CN119166481B - Cloud desktop scheduling method and system - Google Patents
Cloud desktop scheduling method and systemInfo
- Publication number
- CN119166481B CN119166481B CN202411305301.5A CN202411305301A CN119166481B CN 119166481 B CN119166481 B CN 119166481B CN 202411305301 A CN202411305301 A CN 202411305301A CN 119166481 B CN119166481 B CN 119166481B
- Authority
- CN
- China
- Prior art keywords
- cloud desktop
- resource
- usage
- time node
- system performance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3438—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment monitoring of user actions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention relates to the technical field of cloud desktop management, and particularly discloses a cloud desktop scheduling method and system, wherein the method comprises the steps of collecting user behavior data, real usage amount of various resources and system performance data; the method comprises the steps of processing user behavior data to obtain cloud desktop user viscosity, processing system performance data to obtain cloud desktop system performance evaluation indexes, comprehensively analyzing various resources of each time node according to various resource prediction utilization amounts and various resource real utilization amounts of each historical time node, comprehensively analyzing to obtain resource utilization amount prediction precision indexes, and evaluating resource utilization amount prediction results according to the resource utilization amount prediction precision indexes. According to the method and the device for estimating the predicted usage amount of the future time node, the predicted usage amount of the future time node is estimated, so that the stability and the safety of a system are improved, the resource usage is optimized, the user experience is improved, and higher operation efficiency is brought to enterprises.
Description
Technical Field
The invention relates to the technical field of cloud desktop management, in particular to a cloud desktop scheduling method and a cloud desktop scheduling system.
Background
Cloud desktop scheduling is an important ring in the field of cloud computing, and along with the development of artificial intelligence technology and cloud computing technology, providing more humanized cloud desktop service has become a normal state, and an efficient and user-friendly cloud desktop scheduling method has important significance for improving user experience and enhancing market competitiveness.
For example, the invention patent with the bulletin number of CN108769233B is a resource optimization distribution method based on desktop cloud, which comprises the following steps of S1, constructing a laboratory resource optimization model according to professional teaching desktop cloud D taking professional teaching as a starting point, teachers T defined in a teaching course range and students S completing a learning course target, calculating ideal desktop cloud quantity, recalculating the minimum desktop cloud quantity according to the limit of the number of professional teachers, and arranging physical laboratories with experimental teaching as full as possible according to the determined desktop cloud D quantity.
For example, the invention patent with bulletin number of CN104253865B is a two-stage management method of a hybrid desktop cloud service platform, which comprises the following steps of constructing a hybrid desktop cloud data center, sending a registration request to a central management node (Vmm-Server) by a Server node, receiving and processing the registration request of the Server node by the central management node, uniformly receiving a user request by the central management node through a service interface layer, responding to the user request by the central management node according to a two-layer scheduling method, and maintaining system load balance by the central management node by adopting a two-layer migration method according to the current system state and virtual machine running state.
However, in the process of realizing the technical scheme in the embodiment of the application, the technical problems at least exist in the prior cloud desktop scheduling method, such as more emphasis is placed on improving the resource utilization rate, but the situation that resources are unevenly distributed still possibly exists in the peak period, and the personalized demands of users cannot be fully satisfied.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a cloud desktop scheduling method and a cloud desktop scheduling system, which can effectively solve the problems related to the background art.
In order to achieve the purpose, the cloud desktop scheduling method is achieved through the following technical scheme that the cloud desktop scheduling method comprises the steps of collecting user behavior data, real usage amount of various resources and system performance data.
Processing the user behavior data to obtain the viscosity of the cloud desktop user, processing the system performance data to obtain the performance evaluation index of the cloud desktop system, and comprehensively analyzing the real usage of various resources to obtain the predicted usage of various resources of each time node.
And carrying out time node division according to the current time node to obtain each historical time node and each future time node, and comprehensively analyzing to obtain a resource usage prediction accuracy index according to various resource prediction usage amounts and various real resource usage amounts of each historical time node.
And evaluating the resource usage prediction result according to the resource usage prediction precision index, if the evaluation result is qualified, carrying out cloud desktop resource scheduling according to various prediction usage of each future time node, and if the evaluation result is unqualified, carrying out resource usage prediction again.
As a further method, the user behavior data is processed to obtain the viscosity of the cloud desktop user, and the specific processing process is that the user behavior data comprises accumulated login times, accumulated login duration and the number of users.
Extracting a cloud desktop user viscosity evaluation influence factor corresponding to a preset accumulated login frequency and a cloud desktop user viscosity evaluation influence factor corresponding to accumulated login duration from a cloud desktop database.
And comprehensively analyzing according to the accumulated login times and the accumulated login time of each user of each time node in each monitoring period to obtain the viscosity of the cloud desktop user of each time node in each monitoring period.
