Virtual machine optimal placement recommendation device and server operating system including the same

Abstract

The virtual machine optimal placement recommendation device, which is a virtual machine optimal placement recommendation device for optimal placement of virtual machines running on a physical server, may include a collection module that collects operation information, which is information generated while an administrator operates the physical server, a preference evaluation module that calculates a preference rating value related to the administrator's management preference based on the operation information using a predetermined preference rating calculation method, and a recommendation module that calculates recommendation information that reflects the preference rating value and recommends the administrator the placement of virtual machines on the physical server.

Description

[Technical Field]

[0001] The present invention relates to a virtual machine optimal placement recommendation device for placing and managing virtual machines on a physical server, and a server operating system including the same. [Background technology]

[0002] As the technological capabilities of the Internet network improve and spatial constraints on computer work, such as working from home, are eliminated, cloud environments have grown rapidly in recent years. From the perspective of cloud service providers, how to manage virtual machines on physical machines is a very important issue in order to provide cloud services without interruption.

[0003] Previously, based on a first-fit decreasing algorithm, it recommended migrating virtual machines from less used servers to more used servers in stages, minimizing server operation. However, minimizing server operation can cause overload on the remaining operating servers, leading to concerns about server stability. Summary of the Invention [Problem to be solved by the invention]

[0004] The present invention is intended to solve the above-mentioned problems, and provides a virtual machine optimal placement recommendation device that objectively evaluates and determines the placement of virtual machines and recommends optimal virtual machine placement based on the evaluation, and a server operating system including the same. [Means for solving the problem]

[0005] A virtual machine optimal placement recommendation device according to one embodiment of the present invention is a virtual machine optimal placement recommendation device for optimal placement of virtual machines running on a physical server, and includes: a collection module that collects operation information, which is information generated while an administrator operates the physical server; a preference evaluation module that calculates a preference rating value related to the administrator's management preferences based on the operation information using a predetermined preference rating calculation method; and a recommendation module that reflects the preference rating value and calculates recommendation information that recommends the administrator to place virtual machines on the physical server.

[0006] The system further includes a placement calculation module that calculates an optimal placement, which is a placement of virtual machines on a physical server that satisfies predetermined placement conditions, and the placement calculation module The placement evaluation module receives placement evaluation values ​​obtained by evaluating the optimal placements using a predetermined evaluation method, and the recommendation module can sort the optimal placements in order of similarity of the placement evaluation values ​​to the selection value according to the predetermined recommendation method, and select a placement of virtual machines to recommend.

[0007] The selection popularity value and the placement evaluation value may be calculated taking into consideration aspects of stability and efficiency.

[0008] The predetermined preference calculation method may be a method of classifying an operation pattern of the virtual machine management system of the administrator based on the operation information and calculating the preference value.

[0009] In addition, the predetermined recommendation method may be a method of comparing the placement evaluation values ​​based on the popularity value and recommending placement of virtual machines in a similar order, and if the similarity between stability and efficiency is the same, recommending placement of virtual machines by placing weight on stability.

[0010] The system may further include an interface module that calculates an interface for displaying the recommendation information generated by the recommendation module to an administrator, and the interface module may calculate an interface for displaying the optimal placement information and the recommendation information together.

[0011] In addition, the interface module can change the preference value within a range from minimum to maximum, and can calculate an interface that reflects the changed preference value and recommends the placement of virtual machines on a physical server.

[0012] In addition, the collection module may collect personal information of the administrator if the predetermined amount of the operational information is not collected, and the predetermined preference calculation method may be a method of calculating the preference value by considering both the personal information and the operational information.

[0013] The present invention may further include a proficiency determination module that determines the administrator's operation proficiency of the virtual machine management system based on the operation information, and the predetermined preference calculation method may be a method of calculating the preference value by changing a weight value to which the personal information and the operation information are applied based on the operation proficiency.

[0014] A virtual machine optimal placement recommendation method according to one embodiment of the present invention is a virtual machine optimal placement recommendation method for operating virtual machines running on a physical server, which is realized by a virtual machine management system, and may include a step of collecting, by a collection module, operation information, which is information generated while an administrator controls the virtual machine management system; a step of calculating, by a preference evaluation module, a preference rating value regarding the administrator's preferences based on the operation information using a predetermined preference calculation method; and a step of selecting, by a recommendation module, an arrangement of virtual machines on a physical server to be recommended to the administrator, reflecting the preference rating value.

[0015] A virtual machine placement evaluation device according to one embodiment of the present invention is a virtual machine placement evaluation device for calculating a placement evaluation value, which is an evaluation of the placement of virtual machines operated on a physical server, and includes: a receiving module that collects operational information, which is information generated while virtual machines are running; a first calculation module that calculates an efficiency evaluation value, which is a value of the efficiency of virtual machine placement based on the operational information; a second calculation module that calculates a stability evaluation value, which is a value for the stability of virtual machine placement based on the operational information; and a placement evaluation module that calculates the placement evaluation value using the efficiency evaluation value and the stability evaluation value.

[0016] The first calculation module can calculate the efficiency evaluation value by calculating the integrated power consumption based on the CPU operating rate of the physical server.

[0017] In addition, the first calculation module can calculate the integrated power amount by adding up the operating power amount, which is the amount of power generated by running a virtual machine, and the transfer power amount, which is the amount of power generated when the virtual machine is migrated.

