether the actual usage value of the resources in the server used by the target user within the preset period meets the user needs of the target user;

If not, the corresponding resources in the server are re-allocated to the target user.

In one embodiment, calculating user satisfaction based on the actual usage value and the predicted value includes:

Construct an evaluation model based on the actual usage value and the predicted value;

Conduct usability evaluation through the evaluation model to obtain usability evaluation results;

Calculate user satisfaction based on the usability evaluation results.

In one embodiment, calculating user satisfaction based on the usability evaluation results includes:

The predicted value and the usability evaluation result are input into the user satisfaction formula to obtain user satisfaction; the user satisfaction formula is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

A server high availability evaluation device, the device includes:

The high-availability service catalog acquisition module is used to obtain the high-availability service catalog;

An evaluation index determination module is used to select a target feature vector from the high-availability service catalog as an evaluation index;

A server resource allocation module, configured to allocate corresponding resources in the server to target users based on the evaluation indicators;

The resource usage value acquisition module is used to obtain the historical usage value of the resources in the server used by the target user within the historical period;

A prediction module, configured to predict the predicted value of the target user's use of resources in the server within a preset period based on the historical usage value;

The resource usage value acquisition module is also used to obtain the actual usage value of the resources in the server used by the target user within the preset period;

A user satisfaction calculation module, configured to calculate user satisfaction based on the actual usage value and the predicted value;

Availability determining module, configured to determine the availability of the server according to the user satisfaction level.

In one embodiment, the evaluation index determination module is also used to:

Based on the decision tree algorithm, the target feature vector is selected from the high-availability service catalog as an evaluation index through a machine learning algorithm.

In one embodiment, the server resource allocation module is also used to:

Use the consistency analysis of α algorithm to analyze user data to obtain the matching degree of user needs and the actual process and the adaptability of user needs;

According to the evaluation index, the matching degree and the fitness degree, corresponding resources in the server are allocated to the target user.

In one embodiment, the prediction module is also used to:

Get the default time period;

Enter the preset period into a prediction model; the prediction model is a time series model obtained based on the historical usage value;

The predicted value of the target user's use of resources in the server within the preset period is predicted through the prediction model.

In one embodiment, the device further includes:

A user demand judgment module, used to judge whether the actual usage value of the resources in the server used by the target user within a preset period meets the user demand of the target user;

The server resource allocation module is also configured to re-allocate a server to the target user if the actual usage value of the resources in the server used by the target user within the preset period does not meet the user needs of the target user. corresponding resources.

In one embodiment, the user satisfaction calculation module is also used to:

Construct an evaluation model based on the actual usage value and the predicted value;

Conduct usability evaluation through the evaluation model to obtain usability evaluation results;

Calculate user satisfaction based on the usability evaluation results.

In one embodiment, the user satisfaction calculation module is also used to:

The predicted value and the usability evaluation result are input into the user satisfaction formula to obtain user satisfaction; the user satisfaction formula is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it causes the processor to perform the steps of the above method.

A computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the steps of the above method.

The above-mentioned server high availability evaluation method, device, equipment and readable storage medium, after obtaining the high availability service catalog, selects the target feature vector from the high availability service catalog as the evaluation index, and allocates corresponding resources in the server to the target user according to the evaluation index , obtain the historical usage value of the resources in the server used by the target user within the historical period, predict the predicted value of the resources in the server used by the target user in the preset period based on the historical usage value, and obtain the actual usage of resources in the server by the target user in the preset period. Usage values, thereby calculating user satisfaction based on actual usage values and predicted values, and determining server availability based on user satisfaction. In this way, the high availability of the server can be evaluated on the premise of meeting the user's demand for high availability service quality.

