CN103593224A - Virtual machine resource allocation system and method - Google Patents

Abstract

A virtual machine resource allocation system and method, the method comprising: a data acquisition step of obtaining the usage rate of a specific resource of a virtual machine; a calculation step of calculating the average usage rate of the specific resource of the virtual machine every other analysis cycle, and Determine the resource level of the virtual machine according to the average usage rate of the specific resource of the virtual machine; the recommending step is to obtain the recommended resource specification according to the resource level of the virtual machine, and send the recommended resource specification to the client; the configuration step, when When the client accepts the recommended resource specification, it re-allocates resources for the virtual machine according to the recommended resource specification. Utilizing the present invention can provide the user with recommended resource specifications according to the change of specific resource utilization rate during the use of the virtual machine at the user end.

Description

Virtual machine resource configuration system and method

technical field

The present invention relates to a cloud computing service system and method, in particular to a virtual machine resource configuration system and method.

Background technique

Cloud services provide users with a solution to save computing costs. Users do not need to spend a lot of money to purchase hardware, as long as the virtualized application can achieve the computing purpose of multiple physical hosts. The current cloud service providers mainly include Amazon EC2, CloudShare or Citrix, etc., and they all provide different user operation interfaces and services.

When users apply for a virtual machine (Virtual Machine, VM) from a cloud service provider, they can choose resource specifications and prices according to their needs. However, when users choose virtual machine resources, they may not be able to accurately predict how many resources will be used, and choose too many or Too few resources will result in wasted costs. Although some vendors offer a plan to pay as much as resources are used, it may also cause budget overuse because users cannot grasp the usage status of virtual machines.

Contents of the invention

In view of the above, it is necessary to provide a virtual machine resource configuration system, which can provide the user with recommended resource specifications according to the changes in the utilization rate of specific resources during the use of the virtual machine at the client end, and reconfigure the virtual machine according to the recommended resource specifications. Virtual machines allocate resources.

In view of the above, it is necessary to provide a virtual machine resource configuration method, which can provide the user with a recommended resource specification according to the change of the specific resource usage rate during the use of the client virtual machine, and reconfigure the resource specification according to the recommended resource specification. Virtual machines allocate resources.

A virtual machine resource configuration system, the system includes: a data acquisition module, used to obtain the usage rate of a specific resource of a virtual machine; a calculation module, used to calculate the average usage rate of the specific resource of the virtual machine every other analysis cycle , and determine the resource level of the virtual machine according to the average usage rate of the specific resource of the virtual machine; the recommendation module is used to obtain the recommended resource specification according to the resource level of the virtual machine, and send the recommended resource specification to the client; The configuration module is configured to re-allocate resources for the virtual machine according to the recommended resource specification when the client accepts the recommended resource specification.

A method for configuring virtual machine resources, the method comprising: a data acquisition step of obtaining the usage rate of a specific resource of a virtual machine; a calculation step of calculating the average usage rate of the specific resource of the virtual machine every other analysis cycle, and according to the The average utilization rate of the specific resource of the virtual machine determines the resource level of the virtual machine; the recommendation step is to obtain the recommended resource specification according to the resource level of the virtual machine, and send the recommended resource specification to the client; the configuration step is to When the recommended resource specification is accepted, resources are re-allocated to the virtual machine according to the recommended resource specification.

Compared with the prior art, the virtual machine resource configuration system and method can monitor the resource usage status of the client virtual machine in real time, and provide recommendations to the user according to changes in the utilization rate of specific resources during the use of the client virtual machine. resource specifications, and re-allocate resources for the virtual machine according to the recommended resource specifications, so as to prevent users from applying for excessive or insufficient resources.

The present invention is suitable for application in the scenario of providing IaaS (Infrastructure as a Service), which not only saves unnecessary costs for users, but also improves the utilization rate of hardware resources of service providers.

Description of drawings

Fig. 1 is a network architecture diagram of the control server of the present invention.

FIG. 2 is a schematic diagram of the operating environment of the virtual machine resource configuration system of the present invention.

Fig. 3 is a functional block diagram of the virtual machine resource configuration system.

Fig. 4 is a flow chart of a preferred embodiment of the method for configuring virtual machine resources in the present invention.

