CN109491788B - A method and device for implementing load balancing on a virtualization platform - Google Patents

Abstract

The invention discloses a method for realizing load balancing of a virtualization platform, which includes: obtaining the CPU usage rate corresponding to each virtual machine in the virtualization platform; respectively judging whether the CPU usage rate corresponding to each virtual machine exceeds A preset first threshold; if so, determine the available CPU whose CPU usage in the virtualization platform is lower than the preset second threshold, and determine the process of the virtual machine exceeding the first threshold as a process to be adjusted; The adjustment process is bound to the available CPU affinity, so that the available CPU can process the process to be adjusted, and realize the load balancing of the virtualization platform. In this way, the load balancing of the virtualization platform is realized, the performance of the currently busy virtual machine is improved, and unexpected situations such as downtime can be avoided. Correspondingly, a virtualization platform load balancing implementation device, device and readable storage medium disclosed in the present invention also have the above-mentioned technical effects.

Description

A method and device for implementing load balancing on a virtualization platform

technical field

The present invention relates to the technical field of virtualization, and more specifically, to a method, device, device and readable storage medium for realizing load balancing of a virtualization platform.

Background technique

In a virtualization platform, each virtual machine is fixedly bound to one or more CPUs to process its business processes, that is, each CPU corresponds to a business process, generally a qemu-kvm process. The virtual machine corresponds to several CPUs and corresponds to several qemu-kvm processes, which can avoid performance degradation caused by frequent switching of processes on different CPUs.

However, when the CPU usage of a virtual machine in the virtualization platform continues to be high, it indicates that the qemu-kvm process in each CPU is relatively busy. If this situation lasts for too long, downtime is likely to occur Wait for the unexpected. It should be noted that since each virtual machine is fixedly bound to a CPU, when the CPU usage of a certain virtual machine continues to be high, the CPUs corresponding to other virtual machines may be in an idle state. It can be seen that in this case, the load of the virtualization platform cannot be kept balanced. If this problem is not solved, the performance of the busy virtual machine will be degraded, and CPU resources in the virtualization platform will also be wasted.

Therefore, how to realize the load balancing of the virtualization platform and improve the performance of the virtual machine is a problem to be solved by those skilled in the art.

Contents of the invention

The object of the present invention is to provide a method, device, device and readable storage medium for implementing load balancing of a virtualization platform, so as to realize load balancing of a virtualization platform and improve the performance of a virtual machine.

In order to achieve the above object, the embodiment of the present invention provides the following technical solutions:

A method for implementing load balancing on a virtualization platform, comprising:

Obtain the CPU usage corresponding to each virtual machine in the virtualization platform;

Determine whether the CPU usage corresponding to each virtual machine exceeds a preset first threshold within a preset time period;

If so, then determine the available CPU whose CPU usage rate in the virtualization platform is lower than the preset second threshold, and determine the process of the virtual machine exceeding the first threshold as the process to be adjusted; Affinity binding is performed between a process and the available CPU, so that the available CPU can process the process to be adjusted, so as to realize load balancing of the virtualization platform.

Wherein, said obtaining the CPU utilization rate corresponding to each virtual machine in the virtualization platform includes:

The CPU utilization rate corresponding to each virtual machine in the virtualization platform is acquired at a preset interval.

Wherein, after the available CPU processes the process to be adjusted, it further includes:

Acquiring the CPU usage of the process to be adjusted, and judging whether the CPU usage of the process to be adjusted exceeds the first threshold within a preset period of time;

If so, then determine the available CPU, and bind the process to be adjusted with the available CPU for affinity binding, so that the available CPU can process the process to be adjusted, and realize the load balancing of the virtualization platform .

Among them, also include:

When the CPU usage rate of the process to be adjusted is lower than a preset third threshold, affinity unbinding is performed between the process to be adjusted and the available CPU.

Wherein, before obtaining the CPU utilization rate corresponding to each virtual machine in the virtualization platform, it also includes:

Monitor the CPU usage corresponding to each virtual machine in the virtualization platform.