Description of the drawings

Figure 1 is an application environment diagram of the server high availability evaluation method in one embodiment;

Figure 2 is a model diagram of a server high availability evaluation method in one embodiment;

Figure 3 is a schematic flowchart of a server high availability evaluation method in one embodiment;

Figure 4 is a schematic flowchart of a server high availability evaluation method in another embodiment;

Figure 5 is a schematic diagram of a simulated system using a VMware workstation in one embodiment;

Figure 6 is a structural block diagram of a server high availability evaluation device in one embodiment;

Figure 7 is a structural block diagram of a server high availability evaluation device in another embodiment;

Figure 8 is a structural block diagram of a computer device in one embodiment.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

The server high availability evaluation method provided by this application can be applied to the application environment as shown in Figure 1. Among them, the terminal 102 carries out network communication with the high availability server 106 through the proxy server 104. Taking the above server high availability evaluation method executed on the proxy server 104 as an example, and illustrating it in conjunction with the server high availability evaluation model diagram shown in Figure 2, the proxy server 104 obtains a high availability service catalog; selects a target feature vector from the high availability service catalog As an evaluation index; allocate corresponding resources in the server to the target user based on the evaluation index; obtain the historical usage value of the target user's use of the server's resources within the historical period; predict the target user's use of the server's resources within the preset period based on the historical usage value predicted value; obtain the actual usage value of resources in the server used by target users within a preset period; calculate user satisfaction based on the actual usage value and predicted value.

Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets and portable wearable devices. The proxy server 104 can be a service level proxy that stores service level agreements. The high availability server 106 is a terminal. The high-availability server proxy server 104 and the high-availability server 106 that provide business services can be implemented as an independent server or a server cluster composed of multiple servers.

As shown in Figure 3, in one embodiment, a server high availability evaluation method is provided. This embodiment mainly illustrates the application of this method to the proxy server 104 in Figure 1 above. Referring to Figure 3, the server high availability evaluation method specifically includes the following steps:

S302, obtain the high availability service catalog.

Among them, the high-availability service directory contains information such as service types and service contents provided by the high-availability server for different users.

In one embodiment, before obtaining the high-availability service catalog, the proxy server receives the service level agreement (Service Level Agreement, SLA) sent by the high-availability server, then creates a high-availability service catalog according to the SLA, and stores the high-availability service catalog. The service catalog is stored in the database. When evaluating the high availability of a high availability server, the high availability service catalog is read from the database. Among them, the service level agreement is an agreement or contract on the service quality level determined by both the service provider and the user through negotiation. The purpose of formulating this agreement or contract is to make the service provider and the user understand the service quality, level, performance and priority. Reach a consensus on rights, responsibilities, etc. The service level agreement contains relevant parameters of the service level agreement. The parameters and their corresponding descriptions are shown in the table below.

Table 1 Service Level Agreement Parameter Table

parameter describe CPU capacity High availability system virtual machine CPU running speed Memory Capacity High availability system virtual machine flash memory space size Import time Preparation and import time before official use storage The amount of storage space required for user data Availability The availability of the SLA service signed by the user during a specified period of time Business response time The time it takes for a service provider to respond to and process high-availability service requests signed by users with SLA

S304: Select the target feature vector from the high-availability service catalog as the evaluation index.

Among them, the target feature vector is a feature vector that can reflect the individual characteristics of users corresponding to different terminals. The feature vector can be at least one of memory capacity (Memory), response time (Time) and storage (Storage). High-availability service The catalog contains all measurable feature vectors.

In one embodiment, after obtaining the high-availability service directory, the proxy server uses a machine learning algorithm to select the target feature vector from the high-availability service directory. The machine learning algorithm used may be a machine learning algorithm based on a decision tree. Specifically, It is the ID3 algorithm, and the decision tree is a tree built based on decision-making. In machine learning, a decision tree is a prediction model that represents a mapping relationship between object attributes and object values. Each node represents an object, and each bifurcated path in the tree represents an object. Possible attribute values, and each leaf node corresponds to the value of the object represented by the path from the root node to the leaf node. The decision tree has only a single output. If there are multiple outputs, independent decision trees can be established to process different outputs. The ID3 algorithm is a greedy algorithm used to construct decision trees.

