Cloud desktop load balancing method and device, equipment and storage medium

Abstract

The invention relates to the technical field of cloud computing, in particular to a cloud desktop load balancing method and device, equipment and a storage medium, and the method comprises the steps: carrying out the data processing of a cloud desktop message queue set and a host machine message queue set within a preset time limit, and obtaining a cloud desktop limited performance data set and a host machine limited performance data set; respectively filtering the cloud desktop limited performance data set and the host machine limited performance data set based on a preset filtering time period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; respectively obtaining a cloud desktop performance index classification set and a host machine performance index classification set based on the filtered performance data set of the cloud desktop, the filtered performance data set of the host machine, a preset combined prediction model and a classification rule; and scheduling the cloud desktop corresponding to the cloud desktop performance index classification set to the host machine corresponding to the host machine performance index classification set matched with the cloud desktop performance index classification set. The experience of using the cloud desktop by the user can be improved.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a cloud desktop load balancing method, device, equipment and storage medium. Background Art

[0002] In large enterprises, users often use cloud desktops for office work. Since each user has different usage habits, for example, some users are accustomed to opening a lot of software and not closing them, which leads to high memory resource usage; some users are accustomed to using cloud desktops for development and performing a lot of calculations, which leads to high memory resource usage; cloud desktops rely on the host machine they occupy to perform calculations to meet user usage needs.

[0003] Generally, when allocating cloud desktops, the user's usage habits are not known, resulting in them being basically randomly allocated to an idle host machine. This may cause more users with the same type of resource utilization to be allocated to the same host machine at the same time. In addition, cloud desktops are generally allocated according to the over-allocation ratio, which may lead to resource contention. Moreover, when a host machine is too busy, some of its running cloud desktops will be migrated to other host machines. It can be seen that the current user experience of using cloud desktops is not good. Summary of the invention

[0004] In order to improve the user experience of using a cloud desktop, the present application provides a cloud desktop load balancing method, apparatus, device and storage medium.

[0005] In a first aspect, the present application provides a cloud desktop load balancing method, comprising:

[0006] Processing the cloud desktop message queue set and the host machine message queue set within a preset period of time to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set;

[0007] The cloud desktop deadline performance data set and the host machine deadline performance data set are filtered respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set;

[0008] Based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, a cloud desktop performance indicator classification set and a host machine performance indicator classification set are obtained respectively;

[0009] The cloud desktop corresponding to the cloud desktop performance indicator classification set is scheduled to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0010] Through the above implementation, the calculated cloud desktop performance indicator classification set and host machine performance indicator classification set can respectively understand the busyness of each performance indicator of each cloud desktop and host machine, so as to determine the host machine matching each cloud desktop, and then dispatch the cloud desktop to the corresponding host machine, thereby realizing load balancing of the cloud desktop and eliminating the phenomenon of cloud desktop resource competition, which is convenient for improving the user experience of using the cloud desktop.

[0011] Preferably, the cloud desktop message queue in the cloud desktop message queue set includes the cloud desktop CPU usage, cloud desktop memory occupancy, cloud desktop disk IO speed, cloud desktop network rate, cloud desktop IP address, cloud desktop MAC address and cloud desktop timestamp; the host message queue in the host message queue set includes the host CPU usage, host memory occupancy, host disk IO speed, host network rate, host IP address, host MAC address and host timestamp.

[0012] Through the above implementation, the data types in the cloud desktop message queue set and the host message queue set are disclosed, which facilitates the subsequent clarification of the objects for data processing on the cloud desktop message queue set and the host message queue set.

[0013] Preferably, the step of performing data processing on the cloud desktop message queue set and the host machine message queue set within a preset period to obtain the cloud desktop deadline performance data set and the host machine deadline performance data set includes:

[0014] Obtain the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, and cloud desktop network rate in each of the cloud desktop message queues within a preset period to obtain a cloud desktop deadline performance data set;

[0015] The host CPU usage rate, host memory occupancy rate, host disk IO speed, and host network rate in each host message queue within a preset period are obtained to obtain a host time-limited performance data set.

[0016] Through the above implementation, data processing is performed on the data in the cloud desktop message queue set and the host message queue set; it is convenient to remove unimportant data such as IP addresses, MAC addresses and timestamps in the cloud desktop message queue set and the host message queue set, thereby improving the efficiency and accuracy of data processing.

[0017] Preferably, the cloud desktop performance indicator classification set and the host machine performance indicator classification set are obtained based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, respectively, including:

[0018] Based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rules, a cloud desktop performance indicator classification set is obtained;

[0019] Based on the host machine filtered performance data set, a preset combined prediction model, and classification rules, a host machine performance indicator classification set is obtained.