A device for implementing load balancing on a virtualization platform, comprising:

An acquisition module, configured to acquire the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

A judging module, configured to judge whether the CPU usage corresponding to each virtual machine exceeds a preset first threshold within a preset duration;

An execution module, configured to determine an available CPU whose CPU usage in the virtualization platform is lower than a preset second threshold when the CPU usage corresponding to the virtual machine exceeds a preset first threshold within a preset duration, and determining the process of the virtual machine exceeding the first threshold as a process to be adjusted; binding the process to be adjusted with the available CPU for affinity binding, so that the available CPU can process the process to be adjusted, Realize load balancing of the virtualization platform.

Wherein, described acquisition mold body is used for:

The CPU utilization rate corresponding to each virtual machine in the virtualization platform is acquired at a preset interval.

A device for implementing load balancing on a virtualization platform, comprising:

memory for storing computer programs;

The processor is configured to implement the steps of any one of the methods for implementing load balancing on a virtualization platform described above when executing the computer program.

A readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the methods for realizing load balancing of a virtualization platform described above are implemented.

It can be seen from the above solutions that a method for realizing load balancing of a virtualization platform provided by an embodiment of the present invention includes: obtaining the CPU usage rate corresponding to each virtual machine in the virtualization platform; and separately determining the CPU usage rate corresponding to each virtual machine Whether the preset first threshold value is exceeded within the preset duration; Determine the process of the machine as the process to be adjusted; bind the process to be adjusted with the available CPU for affinity binding, so that the available CPU can process the process to be adjusted, and realize the load balancing of the virtualization platform.

It can be seen that, in the method, when the CPU usage corresponding to virtualization in the virtualization platform exceeds the preset first threshold within a preset period of time, the process of the virtual machine exceeding the first threshold is determined as a process to be adjusted, and Binding the process to be adjusted with the determined available CPU for affinity binding, so that the available CPU processes the process to be adjusted, thereby realizing load balancing of the virtualization platform, improving the performance of the currently busy virtual machine, and avoiding The occurrence of unforeseen circumstances such as downtime.

Correspondingly, the virtualization platform load balancing implementation device, device, and readable storage medium provided by the embodiments of the present invention also have the above-mentioned technical effects.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

FIG. 1 is a flow chart of a method for realizing load balancing of a virtualization platform disclosed by an embodiment of the present invention;

FIG. 2 is a flow chart of another method for realizing load balancing of a virtualization platform disclosed by an embodiment of the present invention;

FIG. 3 is a schematic diagram of a virtualization platform load balancing implementation device disclosed in an embodiment of the present invention;

FIG. 4 is a schematic diagram of a device for realizing load balancing of a virtualization platform disclosed by an embodiment of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The embodiment of the present invention discloses a virtualization platform load balance realization method, device, equipment and readable storage medium, so as to realize the load balance of the virtualization platform and improve the performance of the virtual machine.

Referring to FIG. 1, a method for realizing load balancing of a virtualization platform provided by an embodiment of the present invention includes:

S101. Obtain the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

Specifically, there are multiple virtual machines in the virtualization platform, and each virtualization corresponds to a CPU usage rate. Therefore, the CPU usage rate obtained in this embodiment has a plurality of data, and these data are in one-to-one correspondence with each virtual machine.

S102. Determine whether the CPU usage rate corresponding to each virtual machine exceeds a preset first threshold within a preset duration; if yes, execute S103; if not, execute S101;

S103. Determine the available CPUs in the virtualization platform whose CPU utilization rate is lower than the preset second threshold, and determine the process of the virtual machine exceeding the first threshold as the process to be adjusted, and execute S104;

S104. Bind the processes to be adjusted with the available CPUs for affinity binding, so that the available CPUs process the processes to be adjusted, and realize load balancing of the virtualization platform.

It should be noted that since there are multiple CPU usage rates, it is necessary to compare the CPU usage rates with the preset first threshold one by one, so as to determine the corresponding virtual machine process. For example: when there are 5 virtual machines in the virtualization platform, they are A, B, C, D, and E, and the obtained CPU usage rates are 60%, 60%, 80%, 87%, and 90%, assuming The preset first threshold is 85%, then the CPU usage rates exceeding the first threshold are 87% and 90%, and the corresponding virtual machines are D and E, so D and E are the virtual machines exceeding the first threshold. The corresponding processes are the processes to be adjusted. It can be seen that the number of processes to be adjusted is not fixed.