As a further method, the cloud desktop system performance evaluation index is obtained by processing the system performance data, wherein the specific processing process is that the system performance data comprises system response time and system throughput.
And extracting a cloud desktop system performance evaluation influence factor corresponding to the preset system response time and a cloud desktop system performance evaluation influence factor corresponding to the preset system throughput from the cloud desktop database.
And comprehensively analyzing and obtaining a cloud desktop system performance evaluation index according to the system response time and the system throughput of each time node in each monitoring period.
The method comprises the following steps of obtaining the predicted usage amount of various resources of each time node through comprehensive analysis, wherein the specific analysis process comprises the steps of extracting preset cloud desktop user viscosity, cloud desktop system performance evaluation indexes and predicted usage amount evaluation influence factors corresponding to the actual usage amount of various resources from a cloud desktop database.
And comprehensively analyzing and obtaining the predicted usage of various resources of each time node according to the user viscosity, the system performance evaluation index and the actual usage of various resources.
As a further method, the resource usage prediction accuracy index is a quantization index obtained by comparing and analyzing the predicted usage and the actual usage of each resource at each time node, and is used for quantifying the accuracy of the resource usage prediction.
As a further method, the cloud desktop system performance evaluation index is a quantization index obtained by analyzing the system response time and the system throughput of each time node in each monitoring period, and is used for quantizing the performance of the cloud desktop system.
The cloud desktop scheduling method is evaluated according to the resource usage prediction precision index, and the specific evaluation process comprises the steps of extracting a resource usage prediction precision threshold value from a cloud desktop database, comparing the resource usage prediction precision index with the resource usage prediction precision threshold value, evaluating the cloud desktop scheduling method as qualified if the resource usage prediction precision index is greater than or equal to the resource usage prediction precision threshold value, and evaluating the cloud desktop scheduling method as unqualified if the resource usage prediction precision index is less than the resource usage prediction precision threshold value.
As a further method, the predicted usage amount of each time node of each resource is expressed by the following specific numerical expression:
Wherein y ik represents the predicted usage of the kth resource of the ith time node, x ijk represents the usage of the kth resource of the jth time node, σ ik represents the resource prediction compensation parameter of the kth resource of the ith time node, ω 1 represents the predicted usage evaluation influence factor corresponding to the set historical average usage, ω 2 represents the predicted usage evaluation influence factor corresponding to the set resource prediction compensation parameter, i represents the number of time nodes, i=1, 2,3,...
As a further method, the prediction precision index has a specific numerical expression:
Wherein C represents a prediction accuracy index, e represents a natural constant, y ik represents a predicted usage amount of the kth resource of the ith time node, The actual usage amount of the kth resource of the ith time node is represented, Δy represents the set allowable deviation usage amount, i represents the number of time nodes, i=1, 2,3,..n, n represents the total number of time nodes, k represents the number of resources, k=1, 2,3,..h, h represents the total number of resources.
The second aspect of the application provides a cloud desktop scheduling method system which comprises a cloud desktop data acquisition module, wherein the cloud desktop data acquisition module is used for acquiring user behavior data, various real resource usage amounts and system performance data, the user behavior data comprise accumulated login times, accumulated login duration and system response time of a user, and the system performance data comprise system throughput and resource usage amounts.
The cloud desktop prediction usage analysis module is used for processing the user behavior data to obtain the viscosity of the cloud desktop user, processing the system performance data to obtain the cloud desktop system performance evaluation index, combining the real usage of various resources, and comprehensively analyzing to obtain the prediction usage of various resources of each time node.
The cloud desktop prediction accuracy analysis module is used for carrying out time node division according to the current time node to obtain each historical time node and each future time node, and comprehensively analyzing and obtaining a resource usage prediction accuracy index according to various resource prediction usage and various real resource usage of each historical time node.
And the cloud desktop scheduling method evaluation module is used for evaluating the resource usage prediction result according to the resource usage prediction precision index, if the evaluation result is qualified, carrying out cloud desktop resource scheduling according to various prediction usage of each future time node, and if the evaluation result is unqualified, carrying out resource usage prediction again.
The cloud desktop database comprises cloud desktop user viscosity evaluation influence factors corresponding to preset accumulated login times, cloud desktop user viscosity evaluation influence factors corresponding to accumulated login time, cloud desktop system performance evaluation influence factors corresponding to preset system response time, cloud desktop system performance evaluation influence factors corresponding to preset system throughput, predicted usage amount evaluation influence factors corresponding to resource usage amount and resource prediction compensation parameters and a resource usage amount prediction precision threshold.