[0020] Through the above implementation, the cloud desktop filtered performance data set and the host machine filtered performance data set are processed according to the preset combined prediction model and classification rules to obtain the corresponding performance indicator classification set; it is convenient to determine the host machine corresponding to the cloud desktop according to the performance indicator classification set.

[0021] Preferably, the cloud desktop performance indicator classification set is obtained based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rules, including:

[0022] Perform mean calculation on each performance indicator data set in the cloud desktop filtered performance data set to obtain the cloud desktop target performance data;

[0023] Processing the cloud desktop target performance data based on a preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets;

[0024] The predicted values ​​of the performance indicators are processed based on the classification rules to obtain a cloud desktop performance indicator classification set.

[0025] Through the above implementation, the cloud desktop filtered performance data set is first averaged, and then the performance indicator classification set is calculated; this facilitates accurate classification of each performance indicator of the corresponding cloud desktop.

[0026] Preferably, the combined prediction model includes an autoregressive model, a weighted moving average model and an exponential smoothing model; the autoregressive model is preset with a first model weight, the weighted moving average model is preset with a second model weight, and the exponential smoothing model is preset with a third model weight;

[0027] The processing of the cloud desktop target performance data based on the preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets includes:

[0028] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the autoregressive model in the combined prediction model for processing to obtain a first prediction value;

[0029] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into a weighted moving average model in the combined prediction model for processing to obtain a second prediction value;

[0030] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the exponential smoothing model in the combined prediction model for processing to obtain a third prediction value;

[0031] Based on the first prediction value, the second prediction value, the third prediction value, the first model weight, the second model weight and the third model weight, the performance indicator prediction value corresponding to each performance indicator data mean is calculated.

[0032] Through the above implementation, the performance indicator prediction value is calculated through weighted operation, which facilitates improving the calculation accuracy of the performance indicator prediction value.

[0033] In a second aspect, the present application provides a cloud desktop load balancing device, comprising:

[0034] A data processing module, used for processing the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set;

[0035] A data filtering module, used to filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set;

[0036] A model processing module, used to obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules;

[0037] The desktop scheduling module is used to schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0038] Through the above implementation, the calculated cloud desktop performance indicator classification set and host machine performance indicator classification set can respectively understand the busyness of each performance indicator of each cloud desktop and host machine, so as to determine the host machine matching each cloud desktop, and then dispatch the cloud desktop to the corresponding host machine, thereby realizing load balancing of the cloud desktop and eliminating the phenomenon of cloud desktop resource competition, which is convenient for improving the user experience of using the cloud desktop.

[0039] In a third aspect, the present application provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above method when executing the computer program.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps in the above method when executed by a processor.

[0041] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0042] The above-mentioned cloud desktop load balancing method, device, equipment and storage medium obtain a cloud desktop deadline performance data set and a host deadline performance data set by processing the cloud desktop message queue set and the host machine message queue set within a preset period; filter the cloud desktop deadline performance data set and the host machine deadline performance data set based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; obtain a cloud desktop performance index classification set and a host machine filtered performance data set based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and a classification rule. host performance indicator classification set; dispatching the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host corresponding to the host performance indicator classification set matched by the cloud desktop performance indicator classification set; through the above implementation, the busyness of each performance indicator of each cloud desktop and the host can be respectively understood through the calculated cloud desktop performance indicator classification set and the host performance indicator classification set, so as to determine the host matching each cloud desktop, and then dispatch the cloud desktop to the corresponding host, thereby realizing load balancing of the cloud desktop and eliminating the phenomenon of cloud desktop resource competition, which is convenient for improving the user experience of using the cloud desktop.

[0043] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A flow chart of a cloud desktop load balancing method provided in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of the structure of a cloud desktop load balancing device provided in an embodiment of the present application;

[0047] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application;

[0048] Figure 4 This is a diagram of the internal structure of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0050] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] In this article, the term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship.

[0052] Embodiment 1

[0053] Figure 1 A flow chart of a cloud desktop load balancing method provided in Example 1 of the present application, refer to Figure 1 The method may be performed by a device for performing the method, and the device may be implemented by software and / or hardware. The method includes:

[0054] S110 , performing data processing on a cloud desktop message queue set and a host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set.

[0055] Among them, the preset deadline is 15 working days in this embodiment; the cloud desktop message queue set includes multiple cloud desktop message queues, and the cloud desktop message queue contains multiple performance indicators of the corresponding cloud desktop; the host message queue set includes multiple host message queues, and the host message queue contains multiple performance indicators of the corresponding host; data processing in this embodiment specifically uses big data processing to extract data from each cloud desktop message queue in the cloud desktop message queue set within the preset deadline, so as to obtain the performance indicator corresponding to each cloud desktop message queue as the cloud desktop deadline performance data set, and extract data from each host message queue in the host message queue set within the preset deadline, so as to obtain the performance indicator corresponding to each host message queue as the host deadline performance data set.