Similarly, assuming that the preset second threshold is 70%, the available CPUs are the CPUs corresponding to the virtual machines A and B respectively. It can be seen that the number of available CPUs is not fixed. When doing affinity binding, you can select an appropriate available CPU for the process to be adjusted according to actual needs, so this specification does not make specific limitations here.

It can be seen that this embodiment provides a method for implementing load balancing on a virtualization platform. In the method, when the CPU usage corresponding to virtualization in the virtualization platform exceeds a preset first threshold within a preset period of time, it will exceed The process of the virtual machine with the first threshold is determined as the process to be adjusted, and the affinity binding of the process to be adjusted and the determined available CPU is made, so that the available CPU can process the process to be adjusted, thereby realizing the virtualization platform Load balancing improves the performance of currently busy virtual machines and avoids accidents such as downtime.

The embodiment of the present invention discloses another method for realizing load balancing of a virtualization platform. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution.

Referring to Fig. 2, another method for realizing load balancing of a virtualization platform provided by an embodiment of the present invention includes:

S201. Obtain the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

S202. Determine whether the CPU usage rate corresponding to each virtual machine exceeds a preset first threshold within a preset duration; if yes, execute S203; if not, execute S202;

S203. Determine the available CPUs in the virtualization platform whose CPU utilization rate is lower than the preset second threshold, and determine the process of the virtual machine exceeding the first threshold as the process to be adjusted, and execute S204;

S204. Affinity-binding the process to be adjusted with the available CPU, so that the available CPU processes the process to be adjusted;

S205. Obtain and process the CPU usage rate of the process to be adjusted, and determine whether the CPU usage rate of the process to be adjusted exceeds the first threshold within the preset duration; if yes, execute S206; if not, execute S202;

S206. Determine the available CPUs in the virtualization platform whose CPU utilization rate is lower than the preset second threshold, and bind the processes to be adjusted with the available CPUs, so that the available CPUs can process the processes to be adjusted to realize virtualization Platform load balancing.

It should be noted that after the first affinity binding is completed, the CPU usage of the process to be adjusted can be detected again. If it still exceeds the first threshold, an appropriate available CPU can be selected for affinity for the process to be adjusted multiple times. performance binding until the CPU usage of the process to be adjusted is lower than the preset first threshold.

It can be seen that this embodiment provides another method for implementing load balancing on a virtualization platform. In the method, when the CPU usage corresponding to virtualization in the virtualization platform exceeds a preset first threshold within a preset period of time, the The process of the virtual machine exceeding the first threshold is determined as the process to be adjusted, and the process to be adjusted is bound to the determined available CPU affinity, and can be adjusted multiple times when necessary, so that the available CPU can process the process to be adjusted , so that the load balancing of the virtualization platform is realized, the performance of the currently busy virtual machine is improved, and unexpected situations such as downtime can be avoided.

Based on any of the above embodiments, it should be noted that the obtaining the CPU usage rate corresponding to each virtual machine in the virtualization platform includes: obtaining the corresponding CPU usage rate of each virtual machine in the virtualization platform at a preset interval. CPU usage.

Specifically, in order to determine the virtual machines whose CPU usage is consistently high, the CPU usage can be obtained at preset intervals; of course, the CPU usage corresponding to each virtual machine in the virtualization platform can also be monitored in real time.

Based on any of the above embodiments, it should be noted that the method further includes: when the CPU usage of the process to be adjusted is lower than a preset third threshold, performing affinity resolution between the process to be adjusted and the available CPU tie. That is: when the process to be adjusted is not busy, affinity unbinding is performed to restore the original CPU binding relationship of the virtualization platform.

The following describes the specific implementation process according to the core idea of the present invention, which mainly includes: monitoring and adjusting each thread. The monitoring thread mainly monitors the CPU usage of each virtual machine, regularly scans the CPU usage, and sends information to the adjustment thread when the usage reaches a certain threshold, and the adjustment thread adjusts the process according to the information.