FIG. 5 is a schematic diagram of a master/slave architecture of multiple database servers.

FIG. 6 is a schematic diagram of distributed parallel computing of multiple database servers.

Description of main component symbols

user terminal 1 control server 2 database server 3 Virtual machine server 4,5 virtual machine 41, 42, 51, 52 display screen 20 input device twenty two memory twenty three Virtual machine resource configuration system twenty four Processor 25  Data Acquisition Module 240 Computing module 241 Recommended modules 242 configuration module 243

Detailed ways

As shown in FIG. 1, it is a network architecture diagram of the control server of the present invention. In this embodiment, the control server 2 communicates with a plurality of client terminals 1 (only one is shown in the figure), a plurality of database servers 3 (only one is shown in the figure), and a plurality of virtual machine servers (only one is shown in FIG. 1 ) through a network. The virtual machine servers 4 and 5) are connected. The network may be an intranet, the Internet, or other types of communication networks, such as GPRS, Wi-Fi/WLAN, 3G/WCDMA, 3.5G/HSDPA, and the like.

Wherein, each virtual machine server is installed with a virtual machine monitoring program, which is used to monitor the utilization rate of the specific resource of the virtual machine of the client 1 every predetermined time, and send the monitored utilization rate of the specific resource of the virtual machine to Control server 2. For example, virtual machine server 4 monitors resource usage rates of virtual machines 41 and 42 , and virtual machine server 5 monitors resource usage rates of virtual machines 51 and 52 . In this embodiment, the specific resource usage includes three types, CPU usage (CPU%), memory usage (Memory%), and hard disk usage (Disk%). In other embodiments, the type of resource usage may also be increased or decreased.

The control server 2 acts as a virtual machine server center, and stores the resource utilization rate of each virtual machine in a corresponding data table according to the name of the virtual machine. For example, the resource usage rate of the virtual machine 41 is stored in the data table Table41, and the resource usage rate of the virtual machine 42 is stored in the data table Table42. Wherein, each data table includes the following fields, virtual machine name, ID, CPU usage, memory usage, hard disk usage, and storage time.

In this embodiment, the multiple database servers 3 provide distributed parallel computing functions. Referring to Fig. 5, the architecture of the plurality of database servers 3 is a master-slave architecture (Master/Slave), and a master server (Master Server) is responsible for dividing data and distributing it to different slave servers (Slave Server), The slave server is responsible for calculating the allocated data, and returns the result to the master server, and the master server then transmits the aggregated data to the control server 2 for further calculation.

As shown in FIG. 2 , it is a schematic diagram of the operating environment of the virtual machine resource configuration system of the present invention. The virtual machine resource configuration system runs on the control server 2 . The control server 2 also includes a display device 20 , an input device 22 , a memory 23 and a processor 25 connected via a data bus.

The memory 23 is used to store the program code of the virtual machine resource allocation system 24 and data tables of virtual machine resource utilization and other materials. The display device 20 is used for displaying information such as the resource usage rate of the virtual machine. The input device 22 is used to input various data set by the user, such as the name of the virtual machine and so on.

The virtual machine resource configuration system 24 is used to provide the user with recommended resource specifications according to changes in the specific resource usage rate during the use of the virtual machine at the client terminal 1, and to re-allocate resources for the virtual machine according to the recommended resource specifications, specifically The procedure is described below.

In this embodiment, the virtual machine resource configuration system 24 can be divided into one or more modules, and the one or more modules are stored in the memory 23 and configured to be processed by one or more processors (this embodiment is a processor 25) to complete the present invention. For example, referring to FIG. 3 , the virtual machine resource configuration system 24 is divided into a data acquisition module 240 , a calculation module 241 , a recommendation module 242 and a configuration module 243 . The module referred to in the present invention is a program segment that completes a specific function, and is more suitable than a program to describe the execution process of software in the control server 2 . The specific functions of each module will be described below with reference to FIG. 4 .

As shown in FIG. 4 , it is a flowchart of a preferred embodiment of the virtual machine resource configuration method of the present invention.