[0056] Specifically, a big data processing method is used to extract data from each cloud desktop message queue in a cloud desktop message queue set within a preset period, thereby obtaining a cloud desktop deadline performance data set; a big data processing method is also used to extract data from each host message queue in a host message queue set within a preset period, thereby obtaining a host deadline performance data set.

[0057] S120: Filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0058] Among them, the preset filtering period in this embodiment is specifically 22:00-08:00, and the filtering period corresponds to the rest period of the cloud desktop user. Since the user basically does not use the cloud desktop during the filtering period, the reference value of the cloud desktop deadline performance data set and the host deadline performance data set corresponding to the filtering period is not high. Therefore, it is necessary to filter the cloud desktop deadline performance data set and the host deadline performance data set within the preset filtering period respectively, and the data after the cloud desktop deadline performance data set is filtered is recorded as the cloud desktop filtered performance data set, and the data after the host filtered performance data set is filtered is recorded as the host filtered performance data set.

[0059] Specifically, the data in the cloud desktop deadline performance data set that are within the preset filtering period are filtered to obtain the cloud desktop filtered performance data set; the data in the host machine deadline performance data set that are within the preset filtering period are also filtered to obtain the host machine filtered performance data set.

[0060] S130, based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, respectively obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set.

[0061] Among them, the preset combined prediction model is used to process the cloud desktop filtered performance data set, so as to obtain prediction data related to the cloud desktop filtered performance data set, and the prediction data is used to predict the performance indicator data of the cloud desktop corresponding to the cloud desktop filtered performance data set in the future time period; the preset combined prediction model is also used to process the host machine filtered performance data set, so as to obtain prediction data related to the host machine filtered performance data set, and the prediction data is used to predict the performance indicator data of the host machine corresponding to the host machine filtered performance data set in the future time period; the classification rules are used to classify the busyness of each performance indicator data predicted by the cloud desktop and the host machine, and the classification results obtained by classifying the busyness of the cloud desktop are recorded as the cloud desktop performance indicator classification set, and the classification results obtained by classifying the busyness of the host machine are recorded as the host machine performance indicator classification set.

[0062] Specifically, a preset combined prediction model is used to process the cloud desktop filtered performance data set to obtain prediction data related to the cloud desktop filtered performance data set, and then the prediction data is processed through classification rules to determine the busyness classification of each performance indicator data of the corresponding cloud desktop in a future time period, thereby obtaining a cloud desktop performance indicator classification set; a preset combined prediction model is used to process the host machine filtered performance data set to obtain prediction data related to the host machine filtered performance data set, and then the prediction data is processed through classification rules to determine the busyness classification of each performance indicator data of the corresponding host machine in a future time period, thereby obtaining a host machine performance indicator classification set.

[0063] S140: Schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0064] Among them, it is assumed that the cloud desktop performance indicator classification set includes three types of busyness classifications of performance indicator data, namely: the first performance indicator is idle, the second performance indicator is idle, and the third performance indicator is busy; if a host machine performance indicator classification set includes: the first performance indicator is busy, the second performance indicator is busy, and the third performance indicator is idle; then the cloud desktop performance indicator classification set matches the host machine performance indicator classification set; the so-called matching means that the busyness classification of the corresponding performance indicators in the cloud desktop performance indicator classification set and the host machine performance indicator classification set are in a complementary state.

[0065] Specifically, taking a cloud desktop performance indicator category set as an example, first determine the host machine performance indicator category set corresponding to the cloud desktop performance indicator category set, and then schedule the cloud desktop corresponding to the cloud desktop performance indicator category set to the host machine corresponding to the host machine performance indicator category set that matches the cloud desktop performance indicator category set.

[0066] It should be noted that, in this embodiment, a cloud desktop deadline performance data set and a host deadline performance data set are obtained by processing the cloud desktop message queue set and the host machine message queue set within a preset period; the cloud desktop deadline performance data set and the host machine deadline performance data set are filtered respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; a cloud desktop performance indicator classification set and a host machine performance indicator classification set are obtained respectively based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and a classification rule; the cloud desktop corresponding to the cloud desktop performance indicator classification set is scheduled to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set; through the above implementation, the calculated cloud desktop performance indicator classification set and the host machine performance indicator classification set can respectively understand the busyness of each performance indicator of each cloud desktop and the host machine, so that the host machine matching each cloud desktop can be determined, and then the cloud desktop can be scheduled to the corresponding host machine, so that the load balancing of the cloud desktop can be achieved, and the phenomenon of cloud desktop resource competition is eliminated, which is convenient for improving the user experience of using the cloud desktop.