Specifically, the CPU corresponding to each virtual machine and the usage rate of the CPU are queried through the virtual machine platform management software, and the process ID corresponding to each CPU is queried and saved. Because the CPU usage rate changes very frequently, it is likely that the CPU usage rate is high at a certain moment due to sudden tasks inside the virtual machine. It is meaningless to deal with the CPU usage rate being too high in a short period of time. Only make adjustments when usage is persistently high. For example: the CPU usage rate is detected at a frequency per minute, and adjustments are made after the detected values exceed the threshold for two consecutive times.

When two consecutive detections show that the process CPU usage of a certain virtual machine exceeds the threshold, the monitoring thread will send a message to the adjustment thread. The message content includes the name of the virtual machine, the number of CPUs, the corresponding process ID, and the CPU usage.

After the adjustment thread receives the message from the monitoring thread, check whether the CPU usage of the corresponding process in the message still exceeds the preset threshold. If so, query the utilization rate of all CPUs in the platform, select the CPU with the lowest utilization rate, and bind the CPU affinity with the process ID in the message, so that the process can run on the CPU with the lowest utilization rate, so as to improve the performance of the virtual machine. performance. If not, the adjustment is discarded.

If after the adjustment is made, the CPU usage rate of the virtual machine is still high, the adjustment thread will perform a second adjustment, specifically: increase the number of CPUs of the virtual machine.

A virtualization platform load balancing realization device provided by an embodiment of the present invention is introduced below. The virtualization platform load balancing realization device described below and the virtualization platform load balancing realization method described above may refer to each other.

Referring to FIG. 3 , a virtualization platform load balancing implementation device provided by an embodiment of the present invention includes:

An acquisition module 301, configured to acquire the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

A judging module 302, configured to judge whether the CPU usage corresponding to each virtual machine exceeds a preset first threshold within a preset duration;

An execution module 303, configured to determine available CPUs in the virtualization platform whose CPU usage is lower than a preset second threshold when the CPU usage corresponding to the virtual machine exceeds a preset first threshold within a preset duration , and determine the process of the virtual machine exceeding the first threshold as the process to be adjusted; bind the process to be adjusted with the available CPU affinity, so that the available CPU can process the process to be adjusted , realizing load balancing of the virtualization platform.

Wherein, described acquisition mold body is used for:

The CPU utilization rate corresponding to each virtual machine in the virtualization platform is acquired at a preset interval.

Among them, also include:

An adjustment module, configured to obtain and process the CPU usage rate of the process to be adjusted, and determine whether the CPU usage rate of the process to be adjusted exceeds the first threshold within a preset duration; if so, determine the available CPU, And binding the process to be adjusted with the available CPU through affinity binding, so that the available CPU can process the process to be adjusted, so as to realize load balancing of the virtualization platform.

Among them, also include:

An unbinding module, configured to unbind the process to be adjusted and the available CPU with affinity when the CPU usage of the process to be adjusted is lower than a preset third threshold.

Among them, also include:

A monitoring module, configured to monitor the CPU usage corresponding to each virtual machine in the virtualization platform.

It can be seen that this embodiment provides a virtualization platform load balancing implementation device, including: an acquisition module, a judgment module, and an execution module. First, the acquisition module obtains the CPU utilization rate corresponding to each virtual machine in the virtualization platform; then the judging module judges whether the CPU utilization rate corresponding to each virtual machine exceeds a preset first threshold within a preset duration; when the virtual When the CPU utilization rate corresponding to the machine exceeds the preset first threshold within the preset time period, the execution module determines that the CPU usage rate of the virtualization platform is lower than the preset second threshold of the available CPU, and will exceed the first threshold The process of the virtual machine is determined as the process to be adjusted; the process to be adjusted is bound to the available CPU affinity, so that the available CPU can process the process to be adjusted, and the load balancing of the virtualization platform is realized. In this way, the division of labor and cooperation among the various modules, each performing its own duties, thus realizes the load balancing of the virtualization platform, improves the performance of the current busy virtual machine, and can avoid unexpected situations such as downtime.