Compared with the prior art, the embodiment of the invention has at least the following advantages or beneficial effects:
(1) According to the cloud desktop user viscosity evaluation method and device, through evaluating the cloud desktop user viscosity, a more targeted marketing strategy can be formulated, the user retention rate and satisfaction degree are improved, and secondly, abnormal user viscosity change possibly indicates problems in service, so that quick diagnosis and problem solving are facilitated, meanwhile, resource requirements can be predicted more accurately, and cost waste caused by excessive configuration is avoided.
(2) According to the cloud desktop system performance evaluation method, the cloud desktop system performance is evaluated, the resource requirements can be predicted more accurately, cost waste caused by excessive configuration is avoided, and more personalized services and support, such as additional support for high-load time periods, can be provided according to the change of the cloud desktop system performance evaluation index.
(3) According to the method and the device for estimating the predicted usage amount of the future time node, the predicted usage amount of the future time node is estimated, so that the stability and the safety of a system are improved, the resource usage is optimized, the user experience is improved, and higher operation efficiency is brought to enterprises.
Drawings
The present application is further illustrated by the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the present application, and other drawings can be obtained by one of ordinary skill in the art without inventive effort from the following drawings.
Fig. 1 is a schematic flow chart of a cloud desktop scheduling method according to the present invention.
Fig. 2 is a schematic diagram of module connection of the cloud desktop system according to the present invention.
FIG. 3 is a schematic diagram of a functional relationship between a resource prediction correction parameter evaluation index and a cloud desktop user viscosity evaluation index according to the present invention.
Detailed Description
The technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments, and all other embodiments obtained by those skilled in the art without making any inventive effort based on the embodiments of the present application are within the scope of protection of the present application.
Referring to fig. 1, a first aspect of the present application provides a cloud desktop scheduling method, including:
Collecting user behavior data, real usage amount of various resources and system performance data;
processing user behavior data to obtain cloud desktop user viscosity, processing system performance data to obtain cloud desktop system performance evaluation indexes, and comprehensively analyzing the real usage of various resources to obtain predicted usage of various resources of each time node;
Dividing time nodes according to the current time nodes to obtain historical time nodes and future time nodes, and comprehensively analyzing according to various resource prediction usage amounts and various real resource usage amounts of the historical time nodes to obtain a resource usage amount prediction accuracy index;
And evaluating the resource usage prediction result according to the resource usage prediction precision index, if the evaluation result is qualified, carrying out cloud desktop resource scheduling according to various prediction usage of each future time node, and if the evaluation result is unqualified, carrying out resource usage prediction again.
It should be understood that the actual usage of various resources in this embodiment includes CPU usage, memory usage, system disk usage, data disk usage, bandwidth usage, traffic usage, and GPU usage.
The cloud desktop user viscosity is obtained by processing user behavior data, wherein the user behavior data comprises accumulated login times, accumulated login duration and the number of users.
And extracting a cloud desktop user viscosity evaluation influence factor corresponding to the preset accumulated login times and a cloud desktop user viscosity evaluation influence factor corresponding to the accumulated login time from the cloud desktop database.
And comprehensively analyzing according to the accumulated login times and the accumulated login time of each user of each time node in each monitoring period to obtain the viscosity of the cloud desktop user of each time node in each monitoring period.
In a specific embodiment, the cumulative login times, the cumulative login duration and the number of users are monitored through functions provided by the cloud desktop platform. The system can be used for monitoring the accumulated login times and the number of users, so that the load condition of the system in different time periods can be known, the resources are further distributed to realize load balance, future resource demands can be predicted, capacity expansion plans can be made in advance, the system can cope with the load in the peak period, the accumulated login time is monitored, abnormal performance and security holes can be found timely, the problem can be rapidly located and solved, and the abnormally high login times or login time can be signs of security threats.
Specifically, the viscosity evaluation index of the cloud desktop user has the following specific numerical expression:
Wherein A ij represents the cloud desktop user viscosity evaluation index of the ith time node in the jth monitoring period, T ijf represents the accumulated login times of the ith user of the ith time node in the jth monitoring period, T 0 represents the critical accumulated login times, S ijf represents the accumulated login time of the ith user of the ith time node in the jth monitoring period, S 0 represents the critical accumulated login time, f 0 represents the critical user number, Indicating a cloud desktop user viscosity evaluation influence factor corresponding to the set accumulated login times,The viscosity evaluation influence factors of the cloud desktop users corresponding to the set accumulated login time are represented,And (3) indicating a cloud desktop user viscosity evaluation influence factor corresponding to the set number of users, wherein i indicates the number of time nodes, i=1, 2,3, & gt, n, n indicates the total number of time nodes, j indicates the monitoring period number, j=1, 2,3, & gt, m indicates the total monitoring period number, f indicates the number of users, f=1, 2,3, & gt, and z indicates the total number of users.