[0018] In addition, the first calculation module can calculate the operating power amount by adding up the basic power amount, which is the amount of power generated when the CPU is idle, and the additional power amount, which is generated when a virtual machine is running and the CPU is operating.

[0019] Also, the first calculation module may calculate the basic amount of power approximately by dividing the amount of power generated when the CPU is fully operated by a predetermined value.

[0020] The first calculation module may also calculate the additional power amount based on a ratio of the operating frequency of the CPU to a difference between the frequency when the CPU is fully operating and the frequency when the CPU is in an idle state.

[0021] Also, the first calculation module may calculate the amount of transfer power approximately as a predetermined ratio of the amount of power generated when the CPU is fully operated.

[0022] In addition, the first calculation module may calculate the amount of transfer power based on a frequency ratio calculated as a predetermined ratio of the frequency when the CPU is fully operating to the difference between the frequency when the CPU is fully operating and the frequency when the CPU is idle.

[0023] In addition, the second calculation module can calculate the stability evaluation value based on a dirty memory rate and a network link rate.

[0024] A virtual machine placement evaluation method according to one embodiment of the present invention is a virtual machine placement evaluation method for calculating a placement evaluation value, which is an evaluation of the placement of virtual machines operated on a physical server, using a virtual machine placement evaluation device, and may include a step of collecting operation information, which is information generated while virtual machines are running, by a receiving module, a step of calculating an efficiency evaluation value, which is a value of the efficiency of the placement of virtual machines based on the operation information, by a first calculation module, a step of calculating a stability evaluation value, which is a value of the stability of the placement of virtual machines based on the operation information, by a second calculation module, and a step of calculating the placement evaluation value using the efficiency evaluation value and the stability evaluation value by a placement evaluation module. [Effects of the Invention]

[0025] The virtual machine optimal placement recommendation device and the server operating system including the same according to the present invention can maximize the stability of the server.

[0026] Furthermore, the number of personnel required for server management can be minimized.

[0027] Furthermore, the time required to administer the server can be minimized.

[0028] However, the effects of the present invention are not limited to those described above, and any unmentioned effects can be clearly understood by a person having ordinary skill in the art to which the present invention pertains from this specification and the accompanying drawings. [Brief explanation of the drawings]

[0029] [Figure 1] FIG. 2 is a relationship diagram of a server operating system according to an embodiment of the present invention. [Figure 2] FIG. 1 is a diagram illustrating the configuration of a server operating system according to an embodiment of the present invention. [Figure 3] 1 is a flowchart of a method for recommending an optimal virtual machine placement according to an embodiment of the present invention. [Figure 4] 10 is a diagram illustrating utilization weights of personal information and operation information in a virtual machine optimal placement recommendation method according to an embodiment of the present invention; [Figure 5] 10 is an example of an image recommending placement of virtual machines in a method for recommending optimal placement of virtual machines according to an embodiment of the present invention; [Figure 6] 10 is an example of displaying the placement of virtual machines according to a selection value in the method for recommending optimal placement of virtual machines according to an embodiment of the present invention. [Figure 7] 1 is a flowchart of a server operation method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0030] 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 examples shown, and a person skilled in the art who understands the spirit of the present invention may easily propose other inventions or other embodiments within the scope of the present invention by adding, changing, or deleting other components within the scope of the same concept, which are also included in the scope of the present invention. In addition, components having the same functions within the same concept shown in the drawings of each embodiment will be described using the same reference numerals.

[0031] FIG. 1 is a relationship diagram of a server operating system according to an embodiment of the present invention.

[0032] Referring to FIG. 1, a server operating system 100 according to an embodiment of the present invention may be a system that operates, maintains, migrates, etc., a virtual machine installed on a physical server.

[0033] As a specific example, the server operating system 100 can monitor whether a problem occurs in a virtual machine, and if a problem occurs in a virtual machine, can resolve the problem. Furthermore, the server operating system can also optimally allocate virtual machines on a physical server in terms of resource utilization.

[0034] The server operating system 100 can be connected to a physical server 200 and / or an external server 300 via a wired / wireless network. The server operating system can collect and receive all information generated when a virtual machine is running on the physical server, all information generated when the physical server is running, and / or necessary information from an external server.

[0035] The network in this invention may be a core network integrated with a wired public network, a wireless mobile communication network, or the mobile Internet, and may refer to a global open computer network structure that provides the TCP / IP protocol and multiple services at its upper layers, i.e., HTTP (Hyper Text Transfer Protocol), HTTPS (Hyper Text Transfer Protocol Secure), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), etc., and is not limited to these examples, but comprehensively refers to a data communication network that can send and receive data in various forms.

[0036] The physical server in the present invention includes other components for executing the server environment of the server, and can include any type of device.

[0037] As an example, the server may be a digital device, such as a laptop computer, a desktop computer, a web pad, a mobile phone, or the like, that has a processor, memory, and computing capabilities.

[0038] In one example, the server may be a web server, but the type of server may vary widely and be readily apparent to those of ordinary skill in the art.

[0039] FIG. 2 is a diagram showing the configuration of a server operating system according to an embodiment of the present invention.

[0040] Referring to FIG. 2, a server operating system 100 according to an embodiment of the present invention may include a virtual machine optimal placement recommendation device 110, a server management device 120, and a virtual machine placement evaluation device .