Step S10 , the user logs in to the virtual machine management interface through the user terminal 1 and starts a virtual machine (such as the virtual machine 41 ). It includes selecting the resource specification, price and security mechanism of the virtual machine. Assume that the resource specification initially selected by the user is (CPU, memory, hard disk) = (2.5GB, 4GB, 200GB).

In step S11, the data obtaining module 240 obtains the usage rate of the specific resource of the virtual machine detected by the virtual machine server, and stores the usage rate of the specific resource of the virtual machine in a data table. For example, the data acquisition module 240 acquires the usage rate of the specific resource of the virtual machine detected by the virtual machine server, such as the CPU usage rate, the memory usage rate and the hard disk usage rate, every minute.

In step S12, the calculation module 241 calculates the average usage rate of the specific resource of the virtual machine every other analysis period, and determines the resource level of the virtual machine according to the average usage rate of the specific resource of the virtual machine. Wherein, the resource level of the virtual machine includes an average utilization rate level of each specific resource.

Wherein, the computing module 241 uses distributed parallel computing (segmentation and merging) to calculate the average usage rate of the specific resource of the virtual machine. When the specific resource usage of a virtual machine needs to be analyzed, the calculation module 241 divides the specific resource usage of the virtual machine into different blocks (Split) according to the acquisition time sequence of the specific resource usage of the virtual machine, and Each block is assigned to different database servers 3 for parallel addition operations.

For example, as shown in FIG. 6, the calculation module 241 divides every 10 pieces of data into a block to obtain blocks split ₁ , split ₂ , ..., split _m , wherein the block split ₁ includes time points from time ₁ to time The CPU usage, memory usage, and hard disk usage of ₁₀ , block split ₂ includes the CPU usage, memory usage, and hard disk usage from time ₁₁ to time ₂₀ . Then, the calculation module 241 assigns the block split ₁ to the slave server 1 for addition operation, assigns the block split ₂ to the slave server 2 for addition operation, etc., and sends back the operation result to the control server 2 .

The calculation module 241 groups the sums calculated by each slave server, and adds up the data of the same group to obtain the group sum. Next, the calculation module 241 sums up all group sums to obtain a final summation value. Finally, the calculation module 241 divides the final total value by the number of records to obtain the average usage rate of specific resources, such as the average CPU usage rate, the average memory usage rate, and the average hard disk usage rate.

For example, as shown in Figure 6, n represents the period (that is, the number of records), i represents the time point, C _i represents the CPU usage rate at time point i, M _i represents the memory usage rate at time point i, and D _i represents time point i C' represents the average CPU usage, M' represents the average memory usage, and D' represents the average hard disk usage.

In this embodiment, three types of resources are taken as examples for illustration, and the average usage rate of each resource is divided into three levels by two thresholds. The method of determining the resource level of the virtual machine according to the average utilization rate of the specific resource of the virtual machine is as follows:

If the average usage rate of the first type of resource (such as the average CPU usage rate) is less than or equal to the first threshold (such as 20%), then determine the average usage rate of the first type of resource level one, denoted as I;

If the average usage rate of the first type of resource is greater than the first threshold and less than or equal to the second threshold (70%), then determine the average usage rate of the first type of resource as level two, denoted as II;

If the average utilization rate of the first type resource is greater than the second threshold, then determine the average utilization rate of the first type resource as level three, denoted as III.

Similarly, if the average usage rate of the second type of resource (such as the average memory usage rate) is less than or equal to the third threshold (such as 40%), then determine the average usage rate of the second type of resource level 1, denoted as A ;

If the average usage rate of the second-type resource is greater than the third threshold and less than or equal to the fourth threshold (80%), then determine the average usage rate of the second-type resource as Level 2, denoted as B;

If the average usage rate of the second type resource is greater than the fourth threshold, then determine the average usage rate of the second type resource as Level 3, denoted as C.

Similarly, if the average usage rate of the third type of resource (such as the average usage rate of the hard disk) is less than or equal to the fifth threshold (such as 30%), then determine the average usage rate of the third type of resource level 1, denoted as i ;

If the average usage rate of the third type of resource is greater than the fifth threshold and less than or equal to the sixth threshold (60%), then determine the average usage rate of the third type of resource as Level 2, denoted as ii;

If the average utilization rate of the third-type resource is greater than the sixth threshold, then determine the average utilization rate of the third-type resource as Level 3, denoted as iii.

For example, assume that Spec _CPU represents the level of average CPU usage, Spec _MEM represents the level of average memory usage, Spec _Disk represents the level of average hard disk usage, and Spec(Spec _CPU , Spec _MEM , Spec _Disk ) represents the virtual machine the resource level of the virtual machine, the algorithm for determining the resource level of the virtual machine is as follows.

If(C'≤Threshold ₁ )Spec _CPU =I;

Else if(Threshold ₁ <C'≤Threshold ₂ )Spec _CPU =II;

Else if(Threshold ₂ <C')Spec _CPU =III;

If(M'≤Threshold ₃ )Spec _MEM =A;

Else if(Threshold ₃ ＜M'≤Threshold ₄ )Spec _MEM =B;

Else if(Threshold ₄ <M')Spec _MEM =C;

If(D'≤Threshold ₅ )Spec _Disk =i;

Else if(Threshold ₅ ＜D'≤Threshold ₆ )Spec _Disk =ii;

Else if(Threshold ₆ <D')Spec _Disk =iii.

Wherein, Threshold ₁ to Threshold ₆ respectively represent the first threshold to the sixth threshold.

In this embodiment, the average usage rate of each specific resource is divided into three levels by two thresholds, wherein the average CPU usage rate is divided into three levels I, II, and III, and the average memory usage rate is divided into A , B, and C levels, and the average hard disk usage rate is divided into three levels i, ii, and iii. Therefore, 27 different virtual machine resource levels can be obtained according to the arrangement and combination. Each resource level is preset with recommended resource specifications. The lower the level, the less resources are allocated, and the higher the level, the more resources are allocated. For example, Spec(I, A, i) represents that the average usage rates of the three resources are all low, and it is recommended to use a low resource specification. Spec (III, C, iii) means that the average utilization rate of the three resources is high, so it is recommended to use a high resource specification.

In other embodiments, other different virtual machine resource levels can also be combined by increasing or decreasing the threshold, the method is similar, and will not be repeated here.

Step S13 , the recommendation module 242 obtains the recommended resource specification according to the resource level of the virtual machine, and sends the recommended resource specification to the client terminal 1 . For example, assuming that the calculated average CPU usage C'=60%, the average memory usage M'=85, and the hard disk average usage D'=20%, it can be known that:

Threshold ₁ <60%<Threshold ₂ , Spec _CPU =I;

Threshold ₄ <85%, Spec _MEM =C;

20%<Threshold ₅ , Spec _Disk = i;

Then the resource level of the virtual machine Spec(Spec _CPU , Spec _MEM , Spec _Disk )=Spec(I, C, i), assuming that the resource specification recommended by the resource level (I, C, i) is:

CPU = {500MB,...,1.75GB},

memory = {3.2GB,...,4GB},

Hard disk = {minimum capacity, ..., 60GB}.

The user can select a resource item within the above range by himself, or input a value within the above range by himself, or the recommendation module 242 recommends a fixed resource item to the user. Among them, the above modification can be completed while the virtual machine is running (Hot Plug), or the modification can be completed after the virtual machine is suspended.

In step S14, the configuration module 243 determines whether the user terminal 1 accepts the recommended resource specification. For example, if the user selects a specific resource specification (500MB, 3.2GB, 20GB) or selects a fixed resource item recommended by the recommendation module 242, it is determined that the user terminal 1 accepts the recommended resource specification, and step S15 is performed. If the user terminal 1 does not accept the recommended resource specification, the process ends.

In step S15, the configuration module 243 re-allocates resources for the virtual machine according to the recommended resource specification. For example, if the user selects (500MB, 3.2GB, 20GB) in the recommended resource specifications, the configuration module 243 re-allocates resources for the virtual machine according to the resource rules selected by the user. If the user chooses that the control server 2 recommends a fixed resource item, the configuration module 243 selects a specific resource from the recommended resource specifications randomly or in a predetermined order (such as the order of resources from small to large).

Wherein, the resource specification initially selected by the user, ie (CPU, memory, hard disk)=(2.5GB, 4GB, 200GB), will not be overwritten by the recommended resource specification and will be stored in the memory 23 .

In other embodiments, the database server 3 can also be one, and even the control server 2 and the database server 3 can be combined into one server. Further, the virtual machine server 4 can also be one, and even the control server 2 and the virtual machine server 4 can be combined into one server.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be Modifications or equivalent replacements can be made without departing from the spirit and scope of the technical solutions of the present invention.

Claims (14)

1. a resources of virtual machine configuration-system, is characterized in that, this system comprises:

Data acquisition module, for obtaining the utilization rate of a virtual machine specific resources;

Computing module, for every an analytical cycle, calculates the average service rate of this virtual machine specific resources, and according to the average service rate of this virtual machine specific resources, determines the resource level of this virtual machine;

Recommending module, for obtain the resource specification of recommendation according to the resource level of this virtual machine, and is sent to user side by the resource specification of this recommendation; And

Configuration module when accepting the resource specification of this recommendation when user side, is this virtual machine Resources allocation according to the resource specification of this recommendation again.

2. resources of virtual machine configuration-system as claimed in claim 1, is characterized in that, described computing module uses distributed parallel calculation method to calculate the average service rate of this virtual machine specific resources.

3. resources of virtual machine configuration-system as claimed in claim 2, is characterized in that, described distributed parallel calculation method is:

Acquisition time order according to this virtual machine specific resources utilization rate, is divided into different blocks by the utilization rate of this virtual machine specific resources, and each block is divided and tasks different database servers and carry out parallel addition computing;

The operation result that gathers all database servers, the operation result gathering according to this calculates the average service rate of this virtual machine specific resources.

4. resources of virtual machine configuration-system as claimed in claim 1, is characterized in that, the resource level of this virtual machine comprises the average service rate level of every kind of specific resources.

5. resources of virtual machine configuration-system as claimed in claim 4, is characterized in that, the threshold values that the average service rate of every kind of specific resources is preset is divided into a plurality of levels.

6. resources of virtual machine configuration-system as claimed in claim 5, is characterized in that, the quantity of the pre-set threshold value that every kind of specific resources is corresponding is two, and the average service rate of every kind of specific resources is divided into three levels by two threshold values.

7. resources of virtual machine configuration-system as claimed in claim 1, is characterized in that, the resource specification of this recommendation can not cover the resource specification of this virtual machine initial setting.

8. a resources of virtual machine collocation method, is characterized in that, the method comprises:

Data obtaining step, obtains the utilization rate of a virtual machine specific resources;

Calculation procedure, every an analytical cycle, calculates the average service rate of this virtual machine specific resources, and according to the average service rate of this virtual machine specific resources, determines the resource level of this virtual machine;

Recommendation step, obtains the resource specification of recommendation according to the resource level of this virtual machine, and the resource specification of this recommendation is sent to user side; And

Configuration step, when user side is accepted the resource specification of this recommendation, is this virtual machine Resources allocation according to the resource specification of this recommendation again.

9. resources of virtual machine collocation method as claimed in claim 8, is characterized in that, the average service rate of this virtual machine specific resources obtains according to distributed parallel calculation method.

10. resources of virtual machine collocation method as claimed in claim 9, is characterized in that, described distributed parallel calculation method is:

Acquisition time order according to this virtual machine specific resources utilization rate, is divided into different blocks by the utilization rate of this virtual machine specific resources, and each block is divided and tasks different database servers and carry out parallel addition computing;

The operation result that gathers all database servers, the operation result gathering according to this calculates the average service rate of this virtual machine specific resources.

11. resources of virtual machine collocation methods as claimed in claim 8, is characterized in that, the resource level of this virtual machine comprises the average service rate level of every kind of specific resources.

12. resources of virtual machine collocation methods as claimed in claim 11, is characterized in that, the threshold values that the average service rate of every kind of specific resources is preset is divided into a plurality of levels.

13. resources of virtual machine collocation methods as claimed in claim 12, is characterized in that, the quantity of the pre-set threshold value that every kind of specific resources is corresponding is two, and the average service rate of every kind of specific resources is divided into three levels by two threshold values.

14. resources of virtual machine collocation methods as claimed in claim 8, is characterized in that, the resource specification of this recommendation can not cover the resource specification of this virtual machine initial setting.

Images