According to the method, the accumulated login times and the accumulated login duration of the user at a certain time node and the number of users at a current time node are combined, the dependent variable is obtained through comprehensive analysis, the more the accumulated login times of the users are, the longer the accumulated login duration is, the higher the dependence degree of the users on cloud desktop service is indicated, meanwhile, the increase of the number of the users indicates that the cloud desktop service is popular, and the comprehensive analysis can obtain a more comprehensive viscosity evaluation index of the cloud desktop user.
Table 1 cloud desktop user viscosity assessment index data example
As shown in table 1, the viscosity evaluation index of the cloud desktop user is determined by the cumulative login times, the cumulative login duration and the number of users together, in a specific embodiment, z=1, the critical cumulative login times are 10 times, the critical cumulative login duration is 1 hour, the number of critical users is 50, the viscosity evaluation influence factor of the cloud desktop user corresponding to the set cumulative login times is 0.3, the viscosity evaluation influence factor of the cloud desktop user corresponding to the set cumulative login duration is 0.4, and the viscosity evaluation influence factor of the cloud desktop user corresponding to the set number of users is 0.3.
The formula considers the accumulated login times, the accumulated login duration and the number of users of each user, and can provide more personalized services and support according to the behavior mode of the users, such as providing higher-level technical support or customizing services for the users with high-frequency use. Through standardized processing of the accumulated login times, the accumulated login duration and the number of users under different time nodes of different monitoring periods, comparison of the accumulated login times, the accumulated login duration and the number of users on the same level is ensured, fairness and comparability of evaluation are improved, meanwhile, time periods or user groups with higher dependence on cloud desktop service can be known, resource allocation is reasonably arranged, and user experience is optimized. The relative importance of the accumulated login times, the accumulated login duration and the influence of the number of users in the evaluation index is reflected by weighting, and the weights of different factors can be adjusted according to different requirements, so that the model has good adaptability. It can be seen that the greater the cumulative login times or cumulative login duration or number of users, the greater the cloud desktop user viscosity assessment index. Through evaluating cloud desktop user viscosity, a more targeted marketing strategy can be formulated, user retention rate and satisfaction are improved, and secondly abnormal user viscosity change possibly indicates problems in service, so that quick diagnosis and problem solving are facilitated, resource requirements can be predicted more accurately, and cost waste caused by excessive configuration is avoided.
In a specific embodiment, the range of the value of the influence factor of the viscosity evaluation of the cloud desktop user corresponding to the cumulative login times, the cumulative login duration and the number of users of each user is between 0 and 1, and the influence degree of different factors on the viscosity evaluation index of the final cloud desktop user can be flexibly adjusted by adjusting the value of the influence factor.
It should be understood that, in this embodiment, a mapping set of the cumulative login times, the cumulative login time lengths, the number of users and the corresponding viscosity evaluation influence factors of the cloud desktop user is constructed by the relationship between the cumulative login times, the cumulative login time lengths, the number of users and the viscosity evaluation index of the cloud desktop user in the historical data, the real-time cumulative login times, the cumulative login time lengths and the number of users are input, and the viscosity evaluation influence factors of the cloud desktop user corresponding to the mapping set are obtained from the mapping set.
In a specific embodiment, the viscosity evaluation index of the cloud desktop user is a quantization index obtained by analyzing the accumulated login times, the accumulated login duration and the number of users, and is used for quantizing the viscosity of the cloud desktop user.
The cloud desktop system performance evaluation index is obtained by processing the system performance data, wherein the system performance data comprises system response time and system throughput.
And extracting a cloud desktop system performance evaluation influence factor corresponding to the preset system response time and a cloud desktop system performance evaluation influence factor corresponding to the preset system throughput from the cloud desktop database.
And comprehensively analyzing and obtaining a cloud desktop system performance evaluation index according to the system response time and the system throughput of each time node in each monitoring period.
In a specific embodiment, throughput refers to the number of requests processed by the system per unit time, and response time refers to the time it takes for the system to respond to a request, i.e., from the time a user initiates a request at a client to the time the client receives a response back from the server ends. The system response time and the system throughput can be detected through the performance monitoring function built in the cloud desktop platform, the cloud desktop can be ensured to respond to the operation of a user rapidly through the monitoring of the system response time, smooth use experience is provided, meanwhile, the waiting time of the user is reduced, the working efficiency and the satisfaction are improved, the monitoring of the system throughput can help to judge the load condition of the system, reasonable allocation of resources is realized, the resources are ensured to be available when needed, and abnormal response time and throughput can indicate that the system is attacked or potential security holes exist, and the detection can help to identify the field needing to strengthen security measures.