[0041] Each device is connected to the other via a wired / wireless network, allowing them to send and receive the necessary information.

[0042] The server management device 120 can carry out all processes required to operate a physical server.

[0043] The server management device 120 may include a storage module 121 that stores information necessary for managing and operating physical servers, a monitoring module 122 that monitors in real time whether virtual machines running on the physical servers are operating abnormally, and a migration module 123 that moves virtual machines between physical servers.

[0044] In addition, the server management device 120 may further include an input module 124 that receives device control input from the administrator, generates a control signal, and transmits it to the virtual machine optimal placement recommendation device 110, the server management device 120, and the virtual machine placement evaluation device 130.

[0045] In addition, the server management device 120 may further include a display module 125 that displays information necessary for the operation and management of the server management device, such as values ​​calculated by the virtual machine optimal placement recommendation device 110, the server management device 120, and the virtual machine placement evaluation device 130, the status of virtual machines, and the status of physical servers.

[0046] The storage module 121 stores all data necessary for the server management device 120 to operate.

[0047] By way of example, the storage module may include internal and / or external memory.

[0048] By way of example, the embedded memory may include at least one of volatile memory (e.g., DRAM, SRAM, or SDRAM) and non-volatile memory (e.g., one time programmable ROM (OTPROM), PROM, EPROM, EEPROM, mask ROM, flash ROM, flash memory, a hard drive, or a solid state drive (SSD).

[0049] The external memory may include a flash drive (e.g., compact flash (CF), secure digital (SD), Micro-SD, Mini-SD, extreme digital (xD), multi-media card (MMC), or memory stick, etc.

[0050] The monitoring module 122 receives log data and metric data generated by the virtual machine and can monitor in real time whether the virtual machine is operating abnormally, including whether it is underloaded or overloaded.

[0051] The migration module 123 can move virtual machines between physical servers in the event of a problem with a physical server or based on an administrator's control signal.

[0052] Before migrating a virtual machine, the migration module 123 can determine whether migration is necessary and whether the physical server is suitable for undergoing migration.

[0053] Through the input module 124, an administrator can input commands and command signals that can manage and operate the server operating system 100.

[0054] For example, the input module may include a mouse, keyboard, touch panel, (digital) pen sensor, key or ultrasonic input device, etc. The touch panel may use at least one of electrostatic, pressure-reducing, infrared, or ultrasonic input methods.

[0055] Furthermore, the touch panel may further include a control circuit. The touch panel may further include a tactile layer to provide a tactile response to the user. A (digital) pen sensor may, for example, be part of the touch panel or may include a separate recognition sheet.

[0056] Additionally, the keys may include, for example, physical buttons, optical keys, or keypads.

[0057] The ultrasonic input device can detect ultrasonic waves generated from the input tool through a microphone and check data corresponding to the detected ultrasonic waves.

[0058] The display module 125 can include any device capable of displaying an image, such as a display device, a screen device, or a beam projector.

[0059] As an example, a display (eg, a display) may include a panel, a holographic device, a projector, or a control circuit for controlling them.

[0060] A virtual machine optimal placement recommendation device 110 according to one embodiment of the present invention, for optimal placement of virtual machines operated on a physical server, may include a collection module 111 that collects operation information, which is information generated while an administrator operates a physical server, a preference evaluation module 112 that calculates a preference rating value associated with the administrator's management signal using a predetermined preference rating calculation method based on the operation information, and a recommendation module 113 that selects a virtual machine placement on the physical server to be recommended to the administrator, reflecting the preference rating value.

[0061] The virtual machine optimal placement recommendation device 110 may further include a placement calculation module 114 that calculates an optimal placement, which is a placement of virtual machines on a physical server that satisfies predetermined placement conditions.

[0062] In addition, the virtual machine optimal placement recommendation device 110 may further include an interface module 115 that calculates an interface for displaying the recommendation information generated by the recommendation module 113 to the administrator, and a proficiency determination module 116 that determines the administrator's proficiency with the virtual machine management system based on the operation information.

[0063] The collection module 111 can collect operation information, which is information generated while an administrator manages a physical server.

[0064] The operation information may include all of the log information and / or metric data of the physical server / virtual machine, and may also include control data that the administrator used to control the physical server and all of the information related to the current operating status of the physical server and virtual machine when the control data occurred.

[0065] For example, the operational information may include how the administrator controlled the physical server and the virtual machine to resolve abnormal situations in the current state of the physical server and the virtual machine.

[0066] The collection module 111 may collect personal information of the administrator if the operation information is not collected in a predetermined amount.

[0067] For example, the predetermined collection amount may be 1 TB, but is not limited to this, and the predetermined collection amount may be varied in various ways as would be obvious to a person skilled in the art.

[0068] The personal information may include information generated by the administrator while controlling the electronic device externally, regardless of the server operating system 100, and the administrator's identity information.

[0069] For this purpose, the collection module 111 can collect necessary information by being connected to an external server, a public institution server, an open server, etc. through a network.

[0070] For this purpose, prior approval may be obtained from the administrator.

[0071] As an example, personal information may include navigation routing information for an administrator.

[0072] When searching for a navigation route, various routes may be recommended, and the navigation user can select one route from multiple routes according to his / her preference.

[0073] For this reason, the administrator's navigation application usage information can be used to understand the administrator's preferences.

[0074] As an example, personal information may include the administrator's vehicle control information.