The following is an introduction to a virtualization platform load balancing implementation device provided by the embodiment of the present invention. The virtualization platform load balancing implementation device described below and the virtualization platform load balancing implementation method and device described above can be cross-referenced. .

Referring to FIG. 4, a virtualization platform load balancing implementation device provided by an embodiment of the present invention includes:

Memory 401, used to store computer programs;

The processor 402 is configured to implement the steps of the method for implementing load balancing on a virtualization platform described in any of the foregoing embodiments when executing the computer program.

A readable storage medium provided by an embodiment of the present invention is introduced below. The readable storage medium described below and the method, device, and device for implementing load balancing on a virtualization platform described above may be referred to together.

A readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for realizing load balancing of a virtualization platform as described in any of the above embodiments are implemented.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (4)

1. A method for realizing load balancing of a virtualization platform is characterized by comprising the following steps:

acquiring the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

respectively judging whether the CPU utilization rate corresponding to each virtual machine exceeds a preset first threshold value within a preset time length;

if yes, determining an available CPU of which the CPU utilization rate in the virtualization platform is lower than a preset second threshold value, and determining the process of the virtual machine exceeding the first threshold value as a process to be adjusted; performing affinity binding on the process to be adjusted and the available CPU so that the available CPU processes the process to be adjusted to realize load balancing of the virtualization platform;

after the first affinity binding is completed, the CPU utilization rate of the process to be adjusted can be detected again, and if the CPU utilization rate still exceeds a first threshold value, available CPUs are selected for the process to be adjusted for multiple times to be subjected to affinity binding or the number of CPUs of the virtual machines exceeding the first threshold value is increased;

the obtaining of the CPU utilization rate corresponding to each virtual machine in the virtualization platform includes:

acquiring the CPU utilization rate corresponding to each virtual machine in the virtualization platform according to preset interval time;

wherein, still include:

when the CPU utilization rate of the process to be adjusted is lower than a preset third threshold value, performing affinity unbinding on the process to be adjusted and the available CPU;

before obtaining the CPU utilization rate corresponding to each virtual machine in the virtualization platform, the method further includes:

and monitoring the CPU utilization rate corresponding to each virtual machine in the virtualization platform.

2. A virtualization platform load balancing implementation device is characterized by comprising:

the acquisition module is used for acquiring the CPU utilization rate corresponding to each virtual machine in the virtualization platform;

the judging module is used for respectively judging whether the CPU utilization rate corresponding to each virtual machine exceeds a preset first threshold value within a preset time length;

the execution module is used for determining an available CPU of which the CPU utilization rate in the virtualization platform is lower than a preset second threshold value when the CPU utilization rate corresponding to the virtual machine exceeds a preset first threshold value within a preset time length, and determining the process of the virtual machine exceeding the first threshold value as a process to be adjusted; performing affinity binding on the process to be adjusted and the available CPU so that the available CPU processes the process to be adjusted to realize load balancing of the virtualization platform;

after the first affinity binding is finished, the CPU utilization rate of the process to be adjusted can be detected again, and if the CPU utilization rate still exceeds the first threshold value, the available CPU is selected for the process to be adjusted for multiple times to be subjected to affinity binding or the number of CPUs of the virtual machines exceeding the first threshold value is increased;

wherein the obtaining module is specifically configured to:

acquiring the CPU utilization rate corresponding to each virtual machine in the virtualization platform according to preset interval time;

wherein, still include:

the unbinding module is used for performing affinity unbinding on the process to be adjusted and the available CPU when the CPU utilization rate of the process to be adjusted is lower than a preset third threshold value;

wherein, still include:

the monitoring module is used for monitoring the CPU utilization rate corresponding to each virtual machine in the virtualization platform before the CPU utilization rate corresponding to each virtual machine in the virtualization platform is obtained.

3. A virtualization platform load balancing implementation device is characterized by comprising:

a memory for storing a computer program;

a processor for implementing the steps of the virtualization platform load balancing implementation method as claimed in claim 1 when executing said computer program.

4. A readable storage medium, having stored thereon a computer program which, when executed by a processor, implements the steps of the virtualization platform load balancing implementation method as recited in claim 1.

Images