[0067] Embodiment 2

[0068] A cloud desktop load balancing method is provided in the second embodiment of the present application. The method refines the "cloud desktop message queue set" and the "host machine message queue set" in the first embodiment. It should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments. The method includes:

[0069] S210: Process the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set.

[0070] Among them, the cloud desktop message queue in the cloud desktop message queue set includes the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, cloud desktop network rate, cloud desktop IP address, cloud desktop MAC address and cloud desktop timestamp; the host message queue in the host message queue set includes the host CPU usage rate, host memory occupancy rate, host disk IO speed, host network rate, host IP address, host MAC address and host timestamp.

[0071] It should be noted that a Pdh (Performance Data Helper) library is provided in the cloud desktop. The Pdh library is used to periodically obtain various performance indicator data of the cloud desktop. In this embodiment, the various performance indicator data include: cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, cloud desktop network rate, wherein the cloud desktop network rate specifically includes the network receiving packet rate and the network sending packet rate; in addition, the timing time interval is specifically 2s in this embodiment; the cloud desktop subsequently sends the various performance indicators collected by the cloud desktop to the server in the form of a message queue. In order to build a message queue, it is also necessary to collect the cloud desktop IP address, cloud desktop MAC address and cloud desktop timestamp related to the cloud desktop; a message queue obtained periodically is sent to the server. Taking the multiple performance indicator data of the secondary cloud desktop as an example, the collected cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, cloud desktop network rate are combined with the acquired cloud desktop IP address, cloud desktop MAC address and cloud desktop timestamp related to the cloud desktop to form a corresponding cloud desktop message queue; after generating a cloud desktop message queue, the cloud desktop sends the cloud desktop message queue to the server, so that the server receives and stores the cloud desktop message queue, and the server stores the received cloud desktop message queue in the preset hadoop, and the collection of multiple cloud desktop message queues in hadoop is also the cloud desktop message queue set.

[0072] It should be noted that the host machine uses Linux-related technologies to periodically obtain various performance indicator data of the host machine. In this embodiment, the various performance indicator data of the host machine include: host CPU usage, host memory occupancy, host disk IO speed and host network rate; after the host machine completes the collection of various performance indicator data each time, it further uses the form of a message queue to send the various performance indicator data to the server. In order to construct the message queue, it is also necessary to obtain the host IP address, host MAC address and host timestamp of the host machine; taking the various performance indicator data of the host machine obtained once at a time as an example, the obtained host CPU usage, host memory occupancy, host disk IO speed and host network rate are combined with the obtained host IP address, host MAC address and host timestamp to obtain the corresponding host message queue, and the host sends the host message queue to the server, so that the server receives and stores the host message queue, and the server stores the received host message queue in a preset Hadoop, and the collection of multiple host message queues in Hadoop is also the host message queue set.

[0073] S220 , filtering the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0074] S230, based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, respectively obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set.

[0075] S240: Schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0076] Embodiment 3

[0077] A cloud desktop load balancing method is provided in the third embodiment of the present application. The method refines the "data processing of the cloud desktop message queue set and the host machine message queue set within a preset period to obtain the cloud desktop time-limited performance data set and the host machine time-limited performance data set" in the second embodiment; it should be noted that for the part not described in detail in this embodiment, reference can be made to the description of other embodiments. The method includes:

[0078] S311, obtaining the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, and cloud desktop network rate in each of the cloud desktop message queues within a preset period, and obtaining a cloud desktop deadline performance data set.

[0079] Among them, taking a cloud desktop message queue in the cloud desktop message queue set as an example, by parsing the cloud desktop message queue, the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed and cloud desktop network rate in the cloud desktop message queue can be obtained; and the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed and cloud desktop network rate obtained by parsing each cloud desktop message queue are recorded as the cloud desktop deadline performance data set.

[0080] Specifically, the cloud desktop message queues in the cloud desktop message queue set are parsed to obtain the cloud desktop CPU usage, cloud desktop memory occupancy, cloud desktop disk IO speed and cloud desktop network rate therein, and the cloud desktop CPU usage, cloud desktop memory occupancy, cloud desktop disk IO speed and cloud desktop network rate obtained by parsing each cloud desktop message queue are used as the cloud desktop deadline performance data set.

[0081] S3112, obtaining the host CPU usage rate, host memory occupancy rate, host disk IO speed, and host network rate in each host message queue within a preset period, and obtaining a host time-limited performance data set.

[0082] Among them, taking a host message queue in the host message queue set as an example, by parsing the host message queue, the host CPU usage, host memory occupancy, host disk IO speed and host network rate in the host message queue can be obtained; and the host CPU usage, host memory occupancy, host disk IO speed and host network rate obtained by parsing each host message queue are recorded as the cloud desktop deadline performance data set.