The above embodiment is explained as an example. First, let the given training set be TR. The elements of TR are represented by feature vectors and their classification results. The attribute table Attrlist of the classification object is [A _l , A ₂ ,...A _n ]. The set Class composed of all the classification results is {C _l , C ₂ ...C _m }, generally n≥1 and m≥2. For each attribute A _i , its value range is ValueType(A _i ), and the value range is discrete. In this way, an element of TR can be expressed in the form of <X, C>, where X = (a ₁ ,...a _n ), a _i corresponds to the value of the i-th attribute of instance Classification results of instance X; then randomly select a subset of training instances to form a training window, and repeat the following steps:

① Construct a decision tree for the instance set in the window;

②Find a counterexample to the decision tree;

③If the counterexample exists, add it to the training window and perform step ①; if the counterexample does not exist, return the obtained decision tree. Select the attribute with the largest amount of information as the extended attribute. This heuristic rule is also called the principle of minimum entropy, which means that maximizing the amount of information obtained is equivalent to minimizing uncertainty, even if entropy is minimized. The given subset of positive and negative instances is S, which constitutes the training window. When the decision has k different outputs, the entropy of S is:

Among them, _Pi is the probability of taking the i-th value among k different outputs.

In order to detect the importance of each attribute, its importance can be evaluated by the information gain of each attribute. For attribute A, assuming its value range is (V _l , V ₂ ,...V _n ), the information gain of attribute A in training instance S can be defined as follows:

Among them, S _i represents the subset of attribute A in the training instance S whose value is V _i , and |S _i | represents the potential of the set. In actual situations, the largest attribute A of Gain(S,A) includes three feature vectors, namely: memory capacity, response time and storage.

S306: Allocate corresponding resources in the server to the target user according to the evaluation index.

Among them, the corresponding resources in the server are the resources of the high-availability server, where the resources of the server include the computing resources, storage resources and network resources of the server.

In one embodiment, after the proxy server selects the target feature vector from the high-availability service directory as the evaluation index, it uses the consistency analysis of the alpha algorithm to analyze the user data to obtain the matching degree between the user needs and the actual process and the degree of user needs. Fitness, and then allocate corresponding resources in the server to the target user based on the evaluation index, matching degree and fitness.

Explain the function of the α algorithm: match the event log L to a Petri net, then the Petri net becomes the behavioral expression of the event log. The input of the α algorithm is an event log L based on task T, and the output is a Petfi net. Our SLA limits the supplier and ensures quality throughout the entire service process for users.

Since participants may access some states that are not included in the expected model when completing tasks or the expected model may contain events that do not appear in the log file, a variable (log coverage C) is defined to evaluate the expected model (user requirements) Match with the actual process.

Among them, |E| represents the number of all events, and |e∈E| represents the number of actual events.

Fitness refers to the degree of correlation between the actual operation trajectory and the expected model. Through the consistency analysis of the actual instance process through the expectation model, the fitness of each instance and the entire task relative to the expectation model can be obtained (based on Petri net). The fitness of the desired model for each instance process is defined as follows:

Among them: m is the number of tokens lost when the actual process is reproduced on the Petri net model; c is the number of tokens consumed; r is the number of tokens that still exist after the task is completed; p is the number of tokens generated .

S308: Obtain the historical usage value of the resources in the server used by the target user within the historical period.

S310: Predict the predicted value of resources in the server used by the target user within a preset period based on historical usage values.

In one embodiment, after the proxy server allocates corresponding resources in the high-availability server to the target user based on the evaluation indicators, it monitors the practical performance and resource usage of the application running on the corresponding terminal of the target in real time, and assigns the corresponding resources to the target user. Use the value to store into the database.

In one embodiment, the proxy server obtains the historical usage value of the resources in the server used by the target user within the historical period from the database, and then builds a prediction model based on the obtained historical usage value. The created prediction model is used to predict the target user's use of resources in the server during the historical period. The preset period uses the predicted value of the resources in the server, and the prediction model may specifically be a time series prediction model.

In one embodiment, the proxy server obtains a preset period to be predicted, inputs the preset period into a prediction model, and uses the prediction model to predict the predicted value of the target user's use of resources in the high-availability server within the preset period. For example, when the evaluation indicator is memory capacity, what is obtained is the historical usage value of the server memory capacity used by the target user in the historical period, and what is predicted is the predicted value of the server memory capacity used by the target user in the preset period; when the evaluation indicator When it is the response time, what is obtained is the historical response time of the target user within the historical period, and what is predicted is the predicted response time of the target user using resources in the server to respond to the target user's request within the preset period.

S312: Obtain the actual usage value of resources in the server used by the target user within the preset period.

S314: Calculate user satisfaction based on actual usage value and predicted value.