Specifically, the cloud desktop system performance evaluation index has the following specific numerical expression:
wherein B ij represents a cloud desktop system performance evaluation index of an ith time node in a jth monitoring period, P ij represents a system response time of the ith time node in the jth monitoring period, P 0 represents a set maximum allowed system response time, Q ij represents a system throughput of the ith time node in the jth monitoring period, Q 0 represents a set maximum allowed system throughput, ρ 1 represents a cloud desktop system performance evaluation influence factor corresponding to the set response time, ρ 2 represents a cloud desktop system performance evaluation influence factor corresponding to the set throughput, i represents a number of time nodes, i=1, 2,3,...
The algorithm of the embodiment combines the system response time and the system throughput of each time node, the dependent variable is obtained through comprehensive analysis, the system response time and the system throughput are directly related, when the system throughput is increased, the system may need to process more requests, which may result in the longer response time of a single request, otherwise, if the system response time is too long, the system may not effectively process more requests, thereby limiting the throughput, and the comprehensive analysis can evaluate the system performance level on a specific time node in a specific monitoring period, so as to obtain a more comprehensive cloud desktop system performance evaluation index.
It should be explained that, in this embodiment, two key factors, namely, the system response time and the system throughput, are considered, so that it is possible to help identify the performance bottleneck, quickly locate the cause of the performance problem, and identify the period of low resource utilization efficiency, thereby saving the cost. By carrying out standardized processing on the system response time and the system throughput of different time nodes, the comparison of the system response time and the system throughput on the same magnitude is ensured, the fairness and the comparability of the evaluation are improved, and meanwhile, the performance of the cloud desktop service in different time periods can be known, so that resources are better allocated, and the load balancing is realized. By weighting the impact of system response time and system throughput, reflecting their relative importance in the evaluation index, the weights of different factors can be adjusted according to different requirements, so that the model has good adaptability. It is readily seen that the cloud desktop system performance assessment index is greater as the system response time is smaller and or the system throughput is greater. By evaluating the performance of the cloud desktop system, the resource requirements can be predicted more accurately, the cost waste caused by excessive configuration is avoided, and more personalized services and support, such as additional support for high-load periods, can be provided according to the change of the performance evaluation index of the cloud desktop system.
In a specific embodiment, the range of the value of the influence factor of the cloud desktop system performance evaluation corresponding to the system response time and the system throughput of each time node is between 0 and 1, and the influence degree of different factors on the final cloud desktop system performance evaluation index can be flexibly adjusted by adjusting the value of the influence factor.
It should be understood that, in this embodiment, by using the relation between the system response time and the system throughput in the historical data and the cloud desktop system performance evaluation index, a mapping set of the system response time and the system throughput and the corresponding cloud desktop system performance evaluation influence factor is constructed, the real-time system response time and the system throughput are input, and the cloud desktop system performance evaluation influence factor corresponding to the mapping set is obtained from the mapping set.
In a specific embodiment, the cloud desktop system performance evaluation index is a quantization index obtained by analyzing the system response time and the system throughput, and is used for quantizing the cloud desktop system performance.
The method comprises the steps of comprehensively analyzing and obtaining the predicted usage amount of various resources of each time node, wherein the specific analysis process comprises the steps of extracting preset cloud desktop user viscosity, cloud desktop system performance evaluation indexes and predicted usage amount evaluation influence factors corresponding to the actual usage amount of various resources from a cloud desktop database.
And comprehensively analyzing and obtaining the predicted usage of various resources of each time node according to the user viscosity, the system performance evaluation index and the actual usage of various resources.
Specifically, the resource prediction correction parameter evaluation index of each time node has the following specific numerical expression:
Wherein epsilon i represents a resource prediction correction parameter evaluation index of an ith time node, A ij represents a cloud desktop user viscosity evaluation index of the ith time node in the jth monitoring period, B ij represents a cloud desktop system performance evaluation index of the ith time node in the jth monitoring period, theta 1 represents a resource prediction correction parameter evaluation influence factor corresponding to the set user viscosity, theta 2 represents a resource prediction correction parameter evaluation influence factor corresponding to the set system performance evaluation index, i represents the number of time nodes, i=1, 2,3, n, n represents the total number of time nodes, j represents the monitoring period number, j=1, 2,3, m, m represents the total monitoring period number.
According to the method, a cloud desktop user viscosity evaluation index and a cloud desktop system performance evaluation index of a certain time node are combined, a dependent variable is obtained through comprehensive analysis, if the system performance is good, the user viscosity is generally high, and the good performance is beneficial to improving user experience, so that the degree of dependence of a user on services is increased, more resources are required to maintain the service quality, the comprehensive analysis can evaluate the resource requirements more comprehensively, and resource prediction is adjusted accordingly.