[0083] It should be noted that, in this embodiment, the host CPU usage rate specifically includes: user-state CPU resource consumption data, kernel-state CPU resource consumption data, and idle CPU resource data; the host memory occupancy rate specifically includes: total memory data MemTotal, available memory resource data MemAvailable, and unused memory resource data MemFree; this embodiment uses the eBPF (extended Berkeley Packet Filter) technology to collect the host network rate. Specifically, first write a corresponding eBPF program, use the LLVM compiler to compile the eBPF program into BPF bytecode, load the BPF bytecode into the network kernel of the host, and collect the host network rate through the maps provided by eBPF; the host disk IO speed mainly includes: the number of KB read per second, the number of KB written per second, and the average waiting time of each device I / O operation, in ms;

[0084] Specifically, the host message queues in the host message queue set are parsed to obtain the host CPU usage, host memory occupancy, host disk IO speed and host network rate therein, and the host CPU usage, host memory occupancy, host disk IO speed and host network rate obtained by parsing each host message queue are used as the host deadline performance data set.

[0085] S320: Filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0086] S330, based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, respectively obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set.

[0087] S340: Schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0088] Embodiment 4

[0089] A cloud desktop load balancing method is provided in a fourth embodiment of the present application. The method refines the "based on the cloud desktop filtered performance data set, the host machine filtered performance data set, the preset combined prediction model, and the classification rules, respectively obtaining a cloud desktop performance indicator classification set and a host machine performance indicator classification set" in the first embodiment; it should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments. The method includes:

[0090] S410: Process the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set.

[0091] S420: Filter the cloud desktop time-limited performance data set and the host machine time-limited performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0092] S431. Obtain a cloud desktop performance indicator classification set based on the cloud desktop filtered performance data set, a preset combined prediction model, and classification rules.

[0093] Among them, the preset combined prediction model can process the cloud desktop filtered performance data set to obtain performance indicator prediction data that corresponds one-to-one to each performance indicator data in the cloud desktop filtered performance data set; taking one of the performance indicator prediction data as an example, the classification rule can classify the performance indicator prediction data to determine the busyness classification of the performance indicator prediction data; the cloud desktop performance indicator classification set is a set of busyness classifications that correspond one-to-one to various performance indicator prediction data corresponding to the cloud desktop filtered performance data set.

[0094] S432: Based on the host machine filtered performance data set, a preset combined prediction model, and classification rules, a host machine performance indicator classification set is obtained.

[0095] Among them, the preset combined prediction model can process the host machine post-filter performance data set to obtain performance indicator prediction data corresponding one-to-one to each performance indicator data in the host machine post-filter performance data set; taking one of the performance indicator prediction data as an example, the classification rule can classify the performance indicator prediction data to determine the busyness classification of the performance indicator prediction data; the host machine performance indicator classification set is a set of busyness classifications corresponding one-to-one to various performance indicator prediction data corresponding to the host machine post-filter performance data set.

[0096] S440: Schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0097] Embodiment 5

[0098] A cloud desktop load balancing method is provided in the fifth embodiment of the present application. The method refines the "obtaining a cloud desktop performance indicator classification set based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rule" in the fourth embodiment; it should be noted that for the part not described in detail in this embodiment, reference can be made to the description of other embodiments. The method includes:

[0099] S510: Process the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set.

[0100] S520: Filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0101] S5311. Perform mean calculation on each performance indicator data set in the cloud desktop filtered performance data set to obtain cloud desktop target performance data.

[0102] Among them, the cloud desktop filtered performance data set includes multiple performance indicator data sets, and each performance indicator data set includes multiple performance indicator data; exemplarily, the cloud desktop filtered performance data set includes: cloud desktop CPU usage data set, cloud desktop memory occupancy data set, cloud desktop disk IO speed data set and cloud desktop network rate data set. Taking the cloud desktop CPU usage data set as an example, the cloud desktop CPU usage data set includes cloud desktop CPU usage data collected at multiple different times. Mean operation is to calculate the mean of multiple performance indicator data in the performance indicator data set. The cloud desktop target performance data is a collection of means corresponding to various performance indicator data sets.

[0103] Specifically, taking a performance indicator data set in the cloud desktop filtered performance data set as an example, the performance indicator data set is obtained, and then the mean of multiple performance indicator data in the performance indicator data set is calculated; the mean corresponding to each performance indicator data set in the cloud desktop filtered performance data set is calculated in the above manner, and the mean values ​​are combined to obtain the cloud desktop target performance data.

[0104] S5312: Process the cloud desktop target performance data based on a preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets.