In one embodiment, after predicting the predicted value of the target user's use of the resources in the server within the preset period, the proxy server obtains the predicted actual usage value of the target user's use of the resources in the server within the preset period, and based on the actual Calculate user satisfaction using values and predicted values.

In one embodiment, after the proxy server predicts the predicted value of the target user's use of the resources in the server within a preset period, it constructs an evaluation model based on the actual usage value and the predicted value; performs availability evaluation through the evaluation model to obtain the availability evaluation result. Then the satisfaction is calculated based on the usability evaluation results. Specifically, the predicted value and the usability evaluation results can be input into the user satisfaction formula to calculate the user satisfaction.

S316: Determine server availability based on user satisfaction.

In one embodiment, after calculating user satisfaction, the service level agreement corresponding to the target user is obtained, and the availability threshold is extracted from the service agreement level. When the user satisfaction is greater than the availability threshold, the corresponding high availability server is determined. Availability is normal. When user satisfaction is less than the availability threshold, the availability of the corresponding high-availability server is determined to be abnormal.

The above embodiment is explained as an example. Use f ₁ , f ₂ and f ₃ to represent the sampled memory capacity, response time and storage respectively. Organize the historical data into a time series and use Sn to represent it. Use f ₁ , f ₂ and f ₃ to predict the corresponding preset period. The time series S' _n , the prediction results Y _p are represented by Y ₁ , Y ₂ and Y ₃ respectively. Let S _it be the actual usage value of the i-th availability evaluation indicator at time t, and Y _it be the predicted value obtained by predicting the i-th availability evaluation indicator by the prediction model at time t, then the prediction error is:

The sum of squared errors is:

Among them, ω _i is the weight corresponding to the i-th evaluation index. Taking the minimum sum of squares of prediction errors as the optimization goal to determine the optimal weighting coefficients ω ₁ , ω ₂ and ω ₃ for f ₁ , f ₂ and f ₃ , the evaluation model can be expressed as:

f＝f ₁ ω ₁ +f ₂ ω ₂ +f ₃ ω ₃ ，ω ₁ +ω ₂ +ω ₃ =1

After constructing the evaluation model, input the target feature vector corresponding to the evaluation index into the evaluation model, calculate the usability evaluation result Y, and then input the predicted value Y _p and the usability evaluation result Y into the user satisfaction formula to calculate user satisfaction Degree, the formula of user satisfaction is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

After calculating the user satisfaction Q, obtain the availability threshold Q _t corresponding to the target user, and then calculate the availability A of the high-availability server based on the user satisfaction Q and the availability threshold Q _t , where:

Among them, A=1 indicates that the availability of the high-availability server is normal, and A=0 indicates that the availability of the high-availability server is abnormal.

The following table shows the availability of high-availability servers calculated through the simulation system (Figure 3). The simulation system selects time points t ₁ to t ₇ to record the two evaluation indicators T and M determined by the decision tree to calculate user satisfaction, T is Response time, the corresponding unit is μs, M is the memory capacity rate, Q is user satisfaction and A is system availability. Among them, at time point t ₄ , the state of the simulation system was artificially adjusted, causing a large deviation in the predicted value corresponding to time point t _4. As can be seen from the table below, when the availability of the high-availability server is artificially destroyed , T, M and S have also changed accordingly, and when the availability of the server is low, changes in server availability can be accurately reported through the server high availability evaluation method of this application.

Table 2 Availability of high-availability servers

time point t ₁ t _{2 t3 t4 t5 t6 t7} T 121 120 121 507 120 119 121 M 17 18 17 42 19 18 19 Q 4 4 4 2 4 4 4 A 1 1 1 0 1 1 1

In one embodiment, the proxy server monitors the practical performance and resource usage of applications run by the target user's corresponding terminal, and determines whether the actual usage value of the resources in the server used by the target user within a preset period meets the user needs of the target user. ; If not, re-allocate the corresponding resources in the server to the target user. Specifically set the fitness threshold F. When |fitness|>F, the system enters an interruption and selects automatic or manual reset according to the user's specific instance.