[0075] In recent years, various control information of automobiles is being collected based on the expansion of IoT technology and the trend of electronic devices in automobiles. The collection module 111 can collect information on how an administrator controls and manages an automobile from the automobile manufacturer's server and / or the news agency's server.

[0076] The control method of the vehicle can be important information in determining whether the manager is someone who values ​​stability or efficiency.

[0077] The vehicle control information is integrated with the navigation information in the vehicle and can be important information for understanding how the manager controlled the vehicle in terms of traffic conditions and locations.

[0078] For example, the personal information may include information about the administrator's age, gender, alma mater, major, and hometown.

[0079] As people get older, they may tend to value stability more than efficiency.

[0080] Women tend to place more importance on stability than efficiency than men.

[0081] A university's academic culture influences its students, so the emphasis placed on efficiency and stability may differ depending on the school from which the student graduated.

[0082] The major one studies influences the learner, so the emphasis placed on efficiency and stability may differ depending on the major.

[0083] Whether one comes from an urban, rural, or fishing village area affects who lives there, and the weighting placed on efficiency and stability may differ.

[0084] For example, if you live in an urban area, you may place more importance on efficiency than stability, and if you live in a rural / fishing village, you may place more importance on stability than efficiency.

[0085] The preference evaluation module 112 can calculate a preference value associated with the manager's management preference according to a predetermined preference calculation method.

[0086] The preference assessment module 112 may store a pattern classification model that classifies the operational information into predetermined operational patterns.

[0087] The preference evaluation module 112 can generate a pattern classification model by deep learning based on classification information obtained by classifying past motion information and corresponding motion information.

[0088] Deep learning may be, but is not limited to, a back propagation algorithm, which is an algorithm that updates the weights of a neural network using labeled data in the output layer.

[0089] Furthermore, since the deep neural network and the back propagation algorithm are well known in the art, detailed explanations thereof may be omitted.

[0090] In one example, the selection value may consist of efficiency and stability values, with the units being %, such that the sum of efficiency and stability is 100.

[0091] However, the present invention is not limited to this, and the form of the selection value can be varied in various ways as would be obvious to a person skilled in the art.

[0092] Operational patterns can be classified into patterns based on stability and efficiency.

[0093] For example, the first operational pattern may be a pattern with 0% stability and 100% efficiency, and the second operational pattern may be a pattern with 2% stability and 98% efficiency. In this way, operational patterns can be classified in terms of all stability and efficiency.

[0094] The predetermined preference calculation method may be a method of classifying an operation pattern of the virtual machine management system of the administrator based on the operation information and calculating the preference value.

[0095] The selection value can be calculated taking into consideration aspects of stability and efficiency.

[0096] The preference evaluation module 112 allows for an effective analysis of the manager's management tendencies.

[0097] The predetermined preference calculation method may be a method of calculating the preference value by taking into consideration both the personal information and the motion information.

[0098] If the amount of collected behavioral information is not determined in advance, the preference evaluation module 112 may calculate a preference value by utilizing personal information as well.

[0099] The preference evaluation module 112 may store a personal calculation model that calculates efficiency and stability through personal information.

[0100] The preference evaluation module 112 can calculate a personal calculation model by deep learning based on values ​​classified into efficiency and stability for past personal information and corresponding personal information.

[0101] The deep learning algorithm utilizes well-known technology, so detailed explanations can be omitted.

[0102] The preference evaluation module 112 may calculate a final preference value by weighting and averaging the first preference value calculated through the pattern classification model and the second preference value calculated through the personal calculation model.

[0103] The preference evaluation module 112 can calculate the preference value by changing the weight of the first and second preference values ​​according to the amount of collected operation information until a predetermined amount of operation information is collected.

[0104] For example, as the amount of collected operation information increases, the first selection favorable rating may be weighted more heavily than the second selection favorable rating, and the average may be calculated to calculate the final selection favorable rating.

[0105] If the information of the operation information is too small, the first-choice popularity value may not fully reflect the manager's tendency and preference.

[0106] To compensate for this, a considerable amount of personal information has already been accumulated, and the second selection value can be further weighted to calculate the final selection value.

[0107] As the amount of motion information gradually increases, the weight of the first selection favorable rating may increase according to the first ratio, and the weight of the second selection favorable rating calculated using personal information may decrease.

[0108] The predetermined preference calculation method may be a method of calculating the preference value by changing weights to which the personal information and the action information are applied based on the operation proficiency.

[0109] In other words, the predetermined preference calculation method may be a method of changing the weighting by a ratio different from the first ratio, in which the weighting of the second selection favorability value is decreased in accordance with the increase in the weighting of the first selection favorability value depending on the operation proficiency, when the amount of collected operation information is greater than or equal to a predetermined reference value.

[0110] As a specific example, the preference evaluation module may be configured to: If the value is equal to or greater than the standard value (change standard value) and the administrator's operational proficiency is equal to or greater than a predetermined level of proficiency, the final selection favorability value can be calculated by setting the rate at which the weight of the second selection favorability value is reduced in accordance with the increase in the weight of the first selection favorability value to a second rate that is even greater than the first rate.

[0111] In addition, when the amount of collected operation information is equal to or greater than a predetermined reference value and the manager's operational proficiency is less than a predetermined proficiency, the preference evaluation module can calculate the final selection favorability value by setting the rate at which the weight of the second selection favorability value is decreased in accordance with the increase in the weight of the first selection favorability value to a third rate that is even smaller than the first rate.