[0105] Among them, the preset combined prediction model is used to process the mean values corresponding to the respective performance index data sets in the cloud desktop target performance data, and performance index prediction values corresponding one by one to the performance index data sets can be obtained.

[0106] Specifically, the mean values corresponding to the respective performance index data sets in the cloud desktop target performance data are respectively input into the preset combined prediction model for processing, so as to obtain performance index prediction values corresponding one by one to the respective performance index data sets.

[0107] S5313. Process each of the performance index prediction values based on a classification rule to obtain a cloud desktop performance index classification set.

[0108] Among them, the classification rule is used to process the performance index prediction value to determine the cloud desktop performance index classification corresponding to the performance index prediction value; by way of example, taking the performance index prediction value for the performance index data set of the cloud desktop CPU usage rate data set as an example, denote this performance index prediction value as P, and the classification rule for this performance index prediction value is: if 0 < P <= 30%, then classify the cloud desktop CPU usage rate corresponding to this performance index prediction value as idle, if 30 < P <= 70%, then classify the cloud desktop CPU usage rate corresponding to this performance index prediction value as medium, if 70 < P <= 100%, then classify the cloud desktop CPU usage rate corresponding to this performance index prediction value as busy. The cloud desktop performance index classification set is a set of cloud desktop performance index classifications corresponding to various performance index prediction values.

[0109] S532. Obtain a host performance index classification set based on the filtered host performance data set, the preset combined prediction model, and the classification rule.

[0110] S540. Schedule the cloud desktops corresponding to the cloud desktop performance index classification set to the hosts corresponding to the host performance index classification set that matches the cloud desktop performance index classification set.

[0111] Embodiment Six

[0112] A cloud desktop load balancing method provided in Embodiment Six of the present application refines the "processing the cloud desktop target performance data based on a preset combined prediction model to obtain performance index prediction values corresponding one by one to the respective performance index data sets" in Embodiment Five; it should be noted that for parts not detailed in this embodiment, reference can be made to the descriptions of other embodiments. This method includes:

[0113] S610. Perform data processing on the cloud desktop message queue set and the host message queue set within a preset time limit to obtain a cloud desktop time-limited performance data set and a host time-limited performance data set.

[0114] S620: Filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set.

[0115] S6311. Perform mean calculation on each performance indicator data set in the cloud desktop filtered performance data set to obtain cloud desktop target performance data.

[0116] S63121. Input the mean value of the performance indicator data in the cloud desktop target performance data into the autoregressive model in the combined prediction model for processing to obtain a first prediction value.

[0117] It should be noted that the combined prediction model includes an autoregressive model, a weighted moving average model and an exponential smoothing model; the autoregressive model is preset with a first model weight, the weighted moving average model is preset with a second model weight, and the exponential smoothing model is preset with a third model weight.

[0118] Among them, the combined prediction model in this embodiment is a combined model generated by a set of three models: an autoregressive model, a weighted moving average model, and an exponential smoothing model; taking a performance indicator data mean as an example, the combined prediction model needs to process the performance indicator data mean separately and obtain prediction values ​​corresponding to the performance indicator data mean respectively; wherein, the first prediction value is the prediction value obtained after the autoregressive model processes the performance indicator data mean; in order to obtain a more accurate prediction value, it is necessary to perform weighted calculation on the three prediction values ​​output by the three models, and for this purpose, it is necessary to preset corresponding model weights for the three models, and the model weight of the autoregressive model is recorded as the first model weight, the model weight of the weighted moving average model is recorded as the second model weight, and the model weight of the exponential smoothing model is recorded as the third model weight.

[0119] Specifically, taking a mean value of performance indicator data in the cloud desktop target performance data as an example, the mean value of the performance indicator data is input into the autoregressive model for processing to obtain a first prediction value.

[0120] S63122. Input the mean value of the performance indicator data in the cloud desktop target performance data into the weighted moving average model in the combined prediction model for processing to obtain a second prediction value.

[0121] The second prediction value is a prediction value obtained by processing the mean of the performance indicator data using a weighted moving average model.

[0122] Specifically, taking a performance indicator data mean value in the cloud desktop target performance data as an example, the performance indicator data mean value is input into a weighted moving average model for processing to obtain a second prediction value.

[0123] S63123. Input the mean value of the performance indicator data in the cloud desktop target performance data into the exponential smoothing model in the combined prediction model for processing to obtain a third prediction value.

[0124] The third prediction value is the prediction value obtained after the exponential smoothing model processes the mean of the performance indicator data.

[0125] Specifically, taking a mean value of performance indicator data in the cloud desktop target performance data as an example, the mean value of the performance indicator data is input into the exponential smoothing model for processing to obtain a third prediction value.