As shown in fig. 3, in a specific embodiment, m=1, θ 1＝0.5,θ2 =0.3. When B ij =0.1, the functional relation between the resource prediction correction parameter evaluation index and the cloud desktop user viscosity evaluation index is shown as a curve a, when B ij =0.5, the functional relation between the resource prediction correction parameter evaluation index and the cloud desktop user viscosity evaluation index is shown as a curve B, and when B ij =1, the functional relation between the resource prediction correction parameter evaluation index and the cloud desktop user viscosity evaluation index is shown as a curve c.
It should be explained that, in this embodiment, two key factors, namely, the viscosity evaluation index of the cloud desktop user and the performance evaluation index of the cloud desktop system, are comprehensively considered, so that future resource demands can be more accurately predicted, and resource allocation can be dynamically adjusted according to the change trend of the indexes, so as to ensure that resources are available when needed. By carrying out summation and averaging on the cloud desktop user viscosity and the cloud desktop system performance evaluation index of different monitoring periods, accurate distribution of resource requirements of future time nodes according to the cloud desktop user viscosity and the cloud desktop system performance evaluation index of different time nodes is facilitated, load balancing is achieved, and user experience is improved. By weighting the influence of the viscosity of the cloud desktop user and the performance of the cloud desktop system, the relative importance of the cloud desktop user viscosity and the performance of the cloud desktop system in the evaluation index is reflected, and the weight of different factors can be adjusted according to different requirements, so that the model has good adaptability. When the viscosity evaluation index of the cloud desktop user is larger and the performance evaluation index of the cloud desktop system is larger, the evaluation index of the resource prediction correction parameter is larger. By evaluating the resource prediction correction parameters, the influence degree of the viscosity evaluation index of the cloud desktop user and the performance evaluation index of the cloud desktop system on the predicted use amount of the resource can be quantified, and the method is beneficial to helping a management layer to make a more intelligent decision, such as timely increase or decrease of the resource.
In a specific embodiment, the range of the value of the resource prediction correction parameter evaluation influence factor corresponding to the cloud desktop user viscosity evaluation index and the cloud desktop system performance evaluation index of each time node is between 0 and 1, and the influence degree of different factors on the final resource prediction correction parameter evaluation index can be flexibly adjusted by adjusting the value of the influence factor.
It should be understood that, in this embodiment, by using the relationship between the cloud desktop user viscosity evaluation index and the cloud desktop system performance evaluation index in the historical data and the resource prediction correction parameter evaluation index, a mapping set of the cloud desktop user viscosity evaluation index and the cloud desktop system performance evaluation index and the corresponding resource prediction correction parameter evaluation influence factor is constructed, the real-time cloud desktop user viscosity evaluation index and the cloud desktop system performance evaluation index are input, and the corresponding resource prediction correction parameter evaluation influence factor is obtained from the mapping set.
In a specific embodiment, the resource prediction correction parameter evaluation index is a quantization index obtained by analyzing the cloud desktop user viscosity evaluation index and the cloud desktop system performance evaluation index, and is used for quantizing the resource prediction correction parameter.
Specifically, the predicted usage amount of each time node of each resource is represented by the following specific numerical expression:
Wherein y ik represents the predicted usage of the kth resource of the ith time node, x ijk represents the usage of the kth resource of the jth time node, σ ik represents the resource prediction compensation parameter of the kth resource of the ith time node, ω 1 represents the predicted usage evaluation influence factor corresponding to the set historical average usage, ω 2 represents the predicted usage evaluation influence factor corresponding to the set resource prediction compensation parameter, i represents the number of time nodes, i=1, 2,3,...
The algorithm combines the resource prediction using quantity and the resource prediction compensating parameter of the historical time node, comprehensively analyzes to obtain a dependent variable, and if a large deviation exists between the historical average using quantity and the actual using quantity, the resource prediction compensating parameter can be used for adjusting the predicted value to be closer to the actual using condition so as to cope with uncertainty and change, and comprehensively analyzes to more accurately predict the resource using quantity and schedule the resource accordingly.
It should be explained that, in this embodiment, the resource prediction compensation parameters corresponding to the resource prediction correction parameter evaluation indexes of different time nodes are extracted in the cloud desktop data set, and the resource prediction compensation parameters can be ensured to be in the same magnitude as the resource usage.