In the above embodiment, after obtaining the high-availability service catalog, the proxy server selects the target feature vector from the high-availability service catalog as the evaluation index, allocates corresponding resources in the server to the target user according to the evaluation index, and obtains the target user's usage in the historical period. The historical usage value of resources in the server is used to predict the predicted value of the target user's use of the server's resources within the preset period based on the historical usage value, and the actual usage value of the target user's use of the server's resources within the preset period is obtained, so as to calculate the actual usage value and The predicted value calculates user satisfaction and determines server availability based on user satisfaction. In this way, the high availability of the server can be evaluated on the premise of meeting the user's demand for high availability service quality.

In one embodiment, a server high availability evaluation method is also provided, and this method is applied to the proxy server 104 in Figure 1 as an example. Referring to Figure 4 and Figure 5, the server high availability evaluation method specifically includes the following steps:

Step 1, high-availability service evaluation, including the establishment of a high-availability service catalog.

Step 2: SLA evaluation system design, including SLA static evaluation system and implementation evaluation system.

The accuracy of assessments is improved through machine learning algorithms based on recorded data from users' corresponding assessments and simulated systems.

Step 3, SLA allocation.

A consistency analysis based on α algorithm is developed to explore user data and match the event log L to a Petri net, then the Petri net becomes the behavioral representation of the event log.

Step 4, SLA monitoring.

Raw data should be processed and analyzed against SLA targets or SLA thresholds to reflect SLA performance. Specifically, the fitness threshold F is set. When |fitness|>F, the system enters an interruption and selects automatic or manual reset according to the user's specific instance.

Step 5, SLA evaluation.

Evaluate high-availability servers based on the SLA static evaluation system and SLA implementation evaluation system.

Step 6, SLA revision.

Integrate information from SLA static evaluation systems and SLA implementation evaluation systems to obtain the final value of high availability service provider trustworthiness.

It should be understood that although the various steps in the flowcharts of Figures 3 and 4 are shown in sequence as indicated by arrows, these steps are not necessarily executed in the order indicated by arrows. Unless explicitly stated in this article, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in Figures 3 and 4 may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but may be executed at different times. The execution of these steps or stages The sequence is not necessarily sequential, but may be performed in turn or alternately with other steps or at least part of steps or stages in other steps.

In one embodiment, as shown in Figure 6, a server high availability evaluation device is provided, including: a high availability service directory acquisition module 602, a server resource allocation module 606, a resource usage value acquisition module 608, a prediction module 610, a user Satisfaction calculation module 612 and availability determination module 614, wherein:

The high-availability service catalog acquisition module 602 is used to obtain the high-availability service catalog;

The evaluation index determination module 604 is used to select a target feature vector from the high-availability service catalog as an evaluation index;

Server resource allocation module 606, used to allocate corresponding resources in the server to target users according to the evaluation indicators;

The resource usage value acquisition module 608 is used to obtain the historical usage value of the resources in the server used by the target user within the historical period;

The prediction module 610 is configured to predict the predicted value of the resource in the server used by the target user within a preset period according to the historical usage value;

The resource usage value acquisition module 608 is also used to obtain the actual usage value of the resources in the server used by the target user within the preset period;

User satisfaction calculation module 612, used to calculate user satisfaction based on the actual usage value and the predicted value;

The availability determination module 614 is used to determine the availability of the server according to the user satisfaction level.

In one embodiment, the evaluation index determination module 604 is also used to:

Based on the decision tree algorithm, the target feature vector is selected from the high-availability service catalog as an evaluation index through a machine learning algorithm.

In one embodiment, the server resource allocation module 606 is also used to:

Use the consistency analysis of α algorithm to analyze user data to obtain the matching degree of user needs and the actual process and the adaptability of user needs;

According to the evaluation index, the matching degree and the fitness degree, corresponding resources in the server are allocated to the target user.

In one embodiment, the prediction module 610 is also used to:

Get the default time period;

Enter the preset period into a prediction model; the prediction model is a time series model obtained based on the historical usage value;

The predicted value of the target user's use of resources in the server within the preset period is predicted through the prediction model.

In one embodiment, as shown in Figure 7, the device further includes: a user demand judgment module 616, wherein:

The user demand judgment module 616 is used to judge whether the actual usage value of the resources in the server used by the target user within the preset period meets the user demand of the target user;

The server resource allocation module 606 is also configured to re-allocate resources to the target user if the actual usage value of the resources in the server used by the target user within the preset period does not meet the user needs of the target user. The corresponding resources in the server.

In one embodiment, the user satisfaction calculation module 612 is also used to:

Construct an evaluation model based on the actual usage value and the predicted value;

Conduct usability evaluation through the evaluation model to obtain usability evaluation results;

Calculate user satisfaction based on the usability evaluation results.

In one embodiment, the user satisfaction calculation module 612 is also used to:

The predicted value and the usability evaluation result are input into the user satisfaction formula to obtain user satisfaction; the user satisfaction formula is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

In the above embodiment, after obtaining the high-availability service catalog, the proxy server selects the target feature vector from the high-availability service catalog as the evaluation index, allocates corresponding resources in the server to the target user according to the evaluation index, and obtains the target user's usage in the historical period. The historical usage value of resources in the server is used to predict the predicted value of the target user's use of the server's resources within the preset period based on the historical usage value, and the actual usage value of the target user's use of the server's resources within the preset period is obtained, so as to calculate the actual usage value and The predicted value calculates user satisfaction and determines server availability based on user satisfaction. In this way, the high availability of the server can be evaluated on the premise of meeting the user's demand for high availability service quality.

For specific limitations on the server high availability evaluation device, please refer to the above limitations on the server high availability evaluation method, which will not be described again here. Each module in the above-mentioned server high availability evaluation device can be implemented in whole or in part by software, hardware and combinations thereof. Each of the above modules may be embedded in or independent of the processor of the computer device in the form of hardware, or may be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be shown in Figure 8 . The computer device includes a processor, memory, and network interfaces connected through a system bus. Wherein, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage medium stores operating systems, computer programs and databases. This internal memory provides an environment for the execution of operating systems and computer programs in non-volatile storage media. The computer device's database is used to store resource usage value data. The network interface of the computer device is used to communicate with external terminals through a network connection. The computer program, when executed by a processor, implements a method of evaluating server high availability.

Those skilled in the art can understand that the structure shown in Figure 8 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Specific computer equipment can May include more or fewer parts than shown, or combine certain parts, or have a different arrangement of parts.

In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the following steps: obtain a high-availability service directory; Select the target feature vector from the service catalog as the evaluation index; allocate corresponding resources in the server to the target user based on the evaluation index; obtain the historical usage value of the resources in the server used by the target user in the historical period; predict the target user's expected use time based on the historical usage value. Set the predicted value of resources in the server used within a period; obtain the actual usage value of resources in the server used by target users within the preset period; calculate user satisfaction based on the actual usage value and predicted value, and determine the availability of the server based on user satisfaction.

In one embodiment, when the computer program is executed by the processor to select the target feature vector as the evaluation index from the high-availability service catalog, the processor is caused to specifically perform the following steps: based on the decision tree algorithm, using the machine learning algorithm to The target feature vector is selected from the available service catalog as the evaluation index.

In one embodiment, when the computer program is executed by the processor to allocate corresponding resources in the server to the target user according to the evaluation index, the processor is caused to specifically perform the following steps: analyze the user data using consistency analysis of the α algorithm, and obtain The degree of matching between user needs and the actual process and the degree of adaptability of user needs; based on the evaluation indicators, degree of matching and degree of adaptability, allocate corresponding resources in the server to target users.

In one embodiment, when the computer program is executed by the processor to predict the predicted value of the target user's use of resources in the server within a preset period based on historical usage values, the processor is caused to specifically perform the following steps: obtain the preset period; The preset period is input to the prediction model; the prediction model is a time series model based on historical usage values; the predicted value of the target user's use of resources in the server within the preset period is predicted through the prediction model.

In one embodiment, when the computer program is executed by the processor, the processor also performs the following steps: determine whether the actual usage value of the resources in the server used by the target user within the preset period meets the user needs of the target user; if not, then Re-allocate the corresponding resources in the server to the target user.

In one embodiment, when the computer program is executed by the processor to calculate user satisfaction based on the actual usage value and the predicted value, the processor is caused to specifically perform the following steps: construct an evaluation model based on the actual usage value and the predicted value; perform the evaluation through the evaluation model Usability evaluation, obtain the usability evaluation results; calculate user satisfaction based on the usability evaluation results.

In one embodiment, when the computer program is executed by the processor to calculate user satisfaction based on usability evaluation results, the computer program causes the processor to specifically perform the following steps: input the predicted value and usability evaluation results into the user satisfaction formula to obtain user satisfaction ;The formula for user satisfaction is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, it causes the processor to perform the following steps: obtain a high-availability service directory; select from the high-availability service directory. The target feature vector is used as an evaluation index; allocate corresponding resources in the server to the target user according to the evaluation index; obtain the historical usage value of the resources in the server used by the target user in the historical period; predict the use of the target user in the preset period based on the historical usage value Predicted values of resources in the server; obtain actual usage values of resources in the server used by target users within a preset period; calculate user satisfaction based on actual usage values and predicted values, and determine server availability based on user satisfaction.

In one embodiment, when the computer program is executed by the processor to select the target feature vector as the evaluation index from the high-availability service catalog, the processor is caused to specifically perform the following steps: based on the decision tree algorithm, using the machine learning algorithm to The target feature vector is selected from the available service catalog as the evaluation index.

In one embodiment, when the computer program is executed by the processor to allocate corresponding resources in the server to the target user according to the evaluation index, the processor is caused to specifically perform the following steps: analyze the user data using consistency analysis of the α algorithm, and obtain The degree of matching between user needs and the actual process and the degree of adaptability of user needs; based on the evaluation indicators, degree of matching and degree of adaptability, allocate corresponding resources in the server to target users.

In one embodiment, when the computer program is executed by the processor to predict the predicted value of the target user's use of resources in the server within a preset period based on historical usage values, the processor is caused to specifically perform the following steps: obtain the preset period; The preset period is input to the prediction model; the prediction model is a time series model based on historical usage values; the predicted value of the target user's use of resources in the server within the preset period is predicted through the prediction model.

In one embodiment, when the computer program is executed by the processor, the processor also performs the following steps: determine whether the actual usage value of the resources in the server used by the target user within the preset period meets the user needs of the target user; if not, then Re-allocate the corresponding resources in the server to the target user.

In one embodiment, when the computer program is executed by the processor to calculate user satisfaction based on the actual usage value and the predicted value, the processor is caused to specifically perform the following steps: construct an evaluation model based on the actual usage value and the predicted value; perform the evaluation through the evaluation model Usability evaluation, obtain the usability evaluation results; calculate user satisfaction based on the usability evaluation results.

In one embodiment, when the computer program is executed by the processor to calculate user satisfaction based on usability evaluation results, the computer program causes the processor to specifically perform the following steps: input the predicted value and usability evaluation results into the user satisfaction formula to obtain user satisfaction ;The formula for user satisfaction is as follows:

Among them, Q is user satisfaction, Y _P is the predicted value, and Y is the usability evaluation result.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage. In the media, when executed, the computer program may include the processes of the above method embodiments. Any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM).

The technical features of the above embodiments can be combined in any way. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, all possible combinations should be used. It is considered to be within the scope of this manual.

The above-mentioned embodiments only express several implementation modes of the present application, and their descriptions are relatively specific and detailed, but they should not be understood as limiting the scope of the invention patent. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all fall within the protection scope of the present application. Therefore, the scope of protection of this patent application should be determined by the appended claims.

Claims (10)

1. A server high availability assessment method, comprising:

acquiring a high-availability service directory; the high-availability service catalog comprises service types and service contents provided by the high-availability server aiming at different users;

selecting a target feature vector from the high-availability service directory as an evaluation index; the evaluation index comprises a memory capacity index, a response time index and a storage index;

Distributing corresponding resources in the server to the target user according to the evaluation index;

acquiring a historical use value of the target user using the resources in the server in a historical period;

predicting a predicted value of the target user using the resources in the server in a preset period according to the historical use value;

acquiring an actual use value of the resources in the server used by the target user within the preset period;