[0126] S63124. Calculate the performance indicator prediction values ​​corresponding to the means of each performance indicator data based on the first prediction value, the second prediction value, the third prediction value, the first model weight, the second model weight and the third model weight.

[0127] Among them, the product of the first prediction value and the first model weight is recorded as the first product, the product of the second prediction value and the second model weight is recorded as the second product, and the product of the third prediction value and the third model weight is recorded as the third product; the performance indicator prediction value is the sum of the first product, the second product and the third product.

[0128] Specifically, taking the mean value of a performance indicator data in the cloud desktop target performance data as an example, the first product, the second product and the third product corresponding to the mean value of the performance indicator data are calculated, and then the sum of the first product, the second product and the third product is calculated as the performance indicator prediction value corresponding to the mean value of the performance indicator data.

[0129] S6313. Process the predicted values ​​of the performance indicators based on the classification rules to obtain a cloud desktop performance indicator classification set.

[0130] S632: Based on the host machine filtered performance data set, a preset combined prediction model, and classification rules, obtain a host machine performance indicator classification set.

[0131] S640: Schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0132] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0133] Embodiment 7

[0134] Based on the same inventive concept, this embodiment also provides a cloud desktop load balancing device for implementing the cloud desktop load balancing method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more cloud desktop load balancing device embodiments provided below can refer to the limitations of the cloud desktop load balancing method above, and will not be repeated here.

[0135] In this embodiment, Figure 2 As shown, a cloud desktop load balancing device is provided, comprising:

[0136] A data processing module, used for processing the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set;

[0137] A data filtering module, used to filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set;

[0138] A model processing module, used to obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules;

[0139] The desktop scheduling module is used to schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

[0140] Each module in the above-mentioned cloud desktop load balancing device can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0141] It should be noted that, in this embodiment, a cloud desktop deadline performance data set and a host deadline performance data set are obtained by processing the cloud desktop message queue set and the host machine message queue set within a preset period; the cloud desktop deadline performance data set and the host machine deadline performance data set are filtered respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; a cloud desktop performance indicator classification set and a host machine performance indicator classification set are obtained respectively based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and a classification rule; the cloud desktop corresponding to the cloud desktop performance indicator classification set is scheduled to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set; through the above implementation, the calculated cloud desktop performance indicator classification set and the host machine performance indicator classification set can respectively understand the busyness of each performance indicator of each cloud desktop and the host machine, so that the host machine matching each cloud desktop can be determined, and then the cloud desktop can be scheduled to the corresponding host machine, so that the load balancing of the cloud desktop can be achieved, and the phenomenon of cloud desktop resource competition is eliminated, which is convenient for improving the user experience of using the cloud desktop.

[0142] In one embodiment, in terms of performing data processing on a cloud desktop message queue set and a host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set, the data processing module is specifically used to:

[0143] Obtain the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, and cloud desktop network rate in each of the cloud desktop message queues within a preset period to obtain a cloud desktop deadline performance data set;

[0144] The host CPU usage rate, host memory occupancy rate, host disk IO speed, and host network rate in each host message queue within a preset period are obtained to obtain a host time-limited performance data set.

[0145] In one embodiment, in terms of obtaining a cloud desktop performance indicator classification set and a host machine performance indicator classification set based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and a classification rule, the model processing module is specifically used to:

[0146] Based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rules, a cloud desktop performance indicator classification set is obtained;

[0147] Based on the host machine filtered performance data set, a preset combined prediction model, and classification rules, a host machine performance indicator classification set is obtained.

[0148] In one embodiment, in terms of obtaining a cloud desktop performance indicator classification set based on the cloud desktop filtered performance data set, a preset combined prediction model, and a classification rule, the model processing module is specifically used to:

[0149] Perform mean calculation on each performance indicator data set in the cloud desktop filtered performance data set to obtain the cloud desktop target performance data;

[0150] Processing the cloud desktop target performance data based on a preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets;

[0151] The predicted values ​​of the performance indicators are processed based on the classification rules to obtain a cloud desktop performance indicator classification set.

[0152] In one embodiment, in processing the cloud desktop target performance data based on a preset combined prediction model to obtain a one-to-one corresponding performance indicator prediction value for each performance indicator data set, the model processing module is specifically used to:

[0153] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the autoregressive model in the combined prediction model for processing to obtain a first prediction value;

[0154] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into a weighted moving average model in the combined prediction model for processing to obtain a second prediction value;

[0155] Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the exponential smoothing model in the combined prediction model for processing to obtain a third prediction value;

[0156] Based on the first prediction value, the second prediction value, the third prediction value, the first model weight, the second model weight and the third model weight, the performance indicator prediction value corresponding to each performance indicator data mean is calculated.