It should be explained that, in this embodiment, two key factors, that is, the historical time node resource usage amount and the resource prediction compensation parameter, are comprehensively considered, so that the requirements of each time node and each resource can be predicted more accurately. By averaging the sum of the resource usage of the historical time nodes of different monitoring periods, the behavior pattern of the user can be further understood and accordingly more personalized services can be provided. The relative importance of the historical time node resource usage and the resource prediction compensation parameter in the evaluation index is reflected by weighting the influence of the historical time node resource usage and the resource prediction compensation parameter, and the weights of different factors can be adjusted according to different requirements, so that the model has good adaptability. It is readily seen that the greater the resource usage of a historical time node and/or the greater the resource prediction compensation parameter, the greater the predicted usage of a future time node. By evaluating the predicted usage amount of the future time node, the influence degree of the resource usage amount of the historical time node and the predicted usage amount of the future time node by the resource prediction compensation parameter can be quantified, so that the stability and the safety of the system can be improved, the resource usage can be optimized, the user experience can be improved, and higher operation efficiency can be brought to enterprises.
In a specific embodiment, the range of the influence factor of the future time node resource prediction use amount corresponding to the historical time node resource use amount and the resource prediction compensation parameter is between 0 and 1, and the influence degree of different factors on the final future time node resource prediction use amount can be flexibly adjusted by adjusting the value of the influence factor.
It should be understood that, in this embodiment, by using the relation between the historical time node resource usage amount and the resource prediction compensation parameter in the historical data and the future time node resource prediction usage amount, a mapping set of the historical time node resource usage amount and the resource prediction compensation parameter and the corresponding future time node resource prediction usage amount influencing factor is constructed, the real-time historical time node resource usage amount and the resource prediction compensation parameter are input, and the future time node resource prediction usage amount influencing factor corresponding to the mapping set is obtained from the mapping set.
Specifically, the resource usage prediction accuracy index is a quantization index obtained by comparing and analyzing the predicted usage and the actual usage of each resource at each time node, and is used for quantifying the accuracy degree of the resource usage prediction.
Further, the resource usage prediction precision index has the following specific numerical expression:
Wherein C represents a prediction accuracy index, e represents a natural constant, y ik represents a predicted usage amount of the kth resource of the ith time node, The actual usage amount of the kth resource of the ith time node is represented, Δy represents the set allowable deviation usage amount, i represents the number of time nodes, i=1, 2,3,..n, n represents the total number of time nodes, k represents the number of resources, k=1, 2,3,..h, h represents the total number of resources.
It should be explained that, in this embodiment, it is not difficult to see that the closer the predicted usage amount and the actual usage amount of each resource of each time node are, the larger the resource usage amount prediction accuracy index is, and by evaluating the resource usage amount prediction accuracy index, the accuracy of cloud desktop resource usage amount prediction can be effectively improved, so that better services are provided for enterprises and individual users, and meanwhile, the cost is reduced and the efficiency is improved.
The cloud desktop scheduling method is evaluated according to the resource usage prediction precision index, wherein the cloud desktop scheduling method is evaluated as qualified if the resource usage prediction precision index is larger than or equal to the resource usage prediction precision threshold, and is evaluated as unqualified if the resource usage prediction precision index is smaller than the resource usage prediction precision threshold.
Referring to fig. 2, a second aspect of the present application provides a cloud desktop scheduling system, including:
The cloud desktop data acquisition module is used for acquiring user behavior data, real usage amount of various resources and system performance data, wherein the user behavior data comprises accumulated login times, accumulated login duration and system response time of a user;
The cloud desktop prediction usage analysis module is used for processing the user behavior data to obtain the viscosity of the cloud desktop user, processing the system performance data to obtain a cloud desktop system performance evaluation index, combining the real usage of various resources, and comprehensively analyzing to obtain the prediction usage of various resources of each time node;
the cloud desktop prediction precision analysis module is used for carrying out time node division according to the current time node to obtain each historical time node and each future time node, and comprehensively analyzing to obtain a resource usage prediction precision index according to various resource prediction usage and various real resource usage of each historical time node;
the cloud desktop scheduling method evaluation module is used for evaluating the resource usage prediction result according to the resource usage prediction precision index, if the evaluation result is qualified, carrying out cloud desktop resource scheduling according to various prediction usage of each future time node, and if the evaluation result is unqualified, carrying out resource usage prediction again;
In a specific embodiment, when the evaluation result of the resource usage prediction result is qualified, cloud desktop resource scheduling is performed according to various predicted usage amounts of each future time node, for example, by analyzing historical data, it is found that 9 to 11 am each day is a user login peak period, and at this time, the CPU and the memory usage rate are significantly increased, and then the resource allocation is adjusted in advance according to the prediction result, for example, the number of CPUs and the memory capacity are increased to cope with the peak period requirement.