determining a prediction error value corresponding to the evaluation index according to the actual use value and the prediction value corresponding to each evaluation index, and determining a prediction error square sum expression corresponding to the evaluation index; based on the prediction error square sum expression, determining weights corresponding to the evaluation indexes respectively by taking the minimum prediction error square sum as an optimization target; constructing an evaluation model based on weights corresponding to the evaluation indexes respectively, performing availability evaluation through the evaluation model to obtain an availability evaluation result, and calculating user satisfaction according to the availability evaluation result; the usability assessment model is as follows: f=f ₁ ω ₁ +f ₂ ω ₂ +f ₃ ω ₃ ，ω ₁ +ω ₂ +ω ₃ =1, where f ₁ Indicating the memory capacity index, f ₂ Indicating the response time index, f ₃ Representing a storage index, f representing an availability evaluation result; omega ₁ 、ω ₂ And omega ₃ Respectively f ₁ 、f ₂ And f ₃ Is determined by the optimal weighting coefficient of the (a);

and determining the availability of the server according to the user satisfaction.

2. The method of claim 1, wherein selecting the target feature vector from the high availability service directory as an evaluation index comprises:

and selecting a target feature vector from the high-availability service catalogue by a machine learning algorithm based on a decision tree algorithm as an evaluation index.

3. The method according to claim 1, wherein the allocating the corresponding resources in the server to the target user according to the evaluation index comprises:

analyzing the user data by adopting the consistency analysis of the alpha algorithm to obtain the matching degree of the user requirement and the actual process and the fitness of the user requirement;

and distributing corresponding resources in a server for the target user according to the evaluation index, the matching degree and the adaptability.

4. The method according to claim 1, wherein predicting a predicted value of the target user's use of the resource in the server for a preset period of time based on the historical use value comprises:

Acquiring a preset period;

inputting the preset time period into a prediction model; the predictive model is a time series model based on the historical use value;

and predicting a predicted value of the target user using the resources in the server in the preset period through the prediction model.

5. The method according to claim 1, wherein the method further comprises:

judging whether the actual use value of the resources in the server used by the target user in a preset period meets the user requirement of the target user or not;

if not, the corresponding resources in the server are allocated again to the target user.

6. The method according to claim 1, wherein the method further comprises:

a service level agreement sent by the high availability server is received and then a high availability service directory is created according to the service level agreement.

7. The method of claim 1, wherein said calculating user satisfaction from said usability assessment results comprises:

inputting the predicted value and the usability evaluation result into a user satisfaction formula to obtain user satisfaction; the user satisfaction formula is as follows:

Wherein, _Q for user satisfaction, Y _P And Y is the availability evaluation result for the predicted value.

8. A server high availability assessment apparatus, the apparatus comprising:

the high-availability service catalog acquisition module is used for acquiring the high-availability service catalog;

the evaluation index determining module is used for selecting a target feature vector from the high-availability service catalogue as an evaluation index; the evaluation index comprises a memory capacity index, a response time index and a storage index;

the server resource allocation module is used for allocating corresponding resources in the server to the target user according to the evaluation index;

a resource use value acquisition module, configured to acquire a historical use value of the target user using the resource in the server in a historical period;

the prediction module is used for predicting a predicted value of the target user using the resources in the server in a preset period according to the historical use value;

the resource use value acquisition module is further used for acquiring an actual use value of the resources in the server used by the target user in the preset period;

a user satisfaction calculating module for calculating the user satisfaction according to each evaluation fingerDetermining a prediction error value corresponding to the evaluation index by marking the actual use value and the prediction value, and determining a prediction error square sum expression corresponding to the evaluation index; based on the prediction error square sum expression, determining weights corresponding to the evaluation indexes respectively by taking the minimum prediction error square sum as an optimization target; constructing an evaluation model based on weights corresponding to the evaluation indexes respectively, performing availability evaluation through the evaluation model to obtain an availability evaluation result, and calculating user satisfaction according to the availability evaluation result; the usability assessment model is as follows: f=f ₁ ω ₁ +f ₂ ω ₂ +f ₃ ω ₃ ，ω ₁ +ω ₂ +ω ₃ =1, where f ₁ Indicating the memory capacity index, f ₂ Indicating the response time index, f ₃ Representing a storage index, f representing an availability evaluation result; omega ₁ 、ω ₂ And omega ₃ Respectively f ₁ 、f ₂ And f ₃ Is determined by the optimal weighting coefficient of the (a);

and the availability determination module is used for determining the availability of the server according to the user satisfaction.

9. A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method of any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method of any of claims 1 to 7.