[0157] Embodiment 8

[0158] In this embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a cloud desktop load balancing method is implemented.

[0159] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present disclosure, and does not constitute a limitation on the computer device to which the scheme of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0160] Embodiment 9

[0161] In this embodiment, a computer readable storage medium is provided. Figure 4 As shown, a computer program is stored thereon, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0162] Embodiment 10

[0163] In this embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0164] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0165] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present disclosure may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided by the present disclosure may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited thereto.

[0166] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this specification.

[0167] The above-described embodiments only express several implementation methods of the present disclosure, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present disclosure, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the attached claims.

Claims

1. A cloud desktop load balancing method, characterized in that: include: Processing the cloud desktop message queue set and the host machine message queue set within a preset period of time to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set; The cloud desktop deadline performance data set and the host machine deadline performance data set are filtered respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; Based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules, a cloud desktop performance indicator classification set and a host machine performance indicator classification set are obtained respectively; The cloud desktop corresponding to the cloud desktop performance indicator classification set is scheduled to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

2. The method according to claim 1, characterized in that The cloud desktop message queue in the cloud desktop message queue set includes the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, cloud desktop network rate, cloud desktop IP address, cloud desktop MAC address and cloud desktop timestamp; the host message queue in the host message queue set includes the host CPU usage rate, host memory occupancy rate, host disk IO speed, host network rate, host IP address, host MAC address and host timestamp.

3. The method according to claim 2, characterized in that The data processing of the cloud desktop message queue set and the host machine message queue set within the preset period to obtain the cloud desktop deadline performance data set and the host machine deadline performance data set includes: Obtain the cloud desktop CPU usage rate, cloud desktop memory occupancy rate, cloud desktop disk IO speed, and cloud desktop network rate in each of the cloud desktop message queues within a preset period to obtain a cloud desktop deadline performance data set; The host CPU usage rate, host memory occupancy rate, host disk IO speed, and host network rate in each host message queue within a preset period are obtained to obtain a host time-limited performance data set.

4. The method according to claim 1, characterized in that: The cloud desktop performance indicator classification set and the host machine performance indicator classification set are obtained based on the cloud desktop filtered performance data set, the host machine filtered performance data set, the preset combined prediction model, and the classification rules, respectively, including: Based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rules, a cloud desktop performance indicator classification set is obtained; Based on the host machine filtered performance data set, a preset combined prediction model, and classification rules, a host machine performance indicator classification set is obtained.

5. The method according to claim 4, characterized in that The cloud desktop performance indicator classification set is obtained based on the cloud desktop filtered performance data set, the preset combined prediction model, and the classification rules, including: Perform mean calculation on each performance indicator data set in the cloud desktop filtered performance data set to obtain the cloud desktop target performance data; Processing the cloud desktop target performance data based on a preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets; The predicted values ​​of the performance indicators are processed based on the classification rules to obtain a cloud desktop performance indicator classification set.

6. The method according to claim 5, characterized in that The combined prediction model includes an autoregressive model, a weighted moving average model and an exponential smoothing model; the autoregressive model is preset with a first model weight, the weighted moving average model is preset with a second model weight, and the exponential smoothing model is preset with a third model weight; The processing of the cloud desktop target performance data based on the preset combined prediction model to obtain performance indicator prediction values ​​corresponding to each of the performance indicator data sets includes: Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the autoregressive model in the combined prediction model for processing to obtain a first prediction value; Inputting the mean value of the performance indicator data in the cloud desktop target performance data into a weighted moving average model in the combined prediction model for processing to obtain a second prediction value; Inputting the mean value of the performance indicator data in the cloud desktop target performance data into the exponential smoothing model in the combined prediction model for processing to obtain a third prediction value; Based on the first prediction value, the second prediction value, the third prediction value, the first model weight, the second model weight and the third model weight, the performance indicator prediction value corresponding to each performance indicator data mean is calculated.

7. A cloud desktop load balancing device, characterized in that: The device comprises: A data processing module, used for processing the cloud desktop message queue set and the host machine message queue set within a preset period to obtain a cloud desktop deadline performance data set and a host machine deadline performance data set; A data filtering module, used to filter the cloud desktop deadline performance data set and the host machine deadline performance data set respectively based on a preset filtering period to obtain a cloud desktop filtered performance data set and a host machine filtered performance data set; A model processing module, used to obtain a cloud desktop performance indicator classification set and a host machine performance indicator classification set based on the cloud desktop filtered performance data set, the host machine filtered performance data set, a preset combined prediction model, and classification rules; The desktop scheduling module is used to schedule the cloud desktop corresponding to the cloud desktop performance indicator classification set to the host machine corresponding to the host machine performance indicator classification set that matches the cloud desktop performance indicator classification set.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images