The cloud desktop database is used for storing cloud desktop data, the cloud desktop data comprises a cloud desktop user viscosity evaluation influence factor corresponding to a preset accumulated login time, a cloud desktop user viscosity evaluation influence factor corresponding to an accumulated login time, a cloud desktop system performance evaluation influence factor corresponding to a preset system response time, a cloud desktop system performance evaluation influence factor corresponding to a preset system throughput, a resource usage amount and a prediction usage amount evaluation influence factor corresponding to a resource prediction compensation parameter, and a resource usage amount prediction precision threshold, and data in the cloud desktop database can be acquired through data acquisition of a plurality of cloud desktop experiments.
The foregoing is merely illustrative and explanatory of the constructions of the invention, as long as they do not depart from the spirit and scope of the invention, it should be understood by those skilled in the art that various changes or additions may be made to the specific embodiments described herein.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411305301.5A CN119166481B (en) | 2024-09-19 | 2024-09-19 | Cloud desktop scheduling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411305301.5A CN119166481B (en) | 2024-09-19 | 2024-09-19 | Cloud desktop scheduling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119166481A CN119166481A (en) | 2024-12-20 |
| CN119166481B true CN119166481B (en) | 2025-07-15 |
Family
ID=93884849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411305301.5A Active CN119166481B (en) | 2024-09-19 | 2024-09-19 | Cloud desktop scheduling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119166481B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109491782A (en) * | 2017-09-11 | 2019-03-19 | 中兴通讯股份有限公司 | A kind of method and device of cloud desktop intelligent management |
| CN112905909A (en) * | 2019-11-19 | 2021-06-04 | 腾讯科技（深圳）有限公司 | Data prediction method and device, computer readable storage medium and electronic equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10062354B2 (en) * | 2014-10-10 | 2018-08-28 | DimensionalMechanics, Inc. | System and methods for creating virtual environments |
| CN113556762A (en) * | 2021-07-15 | 2021-10-26 | 支付宝(杭州)信息技术有限公司 | Resource allocation method, device, equipment and medium |
| US20230131611A1 (en) * | 2021-10-21 | 2023-04-27 | Dell Products L.P. | Generating temporal predictions for provisioning cloud resources using trained machine learning techniques |
| CN118467177A (en) * | 2024-07-09 | 2024-08-09 | 宜宾职业技术学院 | Method and system for allocating and managing computer cloud desktop resources |
-
2024
- 2024-09-19 CN CN202411305301.5A patent/CN119166481B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109491782A (en) * | 2017-09-11 | 2019-03-19 | 中兴通讯股份有限公司 | A kind of method and device of cloud desktop intelligent management |
| CN112905909A (en) * | 2019-11-19 | 2021-06-04 | 腾讯科技（深圳）有限公司 | Data prediction method and device, computer readable storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119166481A (en) | 2024-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112835698B (en) | Dynamic load balancing method for request classification processing based on heterogeneous clusters | |
| CN106899660B (en) | Implementation method of energy-saving scheduling in cloud data center based on rolling grey prediction model | |
| CN103294546B (en) | The online moving method of virtual machine of multi-dimensional resource performance interference aware and system | |
| CN111132235B (en) | Mobile offload migration algorithm based on improved HRRN algorithm and multi-attribute decision | |
| CN119232741B (en) | Distributed network resource optimal scheduling method and system under load balancing strategy | |
| CN115297112A (en) | Dynamic resource quota and scheduling component based on Kubernetes | |
| CN108628662A (en) | Mix the resource elastic telescopic method based on load estimation under cloud environment | |
| CN113467944A (en) | Resource deployment device and method for complex software system | |
| CN111752706A (en) | Resource allocation method, device and storage medium | |
| CN119383675B (en) | Dynamic bandwidth allocation method, system, electronic equipment and medium based on large model | |
| CN118394592B (en) | A Paas platform based on cloud computing | |
| CN119536958B (en) | Method, system, medium and program product for balancing host power consumption and temperature | |
| Gupta et al. | Long range dependence in cloud servers: a statistical analysis based on google workload trace | |
| CN114629909A (en) | Cloud resource cost analysis method | |
| CN119003138A (en) | Heterogeneous multi-core collaborative management and automatic operation and maintenance system | |
| CN119248490B (en) | Intelligent computing power scheduling method and system based on dynamic programming | |
| CN118939427B (en) | A resource allocation management method and computer equipment for cloud servers | |
| CN114444853A (en) | Performance control method, device and storage medium | |
| CN119166481B (en) | Cloud desktop scheduling method and system | |
| CN111107569B (en) | Method and device for screening problem cells | |
| CN112732451A (en) | Load balancing system in cloud environment | |
| CN115658319B (en) | Resource scheduling method, system, device and storage medium | |
| CN117729204A (en) | K8S container scheduling method and system based on monitoring perception | |
| KR20160044623A (en) | Load Balancing Method for a Linux Virtual Server | |
| Preetham et al. | Resource provisioning in cloud using ARIMA and LSTM technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |