KR102135209B1 - Virtual machine placement simulation method and virtual machine placement simulation device implementing the same - Google Patents

Abstract

A virtual machine placement simulation method according to one embodiment of the present invention is a method of simulating changes of the placement of virtual machines placed on a physical server having a first physical server and a second physical server physically separated from the first physical server, and the method includes the steps of: calculating occurrence state information, which is information related to the state of the first physical server and the second physical server during a first period calculated based on information related to the operation of the first and second physical servers stored in a storage unit; calculating selected placement information related to a selected placement, which is a placement selected as a new placement of the virtual machine on the first physical server and the second physical server; and calculating comparison information calculated based on the occurrence state information and predicted state information, which is calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit in order to compare the difference between a case where the virtual machine is placed in the selected placement selected based on the predicted state information, which is information related to the state of the first physical server and the second physical server during a second period and a case wherein the virtual machine is placed in the past placement, which is the placement of the virtual machine that is the basis of the occurrence state information.

Description

Virtual machine placement simulation method and virtual machine placement simulation device implementing the same {VIRTUAL MACHINE PLACEMENT SIMULATION METHOD AND VIRTUAL MACHINE PLACEMENT SIMULATION DEVICE IMPLEMENTING THE SAME}

The present invention relates to a virtual machine deployment simulation method and a virtual machine deployment simulation apparatus that implements the same, and a virtual machine deployment simulation method and a realization method, which is a method for simulating changing a virtual machine deployment on a physical server It relates to a virtual machine deployment simulation apparatus.

The data center, which accounts for about 2% of the world's electricity consumption, is a high-density energy-consuming building that consumes the most power among single buildings, but has maintained an inefficient operation by adhering to conservative operations that prioritize stability. Due to the proliferation of IoT, big data, and cloud technologies in the future, small data centers are expected to be integrated into'hyperscale' and'mega' data centers, which are large-scale data centers. The complexity of the project is also expected to deepen. To this end, unnecessary servers are switched to the idle or power saving mode, and the available servers are operated as reliably as possible to maximize the efficiency of the data center operation and to reduce the power consumption, contributing to the reduction of operating costs. It is very important to properly arrange them.

In a virtualized environment that organizes physical resources (Physical Machine, PM, and physical servers) into logical resources (Virtual Machine, VM, and virtual machines), a running virtual machine can be moved to another physical server, which is called live migration. This is an essential management method for managing an efficient data center. Here, a technology related to operating a data center by migrating virtual machines in order to efficiently manage the data center (Japanese Patent Publication No. 5827594, October 23, 2015, registration) has been developed and released.

However, the existing technologies are about operating a data center, and there was the inconvenience of not predicting the results before applying the operating program to the data center.

In order to solve the above problems, an object of the present invention is to provide a virtual machine batch simulation method and a virtual machine batch simulation apparatus that can simulate changes in the operation of a data center through virtual machine relocation.

The problems to be solved by the present invention are not limited to the above-described problems, and the problems not mentioned will be clearly understood by those skilled in the art from the present specification and the accompanying drawings. .

According to a virtual machine placement simulation method according to an embodiment of the present invention, the placement of a virtual machine disposed on a physical server having a first physical server and a second physical server physically separated from the second physical server is changed. In the virtual machine placement simulation method, which is a method of simulating doing, the first physical server during a first period calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit And generating state information that is information related to the state of the second physical server. Calculating selection batch information related to a selected batch which is a selected batch as a new batch of the virtual machines on the first physical server and the second physical server; And a prediction state that is calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit and is information related to states of the first physical server and the second physical server during a second period. In order to compare the difference between the case where the virtual machine is placed in the selected arrangement selected based on the information, and when the virtual machine is deployed in the past arrangement which is the arrangement of the virtual machines on which the occurrence status information is based And calculating comparison information calculated based on the generated state information and the predicted state information.

In addition, the occurrence status information includes occurrence share status information, which is information related to the load of the physical server charged by the virtual machine during the first period, and a process in which the virtual machine is operated on the physical server during the first period. Occurrence violation status information, which is information related to whether or not a predetermined criterion is violated, occurrence idle status information, which is information related to a time when the physical server is not operated because the virtual machine is not disposed on the physical server during the first period, And at least one of occurrence movement status information, which is information related to the number of movements of the virtual server during the first period, wherein the prediction status information is the physical predicted to be imposed by the virtual machine during the second period. Prediction share status information, which is information related to the load of the server, and prediction violation status information, which is information related to whether or not the predetermined criterion is violated when it is assumed that the virtual machine is operated on the physical server in the arbitrary arrangement during the second period. , Prediction idle information that is information related to the time when the physical server is not operated because the virtual machine is not disposed on the physical server during the second period, and prediction that is information related to the number of movements of the virtual server during the second period It may include at least one of the movement state information.

The method may further include calculating economical information, which is information related to economic benefits obtained when the virtual machine is placed in the selected batch, as compared with the case in which the virtual machine was deployed in the past batch.

In addition, the economical information may be calculated based on the generated idle state information and the predicted idle state information.

In addition, the predicted load of the physical server charged by the virtual machine during the second period through a predetermined prediction method using information related to the operation of the first physical server and the second physical server stored in the storage unit And calculating the first predictive virtual load information that is information related to the first predictive virtual load associated with. The prediction status information may further be calculated based on the first predictive virtual load information.

The method further includes calculating machine placement information, which is information related to machine placement, which is the placement of the virtual machines on the first physical server and the second physical server, calculated by a predetermined machine placement calculation method. The predetermined machine placement calculation method is a method of calculating a plurality of placements of the virtual machine, which are the placements of the virtual machines when a plurality of objective function values are minimized based on the initial prediction virtual load information. It can be one of the batches.

In addition, the step of calculating candidate prediction status information that is information related to values in a maximum value and a minimum value range of a prediction status value calculated based on a plurality of the prediction status information; And displaying, on the display unit, a candidate prediction status value calculated based on the prediction status value and the candidate prediction status information in a predetermined display method, wherein the predetermined display method includes the prediction status value and It may be a method of displaying the candidate prediction state value in a predetermined range form.

According to the virtual machine placement simulation method and the virtual machine placement simulation device according to the present invention, the results can be simulated in advance.

Also, it is possible to accurately reflect the purpose of the user.

In addition, the data center can be managed stably.

The effects of the present invention are not limited to the above-described effects, and the effects not mentioned will be clearly understood by those skilled in the art from the present specification and the accompanying drawings.

1 is a diagram illustrating the concept of a virtual machine deployment simulation apparatus and a relationship between a virtual machine deployment simulation apparatus and a plurality of physical servers according to an embodiment of the present invention.
Figure 2 is a flow chart of a virtual machine deployment simulation method according to an embodiment of the present invention
3 is a view for explaining a step of inputting information among the virtual machine deployment simulation method according to an embodiment of the present invention
4 is a view for explaining a process for setting a selected deployment among the virtual machine deployment simulation method according to an embodiment of the present invention
5 is a view for explaining the process of calculating and displaying economic information and comparative information among the virtual machine deployment simulation method according to an embodiment of the present invention
6 is a view for explaining a first objective function considered among virtual machine placement simulation methods according to an embodiment of the present invention
7 is a view for explaining a second objective function considered among virtual machine placement simulation methods according to an embodiment of the present invention

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings. However, the spirit of the present invention is not limited to the presented embodiments, and a person skilled in the art who understands the spirit of the present invention may add another component within the scope of the same spirit, change, delete, etc. Other embodiments included within the scope of the invention idea can be easily proposed, but this will also be included within the scope of the invention.

In addition, elements having the same functions within the scope of the same idea appearing in the drawings of the respective embodiments will be described using the same reference numerals.

In the present specification, when it is determined that a detailed description of known configurations or functions related to the present invention may obscure the subject matter of the present invention, a detailed description thereof will be omitted.

1 is a diagram illustrating a concept of a virtual machine deployment simulation apparatus and a relationship between a virtual machine deployment simulation apparatus and a plurality of physical servers according to an embodiment of the present invention.

Referring to FIG. 1, a virtual machine (V10) batch simulation apparatus 10 according to an embodiment of the present invention processes information (eg, data) required to perform a virtual machine (V10) batch simulation. The processing unit 11, the storage unit 12 in which information necessary for the virtual machine (V10) batch simulation is stored, and related to information required to perform the virtual machine (V10) batch simulation by user manipulation Input unit 13 for generating a signal, a display unit for displaying a constant image, a transmitting and receiving unit 14 for transmitting and receiving information necessary to perform the virtual machine V10 batch simulation, a display unit 15 for displaying an image, and Controlling the processing unit 11, the storage unit 12, the input unit 13, the transmission/reception unit 14 and/or the display unit 15 so that the simulation simulation of the virtual machine V10 arrangement can be smoothly performed It may include a control unit 16.

The processing unit 11 may perform the same operations as all information processing required to implement the virtual machine V10 batch simulation method based on the information stored in the storage unit 12.

In one example, the processing unit 11 may mean a central processing unit (CPU).

However, the present invention is not limited thereto, and the processing unit 11 may include all devices or components capable of performing calculations required to implement the virtual machine V10 batch simulation method.

The storage unit 12 may store information (data) required to implement the virtual machine V10 placement simulation method.

In the storage unit 12, a program for performing a method for simulating the placement of the virtual machine V10 may be stored.

In one example, the storage unit 12 may be a hard disk as an auxiliary storage device.

However, the present invention is not limited thereto, and the storage unit 12 may include all elements capable of storing all information necessary for the virtual machine V10 deployment simulation method to be outdated.

For example, the storage unit 12 may be a database.

The transmitting/receiving unit 14 may receive information or transmit information to an arbitrary medium.

For example, the transceiver 14 may receive information from the physical server P10 or may transmit information.

For example, the transceiver 14 may receive information directly from the physical server P10, but may receive information about the physical server P10 by a separate storage medium.

The transmission/reception unit 14 may receive information to be delivered in another configuration provided by the virtual machine V10 deployment simulation apparatus 10.

Alternatively, the transmission/reception unit 14 may receive predetermined information from other configurations of the virtual machine V10 deployment simulation apparatus 10.

The input unit 13 may generate an input signal by user manipulation.

When the processing unit 11 receives the input signal, the processing unit 11 may calculate information corresponding to the received input signal.

As an example, the input unit 13 may be a keyboard.

However, the present invention is not limited thereto, and the type of the input unit 13 may be variously modified at a level apparent to a person skilled in the art.

The display unit 15 (in the following description, the same as the display unit) may display a predetermined image under the control of the control unit 16.

In one example, the display unit may be a display.

However, the present invention is not limited thereto, and the type of the display unit can be variously modified at a level apparent to a person skilled in the art.

In order for the virtual machine (V10) batch simulation apparatus 10 to smoothly perform the virtual machine (V10) batch simulation method, the control unit 16 includes the processing unit 11, the storage unit 12, The input unit 13 and the transmitting and receiving unit 14 can be controlled.

Configurations of the virtual machine (V10) deployment simulation apparatus 10 may be connected to each other wired and/or wirelessly.

Configurations of the virtual machine (V10) deployment simulation apparatus 10 may be connected to each of the physical servers P10 by wire and/or wireless.

The physical server P10 may mean a hardware platform on which the virtual machine V10 can be executed.

For example, the physical server P10 may be a typical desktop, server computer, laptop computer or mobile phone.

However, the present invention is not limited thereto, and the physical server P10 may be a concept including all means in which the virtual machine V10 can be executed.

The physical server P10 may include first physical servers P10 to nth physical servers P10 (n is a natural number of 2 or more).

Hereinafter, a virtual machine (V10) batch simulation method implemented by using the virtual machine (V10) batch simulation method will be described in detail.

2 is a flowchart of a virtual machine deployment simulation method according to an embodiment of the present invention.

Referring to FIG. 2, in a method for simulating a virtual machine arrangement according to an embodiment of the present invention, it is necessary to perform simulation, and predetermined information, which is predetermined information, is calculated by a signal generated by the input unit (S10) ), a virtual machine workload that is information related to a virtual machine workload that is a load of a physical server charged by the virtual machine by a predetermined workload calculation method based on information related to the operation of the physical server stored in the storage unit Step S20 of calculating information, the first physical server and the second physical server during a first period calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit The second period through a predetermined prediction method using the information related to the operation of the first physical server and the second physical server stored in the storage (S30) and the step of calculating the occurrence status information, which is information related to the state of the It may include the step (S40) of calculating the initial predictive virtual load information, which is information related to the initial predictive virtual load associated with the predicted load of the physical server, which is charged by the virtual machine.

In addition, the virtual machine placement simulation method according to an embodiment of the present invention is information related to the machine placement, which is the placement of the virtual machines on the first physical server and the second physical server, calculated by a predetermined machine placement calculation method. The second period when the virtual machine is disposed on the first physical server and the second physical server according to the machine placement based on the step S50 of calculating in-machine placement information and the initial predicted virtual load information. It may further include the step (S60) of calculating the predicted status information, which is information related to the states of the first physical server and the second physical server.

In addition, in the virtual machine placement simulation method according to an embodiment of the present invention, candidate prediction state information, which is information related to values in the maximum and minimum value ranges of the prediction state values calculated based on the plurality of prediction state information, is provided. Calculated step (S70), the predicted predicted status value and the candidate predicted status value calculated based on the candidate predicted status information is displayed on a display unit in a predetermined display method, the first physical server and the second physical server Step (S80) of calculating the selected batch information related to the selected batch which is the selected batch as the new batch of the virtual machine on the screen, compared with the case where the virtual machine was placed in the past batch, the virtual machine has been placed in the selected batch. When the economic information, which is information related to the economic benefits obtained when it is obtained, is calculated (S90), and is calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit, the When the virtual machine is disposed in the selection arrangement selected based on predicted status information that is information related to the states of the first physical server and the second physical server, and the arrangement of the virtual machine that is the basis of the occurrence status information. It may include the step (S100) of comparing information calculated based on the generated state information and the predicted state information, in order to compare the difference when the virtual machine is deployed in the past batch.

Hereinafter, each step will be described in detail.

3 is a view for explaining a step of inputting information in a virtual machine deployment simulation method according to an embodiment of the present invention.

Referring to FIG. 3, a user may input values displayed on a screen displayed on the display unit 15 through an input unit.

The user can select a physical server and/or virtual machine to enter the specified information.

The processing unit may calculate the designation information by a user manipulation of the input unit.

The designation information is information required for simulation, and may be information that is previously designated by a user.

As an example, the designation information may include information related to the specification A10 of each physical server.

As an example, the user may input a CPU value (A11), a memory value (A12), a network I/O value (A13), and/or a disk I/O (A14) value of each physical server through the input unit. have.

As an example, the designation information may include information related to the specification Q10 of each virtual machine.

For example, the user may input a CPU rate Q11 allocated to any virtual machine on the physical server and/or a memory rate Q12 allocated to any virtual machine through the input unit.

In one example, the designation information may include information related to an objective function used in a predetermined method for calculating machine placement.

For example, on the display unit 15, lists of objective functions that can be utilized in the predetermined machine batch calculating method may be displayed, and a user may select an objective function to be used in the predetermined machine batch calculating method. .

The designation information may be calculated by a user's selection.

In one example, the objective function is the first objective function F11, which is a function related to the amount of change in the load of the physical server, the second objective function F12, which is a function related to the difference between the loads of the physical server, and the virtual machine on the physical server. A third objective function F13, which is a function related to the number of times the batch is moved, and a fourth objective function 14, which is a function related to whether or not the physical server is operated, may be provided.

The detailed description of the objective function will be described later.

4 is a view for explaining a process for setting a selected batch among the virtual machine deployment simulation method according to an embodiment of the present invention.

After inputting all the input values displayed on the display, the simulation can be executed.

Information related to the operation of physical servers may be stored in the storage unit.

The processing unit may calculate the virtual machine workload information by a predetermined workload calculation method based on information related to the operation of the physical server.

The virtual machine workload may mean a load of resources of the corresponding physical server when the virtual machine is running on the physical server.

The virtual machine workload may mean a load applied to the physical server when the virtual machine is running on the physical server.

In one example, the virtual machine workload may be calculated as a load imposed on the central processing unit (CPU) of the physical server by operating the virtual machine on the physical server.

Accordingly, the predetermined workload calculation method may include a method of calculating the virtual machine workload as a load imposed on the central processing unit of the physical server.

In other words, in the predetermined workload calculation method, the virtual machine workload information may be calculated based only on the degree of operation of the CPU.

It can be assumed that one virtual machine can be executed only on one physical server on the above-mentioned predetermined workload calculation method.

In one example, this may mean that one virtual machine cannot be implemented on two physical servers.

The meaning of the workload of the virtual machine may mean a load imposed on the physical server by operating the virtual machine.

The meaning of the workload of a physical server may mean a load imposed on the physical server.

For example, the workload of the physical server may be calculated as a load imposed on the central processing unit (CPU) of the physical server.

In one example, the workload of a physical server may be calculated by adding virtual machine workloads, which are loads (work loads) of virtual machines disposed on the physical server.

This can further increase the accuracy of prediction in the steps described later by simplifying the basic data necessary for prediction described below.

The processing unit may calculate occurrence state information related to the operation of the physical server during the first period.

The occurrence status information is the occurrence share status information that is information related to the load of the physical server charged by the virtual machine during the first period, and is determined in advance during the operation of the virtual machine on the physical server during the first period. Occurrence violation status information, which is information related to whether a standard is violated, occurrence idle status information, which is information related to a time during which the physical server is not operated because the virtual machine is not disposed on the physical server during the first period, and the first period During this, it may include occurrence movement state information, which is information related to the number of times the virtual server is moved.

The occurrence occupancy state information may be information related to a load value of a physical server charged by the virtual machine during the first period.

For example, referring to FIG. 4, the load (work load) value of each physical server may be represented on the display unit as an average graph W11, a maximum value graph W12, and a minimum value graph W13.

The average graph W11 may mean a graph showing an average of workloads applied to each physical server during the first period.

The maximum value graph W12 may refer to a graph illustrated by extracting only the maximum value among workloads applied to each physical server during the first period.

The minimum value graph W13 may refer to a graph illustrated by extracting only the minimum value among workloads applied to each physical server during the first period.

Through this graph, the user can clearly grasp the degree of operation of the physical server so far.

The occurrence occupancy state information may be calculated based on the virtual machine workload information.

The occurrence violation status information may be information on whether a predetermined criterion is violated when the physical server on which the virtual machine is deployed is operated.

The predetermined criterion may mean a predetermined criterion recommended for the physical server to operate.

For example, the predetermined criterion may include a criterion in which the workload applied to the physical server is 90% or less based on the total workload.

However, the present invention is not limited thereto, and the predetermined criteria may be variously changed and added according to the use mode, purpose and place.

For example, referring to FIG. 4, an occurrence violation status value E11 calculated based on the occurrence violation status information may be displayed on the display unit.

The generated idle state information may be information related to a time when a physical server is not operated because a virtual machine is not disposed on the physical server.

The generated idle state information may be calculated as a total sum of all physical servers.

For example, referring to FIG. 4, the generated idle state value R11 calculated based on the generated idle state information may be displayed on the display unit.

The occurrence movement state information may be information related to the number of times the virtual machine has moved on the physical server.

The generated movement state information may be calculated as a total sum of all physical servers.

For example, referring to FIG. 4, an occurrence movement state value T11 calculated based on the occurrence movement state information may be displayed on the display unit.

For example, referring to FIG. 4, the display unit 15 may simultaneously display values related to the occurrence status information and values related to the prediction status information.

In this way, the user can clearly compare and judge before and after the change.

Hereinafter, processes for calculating the prediction state information will be described in detail.

The processing unit uses the virtual machine workload information to predict the initial prediction associated with the predicted load of the physical server charged by the virtual machine during the second period, which is a future period through a predetermined prediction method. Virtual load information can be calculated.

That is, it may be a process of predicting a workload of a physical server applied to a virtual machine in the future based on data stored in the storage unit.

For example, the predetermined prediction method may be a method of calculating virtual machine workload information for the second period by using an algorithm such as multi-regression model, ARIMA analysis, or deep learning based on the virtual machine workload information. have.

However, not limited to this, all known technologies that can implement a method for predicting a future workload may be included, and detailed description thereof may be omitted as it is a technique obvious to those skilled in the art.

For example, the initial predictive virtual load may mean a workload of a virtual machine predicted for a second period from a time point when the virtual machine workload information is calculated or for a second period from an arbitrary time point.

The processing unit may calculate the machine placement information by a predetermined machine placement calculation method based on the calculated first predicted virtual load information.

The predetermined machine placement calculation method will be described in detail below.

The number of machine arrangements may vary depending on the number of objective functions specified by the user.

For example, if three objective functions are designated by a user, the machine arrangement may be calculated in three ways.

This may be because each of the objective functions is non-dominant to each other.

The processing unit may calculate predicted state information when a virtual machine is disposed on a physical server according to each machine arrangement.

That is, if there are a plurality of objective functions, a plurality of prediction state information may also be formed.

The prediction state information may be calculated based on initial prediction virtual load information.

The predicted status information is predicted occupancy status information, which is information related to the load of the physical server predicted to be charged by the virtual machine during the second period, and the virtual machine is physically configured in the arbitrary arrangement during the second period. Presumed violation status information, which is information related to whether or not the predetermined criterion is violated when it is assumed to operate on a server, and information related to a time when the virtual server is not operated because the virtual machine is not disposed on the physical server during the second period. It may include at least one of the predicted idle state information and the predicted moving state information that is information related to the number of times the virtual server moves during the second period.

The predicted occupancy state information may be information related to a load value of a physical server charged by the virtual machine during a second period.

For example, on the display unit, the load (work load) value of each physical server may be represented by an average graph (Z11), a maximum value graph (Z12), and a minimum value graph (Z13).

The average graph Z11 may mean a graph showing an average of workloads applied to each physical server during the second period.

The maximum value graph Z12 may refer to a graph illustrated by extracting only the maximum value among workloads applied to each physical server during the second period.

The minimum value graph Z13 may refer to a graph illustrated by extracting only the minimum value among workloads applied to each physical server during the second period.

Through this graph, the user can clearly grasp the degree of operation of the physical server so far.

The occurrence occupancy state information may be calculated based on the virtual machine workload information.

Prediction violation status information may be information on whether or not a predetermined criterion is violated when the physical server is operated for a second period.

A prediction violation status value C11 that is a value for the prediction violation status information may be displayed on the display unit.

Detailed description of the predetermined criteria may be omitted in the overlapping with the above.

The predicted idle state information may be information related to a time when the physical server is not operated because a virtual machine is not disposed on the physical server during the second period.

The predicted idle state information may be calculated as a total sum of all physical servers.

The predicted movement state information may be information related to the number of times the virtual machine has moved on the physical server during the second period.

The predicted movement state information may be calculated as a total sum of all physical servers.

The processing unit may calculate the prediction state information for each machine arrangement.

An input mark for selecting each machine arrangement may be displayed on the display unit.

Prediction status information corresponding thereto may be displayed on the display unit based on the machine arrangement information input through the display unit.

As an example, referring to FIG. 4, it can be assumed that there are four machine arrangements by a user designating four objective functions.

At this time, when the user clicks on the first batch, values related to prediction status information when the virtual machine is placed on the first batch for a second period may be displayed on the display unit.

On the other hand, when the user clicks on the second batch, values related to prediction status information when the virtual machine is placed on the second batch for a second period may be displayed on the display unit.

However, the prediction idle state values (B11, B12, B13, and B14) related to the prediction idle state information and/or the predicted motion state values (N11, N12, N13) related to the predicted motion state information, regardless of the user's choice of machine placement , N14), values for all arrangements may be displayed on the display unit.

At this time, the predicted idle state value and the predicted movement state value related to the arrangement of the virtual machine specified by the user may be displayed on the display unit as separate markers L10.

The processing unit may calculate candidate prediction status information based on the prediction status information.

Specifically, the processor may calculate first candidate predicted status information calculated based on the predicted idle status information corresponding to each machine arrangement.

To describe the first candidate prediction state information in detail, a prediction idle state value corresponding to each machine arrangement may be calculated.

In this case, values between the prediction idle state value having the maximum value and the prediction idle state value having the minimum value may be a first candidate prediction state value, and data related to the first candidate prediction state value may include first candidate prediction state information. Can be

As a specific example, referring to FIG. 4, assuming that the user has designated four objective functions, when the virtual machine is the first batch, the predicted idle state value B11 may be 2.3 days, and the virtual machine is the second The predicted idle state value B12 when in a batch may be 1.1 days, the predicted idle state value B13 when the virtual machine is a third batch, and the predicted idle state when the virtual machine is a fourth batch The value B14 may be 5 days.

Here, the predicted idle state value having the maximum value is the predicted idle state value (B14) when the virtual machine is in the fourth batch, and the predicted idle state value having the minimum value is the predicted idle state value when the virtual machine is in 2 batches ( B12).

Accordingly, the first candidate predicted state value B15 may be a value that is greater than 1.1 days and less than 5 days except for the predicted idle state value when the virtual machine is the first batch and the predicted idle state value when the virtual machine is the third batch.

Also, the processing unit may calculate second candidate predicted status information calculated based on the predicted movement status information corresponding to each machine arrangement.

Similarly, a predicted movement state value corresponding to each machine arrangement can be calculated.

At this time, values between the predicted movement state value having the maximum value and the predicted movement state value having the minimum value may be the second candidate prediction state value, and the data related to the second candidate prediction state value is the second candidate prediction state information. Can be

As a specific example, referring to FIG. 4, assuming that the user has designated four objective functions, the predicted movement state value N11 may be 15 times when the virtual machine is the first batch, and the virtual machine is the second batch The predicted movement state value N12 may be 12 times, and the predicted movement state value N13 may be 23 times when the virtual machine is the third batch, and the predicted movement state value N14 when the virtual machine is the fourth batch ) May be 7 times.

Here, the predicted movement state value having the maximum value is the predicted movement state value (N13) when the virtual machine is the third batch, and the predicted movement state value having the minimum value is the predicted movement state when the virtual machine is the fourth batch It may be a value N14.

Therefore, the second candidate predicted state value N15 exceeds 7 times except for the predicted movement state value N12 when the virtual machine is the second batch and the predicted movement state value N11 when the virtual machine is the first batch 23 It may be a value that is more than once.

The processor may display the mall candidate prediction status values B15 and N15 in a display unit using a predetermined display method.

The predetermined display method may be a method of displaying the predicted status value and the candidate predicted status value in a predetermined range form (for example, a bar format).

The predicted status value may mean a value calculated based on the predicted status information.

It may mean a value calculated based on the predetermined information.

As a specific example, referring to FIG. 4, the predetermined display method is a method of displaying the predicted idle state values B11, B12, B13, and B14 and the first candidate predicted state value B15 in a predetermined range form. It can contain.

In addition, referring to FIG. 4, the predetermined display method is a method of displaying the predicted movement state values N11, N12, N13, and N14 and the second candidate predicted state value N15 in a predetermined range. It can contain.

Through this, the user can effectively compare the state of the physical server when the virtual machine is deployed with each of the machine arrangements.

The user can select the placement of one virtual machine from among the machine placements.

5 is a view for explaining the process of calculating and displaying economic information and comparison information among the virtual machine placement simulation method according to an embodiment of the present invention.

When the user selects the placement of the virtual machine, selection placement information may be calculated.

The processing unit may calculate the economical information based on selection batch information related to a selection batch, which is a batch of virtual machines selected by a user.

Economic information can mean data related to economic indicators.

Here, the economic indicator for economic information may be an indicator related to economic benefits generated when the virtual machine is deployed in a selective batch during a second period compared to the virtual machine deployed in a first batch during the first period.

As an example, the economic indicator may include indicators such as ROI, net present price (NPV), internal rate of return (IRR), and payback period (BEP).

In one example, the processing unit may calculate a fee incurred for operating the physical server based on the operating time of the physical server and the electrical charge generated per unit when the virtual machine is deployed on the physical server in the past deployment.

In addition, when the virtual machine is disposed on the physical server in a selective arrangement, the processing unit may calculate an estimated fee required to operate the physical server based on the operating time of the physical server and electricity generated per unit.

Based on this, the processing unit may calculate the profit amount to calculate the economic indicators.

In other words, the economical information may be calculated based on the generated idle state information and the predicted idle state information.

Here, the information on the investment money may be stored in advance in the storage unit.

Alternatively, information on the investment money may be generated by a signal generated from the input unit.

In addition, the processing unit may calculate comparison information comparing each other when the virtual machines are disposed in the selected batch during the second period compared to those in which the virtual machines were deployed in the first batch during the first period.

The comparison information may be related to information that provides a further improvement than before when a virtual machine is placed on a physical server in a selective arrangement.

Comparison values G11 calculated based on the comparison information may be displayed on the display unit.

In addition, economic indicator values (K11, K12, K13, and K14) calculated by the economic indicator may be displayed on the display unit.

As an example, the comparison information may include information calculated based on occurrence occupancy state information and predicted occupancy state information.

In one example, the comparison information may be information about an improvement in the workload of the physical server.

For example, the comparison information may be information calculated based on occurrence violation status information and prediction violation status information.

For example, the comparison information may include information calculated based on the occurrence violation status value and the prediction violation status value.

For example, if the occurrence violation status value is 10 and the prediction violation status value is 5, the comparison value calculated based on the occurrence violation status information and the prediction violation status information may be 5.

In the same way, the comparison information may be information calculated based on the generated idle state information and the predicted idle state information.

That is, the comparison information may include information calculated based on the generated idle state value and the predicted idle state value.

Also, the comparison information may be information calculated based on the generated movement state information and the predicted movement state information.

That is, the comparison information may include information calculated based on the generated movement state value and the predicted movement state value.

While viewing the comparison value displayed on the display unit, the user can easily recognize the economic benefits that can be obtained when placed in a selective arrangement.

Further, on the display unit, a predicted state value corresponding to when a virtual machine is placed on a physical server in a selective arrangement may be displayed together with the comparison value.

The first period may mean a certain period before performing the simulation.

Information related to the operation of the physical server during the first period may be stored in the storage unit.

Through the above-described simulation method, the virtual machine placement effect can be analyzed in advance without having to re-deploy the virtual machine.

Hereinafter, a predetermined machine placement calculation method will be described in detail.

The predetermined machine placement calculation method includes a first predetermined machine placement calculation method that allows the placement movement of the virtual machine during the placement movement period (second period) and a second predetermined time in which the placement movement of the virtual machine is not considered during the placement movement period A machine batch calculation method can be provided.

The user can select which type of predetermined machine placement calculation method to adopt in the information input step above.

Hereinafter, the first predetermined machine placement calculation method will be described in detail.

The processing unit may calculate machine placement information, which is information related to the placement of the virtual machine on the physical server (the first physical server and the second physical server), by the first predetermined machine placement calculation method.

Here, the first predetermined machine placement calculation method may mean a method in which the placement of the virtual machine is calculated when an objective function value is minimized based on the initial prediction virtual load information.

The first predetermined machine placement calculation method may include a method in which the machine placement of the virtual machine is calculated such that the load of the physical server is imposed only within a predetermined range for any physical server. (1st condition)

As an example, the predetermined range may be 10% to 80% of the utilization rate of the physical server (the degree to which the load is imposed).

However, the present invention is not limited thereto, and the predetermined range may be variously modified at a level apparent to those skilled in the art.

In the first predetermined machine placement calculation method, the load of the arbitrary physical server is the first criterion (X10, FIG. 6 and 6) so that the load of the physical server can be charged only within a predetermined range for the arbitrary physical server. If it is predicted to be abnormal, it may include a method of moving at least one of the virtual machines disposed on the arbitrary physical server to another physical server. (Condition 1-1)

As an example, the first criterion (see X10, FIGS. 3 and 4) may mean that the utilization rate (a degree to which a load is applied) of the physical server is 80%.

The processing unit is a solution to prevent the load of the physical server from exceeding the first criterion, based on a time point when the workload of the target physical server is predicted to exceed the first criterion, at least one disposed on the target physical server The machine placement can be calculated using a method of moving a virtual machine to another physical server.

Due to this, it is possible to effectively prevent the workload (operation rate) of the physical server from being applied with an excessive load beyond the first criterion.

However, the present invention is not limited thereto, and the first criterion can be variously modified at a level apparent to a person skilled in the art.

In addition, in the first predetermined machine placement calculation method, the load of the arbitrary physical server is the second criterion (X20, FIG. 4) so that the load of the physical server can be charged only within a predetermined range for the arbitrary physical server. 3 and FIG. 4) If it is predicted to be below, it may include a method of moving all of the virtual machines disposed on the arbitrary physical server to another physical server. (Condition 1-2)

As an example, the second criterion (see X20, FIGS. 6 and 7) may mean that the utilization rate (the degree to which the load is applied) of the physical server is 10%.

The processing unit is a solution to prevent the load of the physical server from falling below the second criterion, based on a time point when the workload of the target physical server is predicted to fall below the second criterion, all virtual servers disposed in the target physical server The machine placement can be calculated using a method of transferring machines to another physical server.

This may mean a start condition of a condition in which the physical server is not operated.

In other words, the processing unit may stop the operation of the target physical server by disposing the virtual machine disposed on the target physical server as another physical server as a solution to prevent the physical server load from being lower than the second reference.

The first predetermined machine placement calculation method is a method in which the machine placement is calculated such that the running physical server is not changed to an inactive state due to movement of the virtual machine that causes the inactive physical server to operate. It can contain. (Second condition)

When the operation of another physical server is started simultaneously with stopping one physical server, there is no benefit of stopping the physical server, and a problem occurs that it takes more power and time to operate the server.

Therefore, through the second condition, it may be intended to solve such a problem.

In addition, when calculating the machine arrangement using the first predetermined machine arrangement calculation method, the machine arrangement may be calculated to allow movement of the virtual machines between physical servers during the second period. (3rd condition)

However, since the data loss may be caused when the virtual machine is frequently moved, the movement of one virtual machine during the second period (division period) may be limited to a predetermined number of times. (4th condition)

For example, the predetermined number of times may be two times.

However, the present invention is not limited thereto, and the predetermined number of times can be variously modified at a level apparent to those skilled in the art.

In addition, when calculating the machine arrangement using the first predetermined machine arrangement calculation method, it is possible to calculate the machine arrangement on the assumption that the specifications of each physical server are the same (homogenous condition). (The fifth condition)

The objective function may include a first objective function, a second objective function, a third objective function, and a fourth objective function.

The processing unit may calculate four non-dominant values of each objective function having a minimum value.

Hereinafter, the objective function will be described in detail.

6 is a view for explaining a first objective function considered among virtual machine placement simulation methods according to an embodiment of the present invention.

Referring to FIG. 6, the objective function may include a first objective function that is a function related to a change amount of a load (work load) of a physical server calculated by the initial predictive virtual load during a second period.

As an example, FIG. 6(a) is a graph illustrating workloads of the first to third physical servers when virtual machines are disposed on the first to third physical servers according to the first schedule.

6(b) is a graph showing workloads of the first to third physical servers when virtual machines are disposed on the first to third physical servers according to the second schedule.

It can be seen from the above that the workload of each physical server can be calculated as the sum of the virtual machine workloads installed in each physical server.

Referring to FIG. 6(a), when virtual machines are disposed on a physical server according to a first schedule, workloads of all physical servers may be less than a first criterion (X10) and a second criterion (X20). have.

Also, referring to FIG. 6(b), when virtual machines are disposed on a physical server according to a second schedule, workloads of all physical servers are less than the first criterion (X10) and greater than the second criterion (X20). Can be.

However, when the virtual machines are placed on the physical server based on the second schedule, the amount of change (slope) of the workload of each physical server is higher than when the virtual machines are placed on the physical server based on the first schedule. It can be smaller.

This may be a schedule in which the second schedule is more preferable than the first schedule as the machine arrangement of the virtual machine on the physical server.

In addition, the value of the first objective function that causes the second schedule to be derived may be smaller than the value of the first objective function that causes the first schedule to be derived.

If the rate of change of the workload of the physical server is large, the workload of the physical server exceeds a certain criterion for a short time and migration (previous) of the virtual machine can be performed.

In this case, since the frequent migration of the above-described virtual machine may occur, the risk of data loss of the virtual machine may be further increased.

The first objective function may be a function for preventing this.

7 is a view for explaining a second objective function considered in a virtual machine placement simulation method according to an embodiment of the present invention.

Referring to FIG. 7, the objective function may include a second objective function that is a function related to a difference between loads (work loads) of a physical server calculated by the initial predictive virtual load during a second period.

As an example, FIG. 7(a) is a graph illustrating workloads of the first to third physical servers when virtual machines are disposed on the first to third physical servers according to the third schedule.

7(b) is a graph showing workloads of the first to third physical servers when the virtual machines are disposed on the first to third physical servers according to the fourth schedule.

It can be seen from the above that the workload of each physical server can be calculated as the sum of the virtual machine workloads installed in each physical server.

Referring to FIG. 7(a), when virtual machines are disposed on a physical server according to a third schedule, workloads of all physical servers may be less than the first criterion (X10) and exceed the second criterion (X20). have.

Further, referring to FIG. 7(b), when virtual machines are disposed on a physical server according to a fourth schedule, workloads of all physical servers are less than the first criterion (X10) and greater than the second criterion (X20). Can be.

However, when the virtual machines are placed on the physical server based on the fourth schedule, the difference between the workloads of each physical server is smaller than when the virtual machines are placed on the physical server based on the third schedule. Can.

This may be a schedule in which the fourth schedule is more preferable than the third schedule as the machine arrangement of the virtual machine on the physical server.

In addition, the value of the second objective function that causes the fourth schedule to be derived may be smaller than the value of the second objective function that causes the third schedule to be derived.

Here, in order to compare the workload between the physical servers, the physical servers may be compared with each other by using the average value of the workload of the physical server during the first predetermined period.

The meaning of the second objective function may be to use the physical server evenly.

The objective function may include a third objective function that is a function related to the number of times the virtual machine is deployed and moved on the first physical server and the second physical server in order for the virtual machine to be disposed according to the placement of the machine.

In other words, the third objective function may relate to the number of movements of the virtual machine between physical servers in order to be deployed along the machine deployment in the current virtual machine deployment.

The meaning of the third objective function may be to find an arrangement that minimizes the number of movements of the virtual machine.

Whenever a virtual machine is migrated, there is always the risk of data loss.

The meaning of the third objective function may be to minimize the possibility of data loss.

The objective function may include a fourth objective function that is a function related to whether the first physical server is operating and whether the second physical server is operating during the second period.

In other words, the fourth objective function may be a function related to whether the physical server is operated during the predetermined first period.

As the fourth objective function is smaller, the number of physical servers that are not operated and the time when the physical server is not operated may be maximized.

When the fourth objective function is minimum, the sum of the times of inactivity per physical server may be maximum for all physical servers.

The meaning of the fourth objective function may be meaningful in minimizing server uptime and thus minimizing power loss due to physical server operation.

The machine placement may refer to the placement of the virtual machines on the physical server on how the virtual machines are placed on the physical servers so that each objective function can be minimized while the first to fifth conditions are satisfied.

The objective function is a first objective function that is a function related to an amount of change in the load of the physical server calculated by the initial predictive virtual load during a second period, and a physical server calculated by the initial predictive virtual load during the second period. A second objective function that is a function related to the difference between loads, a third objective function that is a function related to the number of times the virtual machine is deployed and moved on the physical server in order to deploy the virtual machine to the machine deployment, and during the second period , A fourth objective function that is a function related to whether the physical server is operated or not.

As an example, the physical server may include a first physical server and a second physical server separate from the first physical server.

The machine arrangement is based on a state in which the value of the first objective function is minimum, physical servers (eg, first, the values of the second objective function, the values of the third objective function, and the fourth objective function are minimized) A first batch, which is the placement of the virtual machine on the first physical server and the second physical server), based on a state in which the value of the second objective function is minimum, the value of the first objective function, the third objective function, and Based on a state in which the value of the fourth objective function is minimized, the second configuration that is the placement of the virtual machine on the physical servers (for example, the first physical server and the second physical server), and the value of the third objective function is minimum In this case, the value of the first objective function, the second objective function, and the fourth objective function are minimized, which is the placement of the virtual machine on the physical servers (eg, the first physical server and the second physical server). Based on a state in which the value of the third batch and the fourth objective function is minimum, physical servers (for example, the first of which the values of the first objective function, the values of the second objective function, and the third objective function are minimized) And a fourth arrangement, which is the arrangement of the virtual machine on the physical server and the second physical server).

If only the first objective function and the second objective function are selected by the user, the machine arrangement may include only the first and second arrangements, and the detailed description thereof is the same as the above-described content. Can be omitted from

Hereinafter, the second predetermined machine placement calculation method will be described in detail.

The processing unit may calculate machine placement information, which is information related to the placement of the virtual machine on the first physical server and the second physical server, by a second predetermined machine placement calculation method.

Here, when calculating the machine arrangement using the second predetermined machine arrangement calculation method, it is possible to calculate the machine arrangement on the assumption that the specifications of each physical server are the same (homogenous condition).

The first to fourth conditions applied to the first predetermined machine placement calculation method may not be applied to the second predetermined machine placement calculation method.

Only the first objective function, the second objective function, and the third objective function may be considered as the second predetermined machine placement calculation method.

Hereinafter, a detailed description of the second predetermined machine placement calculation method may be omitted in a limit overlapping with the first predetermined machine placement calculation method.

In order to more clearly represent the technical spirit of the present invention, the accompanying drawings are briefly expressed or omitted for configurations that are not related or inferior to the technical spirit of the present invention.

In the above, the configuration and features of the present invention have been described based on the embodiment according to the present invention, but the present invention is not limited to this, and various modifications or changes can be made within the spirit and scope of the present invention. It is apparent to those skilled in the art, and thus, such changes or modifications are revealed to be within the scope of the appended claims.

11: processing unit 12: storage unit
13: input unit 14: transceiver
15: display unit 16: control unit

Claims (7)

In the virtual machine deployment simulation method, which is a method of simulating changing the arrangement of a virtual machine disposed on a physical server having a first physical server and a second physical server physically separated from the first physical server,
Occurrence status information that is information related to states of the first physical server and the second physical server during a first period calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit Step is calculated;
The first associated with the predicted load of the physical server charged by the virtual machine during the second period through a predetermined prediction method using information related to the operation of the first physical server and the second physical server stored in the storage unit Calculating initial predictive virtual load information that is information related to the predictive virtual load;
Calculating machine placement information, which is information related to the placement of the virtual machine on the first physical server and the second physical server, calculated by a predetermined machine placement calculation method;
Calculating selection batch information related to a selected batch which is a selected batch as one of the machine batches as a new batch of the virtual machines on the first physical server and the second physical server;
Prediction state that is calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit and is information related to the states of the first physical server and the second physical server during the second period Comparing the difference, based on information, when the virtual machine is deployed in the selected selected batch and when the virtual machine is deployed in a past batch which is the batch of the virtual machine on which the occurrence status information is based In order to do so, calculating comparison information calculated based on the generated state information and the predicted state information;
Calculating candidate prediction status information that is information related to values in a maximum value and a minimum value range of the prediction status value calculated based on a plurality of the prediction status information; And
And a step of displaying a candidate prediction status value calculated based on the prediction status value and the candidate prediction status information on a display unit in a predetermined display method.
The predicted status information,
Calculated based on the first predicted virtual load information,
The predetermined machine batch calculation method,
It is a method of calculating a plurality of machine arrangements, which are arrangements of the virtual machines, when a plurality of objective function values are minimized based on the first predicted virtual load information.
The predetermined display method,
A method of displaying the predicted state value and the candidate predicted state value in a predetermined range form,
Virtual machine placement simulation method.
According to claim 1,
The occurrence status information,
Occurrence share status information, which is information related to the load of the physical server charged by the virtual machine during the first period, whether the virtual machine violates a predetermined criterion during the operation of the virtual machine on the physical server during the first period, and Occurrence violation status information, which is related information, occurrence idle status information, which is information related to a time during which the physical server is not operated because the virtual machine is not disposed on the physical server during the first period, and of the virtual machine during the first period. And at least one of occurrence movement status information, which is information related to the number of movements,
The predicted status information,
Predicted occupancy status information, which is information related to the load of the physical server predicted to be charged by the virtual machine during the second period, and assumes that the virtual machine is operated on the physical server in an arbitrary arrangement during the second period In case, the prediction violation status information, which is information related to whether or not the predetermined criterion is violated, the prediction idle state information, which is information related to the time when the physical server is not operated because the virtual machine is not disposed on the physical server during the second period, and And at least one of predicted movement state information that is information related to the number of times the virtual machine is moved during the second period.
Virtual machine placement simulation method.
According to claim 2,
The method further includes calculating economical information, which is information related to economic benefits obtained when the virtual machine is placed in the selected batch compared to when the virtual machine is deployed in the past batch.
Virtual machine placement simulation method.
According to claim 3,
The economic information,
Calculated based on the generated idle state information and the predicted idle state information,
Virtual machine placement simulation method.
In the virtual machine deployment simulation method, which is a method of simulating changing the arrangement of a virtual machine disposed on a physical server having a first physical server and a second physical server physically separated from the first physical server,
Occurrence status information that is information related to states of the first physical server and the second physical server during a first period calculated based on information related to the operation of the first physical server and the second physical server stored in the storage unit Step is calculated;
The first associated with the predicted load of the physical server charged by the virtual machine during the second period through a predetermined prediction method using information related to the operation of the first physical server and the second physical server stored in the storage unit Calculating initial predictive virtual load information that is information related to the predictive virtual load;
Selecting an objective function to be used in a predetermined machine batch calculation method from a plurality of objective function lists;
Calculating machine placement information, which is information related to machine placement, which is the placement of the virtual machines on the first physical server and the second physical server, calculated by the predetermined machine placement calculation method; And
It includes; a new batch of the virtual machines on the first physical server and the second physical server, and as one batch of the machine batches, calculating the selected batch information related to the selected batch which is the selected batch; and
The predetermined machine batch calculation method,
Based on the initial prediction virtual load information, when the value of the selected objective function is minimized, a method of calculating the placement of the virtual machine, the machine placement,
Virtual machine placement simulation method.
The method of claim 5,
The above objective function list,
It has a first objective function that is a function related to the amount of change in the load of the physical server calculated by the initial predictive virtual load during the second period,
Virtual machine placement simulation method.
The method of claim 5,
The above objective function list,
And a second objective function that is a function related to a difference between loads of the physical server calculated by the initial predictive virtual load during the second period,
Virtual machine placement simulation method.

Images