[0112] This may be because when the administrator's proficiency level is low, the consistency of the operation information is somewhat low, resulting in somewhat low information reliability. Conversely, when the administrator's proficiency level is high, the consistency of the operation information is somewhat high, resulting in somewhat high information reliability.

[0113] In the unlikely event that the operation skill level is not calculated by the skill level determination module 116, the preference evaluation module can maintain the first ratio and calculate the final preference value.

[0114] The skill level determination module 116 can calculate the manager's skill level in operating the server operation system 100 based on the operation information.

[0115] For this purpose, the proficiency judgment module 116 may store a proficiency judgment model.

[0116] The proficiency determination module 116 can generate a proficiency determination model by deep learning the way in which the physical server and the virtual machine were operated in accordance with the past states of the physical server and the virtual machine and the evaluation scores for this.

[0117] The specific algorithm for deep learning utilizes known techniques, and a detailed description thereof may be omitted.

[0118] The skill level determination model can calculate the operational skill level of the manager.

[0119] However, since the proficiency assessment model only requires the control information necessary to assess proficiency from the operational information and the status information of the physical server and virtual machine at the time of control, even if there is a lot of operational information, it may not appear that there is a lot of information necessary to assess proficiency.

[0120] The skill level determination model can calculate the operation skill level only when the amount of data required for determining the skill level is equal to or greater than a predetermined standard.

[0121] Here, the amount of data required to determine the proficiency level may be smaller than the change reference value.

[0122] This may be due to the effective operation of the aforementioned algorithm.

[0123] The placement calculation module 114 can calculate the optimal placement of the virtual machines.

[0124] To this end, the placement calculation module 114 can predict the future workload of the virtual machines and calculate the placement of the virtual machines on the physical servers that minimizes a preset objective function based on the predicted workload.

[0125] The method by which the placement calculation module 114 calculates the optimal placement of virtual machines is a known technique, and therefore a detailed description thereof may be omitted.

[0126] The placement calculation module 114 can request an evaluation of the calculated optimal placement from the placement evaluation module 134, which will be described later.

[0127] The placement calculation module 114 can receive a placement evaluation value obtained by the placement evaluation module 134 evaluating the optimal placement using a predetermined evaluation method.

[0128] The placement calculation module 114 can transmit the placement evaluation value and the optimal placement received from the placement evaluation module 134 to the recommendation module 113 .

[0129] The layout evaluation value can be a form that takes into account aspects of stability and efficiency.

[0130] For example, stability and efficiency are expressed as percentages based on an overall value of 100, so that the placement evaluation value and (final) selection value can be easily compared.

[0131] The recommendation module 113 can select the placement of the virtual machine to be recommended by classifying the optimal placements in order of placement evaluation value similar to the popularity value according to a predetermined recommendation method.

[0132] The recommendation module 113 can compare the preference value with the placement evaluation value to recommend a virtual machine placement that reflects the administrator's preference among the optimal placements.

[0133] The recommendation module 113 may calculate recommendation information for recommending the recommendation ranking of the optimal layouts in descending order of layout evaluation value most similar to the selection popularity value among the optimal layouts.

[0134] The predetermined recommendation method may be a method of comparing the placement evaluation values ​​based on the popularity value, recommending placement of virtual machines in a similar order, and recommending placement of virtual machines by placing weight on stability when the similarity between stability and efficiency is the same.

[0135] That is, in the case of virtual machine placements where the distance between the popular selection value and the placement evaluation value is equal, more emphasis is placed on stability, and virtual machine placements with higher stability can be recommended with a higher priority.

[0136] The interface module 115 can calculate an interface such that the information on the optimal placement and the recommendation information are displayed together.

[0137] The display module 125 displays the interface calculated by the interface module 115 , allowing the administrator to easily check the information necessary to manage and operate the server operating system 100 .

[0138] A virtual machine placement evaluation device 130 according to an embodiment of the present invention is a virtual machine placement evaluation device 130 for calculating a placement evaluation value, which is an evaluation of the placement of virtual machines operated on a physical server, and may include a receiving module 131 that collects operation information, which is information generated when virtual machines are running, a first calculation module 132 that improves the efficiency of the operation information, a second calculation module 133 that calculates a stability evaluation value, which is a value for the stability of the placement of virtual machines based on the operation information, and a placement evaluation module 134 that calculates the placement evaluation value using the efficiency evaluation value and the stability evaluation value.

[0139] The receiving module 131 can collect operation information that is information generated when a virtual machine is running.

[0140] The operational information may include hardware operation information of the physical server by the virtual machine.

[0141] As an example, the operational information may include CPU operation information, memory operation information, network information, etc., and the present invention is not limited to the above-mentioned examples, and specific examples of the operational information can be varied in various ways at a level that is obvious to ordinary engineers.

[0142] The first calculation module may calculate an efficiency rating.

[0143] The first calculation module 132 may calculate the integrated power consumption based on the CPU operating rate of the physical server to calculate the efficiency evaluation value.

[0144] For example, the greater the integrated power amount, the greater the efficiency evaluation value may be, and the smaller the integrated power amount, the smaller the efficiency evaluation value may be.

[0145] As an example, CPU utilization can be expressed as CPU frequency in MHZ.

[0146] The first calculation module 132 can calculate the integrated power amount by adding up the operating power amount, which is the amount of power generated in the physical server when the virtual machine is running, and the transfer power amount, which is the amount of power generated when the virtual machine is migrated.

[0147] TIFF2026501375000002.tif16166

[0148] TIFF2026501375000003.tif32166

[0149] That is, when a migration occurs, the amount of power generated for the migration and the amount of power generated by the physical server due to the virtual machine placement after the migration can be combined to calculate the integrated amount of power.

[0150] The first calculation module 132 can calculate the operating power amount by adding up the basic power amount, which is the amount of power generated when the CPU is idle, and the additional power amount generated when a virtual machine is running and the CPU is operating.

[0151] Also, the first calculation module 132 may calculate the approximate basic amount of power by dividing the amount of power generated when the CPU is fully operated by a predetermined value.

[0152] In addition, the first calculation module 132 can calculate the additional power amount based on the ratio of the operating frequency of the CPU to the difference between the frequency when the CPU is fully operating and the frequency when the CPU is in an idle state.

[0153] TIFF2026501375000004.tif29166

[0154] TIFF2026501375000005.tif29166

[0155] TIFF2026501375000006.tif23166

[0156] TIFF2026501375000007.tif29166

[0157] TIFF2026501375000008.tif42166

[0158] Here, the operating frequency of a CPU may refer to the frequency of the CPU when a specific operation is performed in a virtual machine, such as an application program, on a physical server.

[0159] TIFF2026501375000009.tif13166

[0160] TIFF2026501375000010.tif39166

[0161] TIFF2026501375000011.tif26166

[0162] The first calculation module 132 may calculate the approximate amount of power transfer as a predetermined ratio of the amount of power generated when the CPU is fully operated.

[0163] Specifically, the first calculation module 132 can calculate the amount of transfer power based on a frequency ratio calculated based on a predetermined ratio of the frequency when the CPU is fully operating to the frequency when the CPU is idle, which is the difference between the frequency when the CPU is fully operating and the frequency when the CPU is idle.

[0164] TIFF2026501375000012.tif20166

[0165] TIFF2026501375000013.tif10166

[0166] The predetermined ratio may vary depending on the performance of the physical machine, the size of the virtual machine to be migrated, the type of virtual machine, and the like.

[0167] For example, the higher the performance of the physical machine, the smaller the predetermined ratio may be.

[0168] For example, the larger the size of the virtual machine to be migrated, the larger the predetermined percentage may be.

[0169] As an example, the predetermined percentage may be 0.4 (40%).

[0170] However, the present invention is not limited to this, and the predetermined ratio can be variously modified at a level that is obvious to those skilled in the art.

[0171] TIFF2026501375000014.tif26166

[0172] TIFF2026501375000015.tif13166

[0173] The second calculation module 133 can calculate the stability evaluation value based on the dirty memory rate and the network link rate.

[0174] As a specific example, the second calculation module 133 can calculate the migration time by comparing the memory of the virtual machine, the dirty memory rate of the virtual machine, and the network link rate.

[0175] The second calculation module 133 compares the migration time required between the size of the physical server and the size of the virtual machine, and can evaluate and classify the stability value into five stages (Safety, Stable, Normal, Warning, Danger).

[0176] Here, the safety level may have a greater stability evaluation value than the danger level.

[0177] The larger the memory of the virtual machine, the higher the dirty memory ratio of the virtual machine, and the smaller the network link ratio, the smaller the stability evaluation value may become.

[0178] Conversely, the smaller the memory of the virtual machine, the smaller the dirty memory rate of the virtual machine, and the larger the network link rate, the larger the stability evaluation value may be.

[0179] The placement evaluation module 134 can calculate a placement evaluation value using the efficiency evaluation value and the stability evaluation value.

[0180] The placement evaluation module 134 calculates an efficiency evaluation value and a stability evaluation value for the physical machine at the time of evaluation, and can compare the efficiency evaluation value and the stability evaluation value when the placement of the virtual machine is changed according to the optimal placement transmitted from the virtual machine optimal placement recommendation device 110, to finally calculate a placement evaluation value.

[0181] The placement evaluation module 134 calculates an efficiency evaluation value and a stability evaluation value for the physical machine at the time of evaluation, and can compare the efficiency evaluation value and the stability evaluation value when the placement of the virtual machine is changed according to the corrected placement transmitted from the server management device 120 to calculate a final placement evaluation value.

[0182] If the efficiency evaluation value of the layout to be changed (optimal layout or modified layout) is greater than the efficiency evaluation value of the current layout, the efficiency may be increased.

[0183] Conversely, if the efficiency evaluation value of the new layout is smaller than the efficiency evaluation value of the current layout, the efficiency may decrease.

[0184] If the stability evaluation value of the new arrangement is greater than the stability evaluation value of the current arrangement, the stability may be increased.

[0185] Conversely, if the stability evaluation value of the new arrangement is smaller than the stability evaluation value of the current arrangement, the stability may decrease.

[0186] The placement evaluation value may consist of values ​​for efficiency and stability, and may be expressed in units of %, with the sum of efficiency and stability being 100.

[0187] For this reason, the efficiency evaluation value and the layout evaluation value can be numerically processed.

[0188] For example, using a predetermined processing table, the difference between the efficiency evaluation value of the current arrangement and the efficiency evaluation value of the arrangement to be changed may be classified according to the difference between the stability evaluation value of the current arrangement and the stability evaluation value of the arrangement to be changed, and efficiency and stability may be specified in advance for each classification of the difference between the efficiency evaluation value of the current arrangement and the efficiency evaluation value of the arrangement to be changed.

[0189] However, the method of processing the efficiency evaluation value and the placement evaluation value is not limited to this, and can be variously modified at a level that is obvious to an ordinary engineer.

[0190] While the base frequency is easy to measure, the idle frequency is difficult to measure. This may be because it is practically impossible to measure the idle frequency without actually running the server in the field and turning it on.

[0191] FIG. 3 is a flowchart of a method for recommending optimal virtual machine placement according to an embodiment of the present invention.

[0192] Hereinafter, detailed explanations may be omitted to the extent that they overlap with the above content.

[0193] Referring to FIG. 3, an optimal placement recommendation method according to an embodiment of the present invention is a virtual machine management method for operating a virtual machine that is implemented by a virtual machine management system and runs on a physical server, the method comprising the steps of: collecting operation information, which is information generated while an administrator controls the virtual machine management system, by a collection module; The method may include a step in which a preference evaluation module calculates a preference rating value related to the administrator's preference using a predetermined preference calculation method based on the operation information, and a step in which a recommendation module reflects the preference rating value and selects a virtual machine placement on a physical server to recommend to the administrator.

[0194] Operational information may be collected by a collection module.

[0195] If the predetermined amount of operational information is not collected, the administrator's personal information may be collected, their proficiency level may be determined, and the personal information and proficiency level may be used to calculate the selection value.

[0196] If only a predetermined amount of motion information is collected, the preference evaluation index can be calculated using only the motion information.

[0197] The preference value may be continuously updated until the administrator requests a recommendation for virtual machine placement via the input module.

[0198] When an administrator requests placement recommendations for virtual machines, an optimal placement can be calculated by a placement calculation module, and the placement calculation module can request a placement evaluation of the optimal placement from a virtual machine placement evaluation device.

[0199] The virtual machine placement evaluation device may calculate a placement evaluation value by comparing the current placement of the virtual machines with the placement of the virtual machines when placed according to the optimal placement.

[0200] The recommendation module compares the selection value with the placement evaluation value, calculates the optimal placement that is suitable for the administrator's operational tendencies and patterns in a similar order, and can recommend virtual machine placement to the administrator through the interface module and display module.

[0201] FIG. 4 is a diagram illustrating utilization weights of personal information and operation information in the virtual machine optimal placement recommendation method according to an embodiment of the present invention.

[0202] Referring to FIG. 4, for example, when the X-axis motion information is fully utilized, it may mean that a predetermined amount of motion information is collected.

[0203] If no operational information is collected, the selection value can be calculated using only personal information.

[0204] As the amount of motion information gradually increases, the weight of the second evaluation value calculated using the personal information according to the first ratio may be decreased.

[0205] However, the present invention is not limited to this, and the first ratio can be variously modified at a level that is obvious to those skilled in the art.

[0206] FIG. 4(a) shows a case where the operational proficiency level is equal to or higher than a predetermined level. Referring to FIG. 4(a), the predetermined reference value X10 of the amount of collected operation information may be half of the predetermined amount of collection.

[0207] However, the predetermined reference value is not limited to this, and can be variously modified at a level that is obvious to an ordinary engineer.

[0208] When the amount of collected operation information is less than a predetermined reference value (X10), the first and second preference rating values ​​are adjusted by the first ratio (B11) to calculate the final preference rating value, but when the amount of collected operation information is greater than or equal to the predetermined reference value, the first and second preference rating values ​​are adjusted by the second ratio (B12) to calculate the final preference rating value.

[0209] As a result, even if only 75% of the predetermined amount of operation information can be collected, the first selection favorable rating value can be fully utilized, and the final selection favorable rating value can be calculated using only the first selection favorable rating value.

[0210] FIG. 4(b) shows a case where the operational proficiency level is less than a predetermined level. Referring to FIG. 4(b), the predetermined reference value X10 of the amount of collected operation information may be half of the predetermined amount of collection.

[0211] However, the predetermined reference value is not limited to this, and can be variously modified at a level that is obvious to an ordinary engineer.

[0212] When the amount of collected action information is less than a predetermined reference value (X10), the first and second preference rating values ​​are adjusted by the first ratio (B11) to calculate the final preference rating value, but when the amount of collected action information is greater than or equal to a predetermined reference value, the first and second preference rating values ​​are adjusted by the third ratio (B13) to calculate the final preference rating value.

[0213] As a result, for example, even if 100% of the predetermined collection amount of operational information is collected, the first selection popularity value is utilized at 75% and the second selection popularity value is utilized at 25%, and the final selection popularity value can be calculated.

[0214] FIG. 5 is an example of an image recommending placement of virtual machines in the virtual machine optimal placement recommendation method according to one embodiment of the present invention.

[0215] Referring to FIG. 5, the recommendation module can recommend an optimal layout and a recommended layout that reflects the administrator's preference, and the interface module can display the recommended information to the administrator.

[0216] When there are multiple objective functions, three solutions for the optimal allocation can be calculated as the first, second, and third plans, which are methods for minimizing each objective function.

[0217] Here, the selection evaluation value may be 40% stability and 60% efficiency, the second plan's layout evaluation value may be 41% stability and 59% efficiency, the third plan's layout evaluation value may be 45% stability and 55% efficiency, and the first plan's layout evaluation value may be 35% stability and 65% efficiency.

[0218] The recommendation module can recommend the second plan having the placement evaluation value closest to the selection popularity value in first place, and if the distance between the selection popularity value and the placement evaluation value is equal, it can recommend the third plan in second place with more emphasis on stability, and can recommend the next-highest-ranking first plan in third place.

[0219] FIG. 6 is an example of displaying the placement of virtual machines according to the selection value in the method for recommending optimal placement of virtual machines according to an embodiment of the present invention.

[0220] The interface module can provide an interface that changes the administrator's preference rating value and re-recommends the optimal placement in order of placement evaluation values ​​similar to the preference rating value according to a predetermined recommendation method in response to the change in preference rating value.

[0221] The interface module can change the preference value within a range from a minimum to a maximum, and can calculate an interface that reflects the changed preference value and recommends placement of virtual machines on a physical server.

[0222] FIG. 6 is a diagram showing an image screen on which the calculated selection value can be changed within a minimum and maximum range.

[0223] The administrator can change the efficiency and stability of the selection reputation value within a range of 0 to 100 (T11), and the recommendation module can recalculate the recommendation information to re-recommend the recommendation ranking from the optimal placement in real time according to the predetermined recommendation method based on the changed selection reputation value.

[0224] This allows the administrator to change the initial selection if there is any doubt about its suitability and recommend the optimal virtual machine placement.

[0225] FIG. 7 is a flowchart of a server operation method according to an embodiment of the present invention.

[0226] Referring to FIG. 7, the monitoring module may monitor the virtual machine in real time to determine whether the virtual machine and / or the physical machine is operating abnormally.

[0227] If the monitoring module determines that a physical machine and / or virtual machine is operating abnormally, the migration module can determine whether migration is necessary to resolve the issue.

[0228] In one example, the abnormal behavior may be an undersized / oversized virtual machine issue.

[0229] However, the abnormal operation is not limited to this, and can be modified in various ways at a level that is obvious to a person skilled in the art.

[0230] If the migration module determines that migration is necessary, the migration module can determine the physical machine to which the virtual machine will be migrated to resolve the issue and calculate the expected placement of the virtual machine.

[0231] The migration module may transmit the changed virtual machine placement to the virtual machine placement assessment device and request that placement assessment values ​​in terms of stability and efficiency be calculated.

[0232] The placement evaluation value may be calculated by comparing the placement of the current virtual machine with the placement of the future virtual machine to be migrated.

[0233] The migration module determines whether to finally migrate based on the placement evaluation value, and if the placement evaluation value meets a predetermined evaluation criterion, the migration module can proceed with the migration.

[0234] As an example, the predetermined evaluation criterion may be a condition that stability or efficiency is equal to or greater than a predetermined standard (20%).

[0235] However, the predetermined evaluation criteria are not limited to this, and can be variously modified at a level that is obvious to ordinary engineers.

[0236] A virtual machine placement evaluation method according to one embodiment of the present invention is a virtual machine placement evaluation method for calculating a placement evaluation value, which is an evaluation of the placement of virtual machines operated on a physical server, using a virtual machine placement evaluation device, and includes a step of collecting operation information, which is information generated while virtual machines are running, by a receiving module; calculating an efficiency evaluation value, which is a value for the efficiency of the placement of virtual machines based on the operation information, by a first calculation module; calculating a stability evaluation value, which is a value for the stability of the placement of virtual machines based on the operation information, by a second calculation module; and calculating the placement evaluation value using the efficiency evaluation value and the stability evaluation value by a placement evaluation module.

[0237] The virtual machine placement evaluation device can evaluate the placement of virtual machines in terms of stability and efficiency according to the situation and transmit a placement evaluation value when requested by the virtual machine optimal placement recommendation device and the server management device.

[0238] For this purpose, it is possible to receive information required for evaluation from the server management device and / or the virtual machine optimal placement recommendation device.

[0239] In the accompanying drawings, in order to more clearly express the technical idea of ​​the present invention, components that are not relevant to or irrelevant to the technical idea of ​​the present invention are simply depicted or omitted. Although the configuration and features of the present invention have been described above based on the embodiments of the present invention, the present invention is not limited thereto, and it will be apparent to those skilled in the art to which the present invention pertains that various modifications and variations can be made within the spirit and scope of the present invention, and therefore such modifications and variations are within the scope of the appended claims.

Claims

1. A virtual machine optimal placement recommendation device for optimally placing virtual machines running on a physical server, comprising: a collection module that collects operational information generated while an administrator is operating a physical server; a preference evaluation module that calculates a preference value related to the manager's management preference by a predetermined preference calculation method based on the operation information; a recommendation module that calculates recommendation information that reflects the selected popularity value and recommends the placement of virtual machines on the physical server to an administrator. A virtual machine optimal placement recommendation device characterized by:

2. and a placement calculation module that calculates an optimal placement, which is a placement of virtual machines on a physical server that satisfies predetermined placement conditions, The placement calculation module: a placement evaluation module receives a placement evaluation value obtained by evaluating the optimal placement using a predetermined evaluation method; The recommendation module: According to a predetermined recommendation method, the optimal placements are sorted in order of similarity of the placement evaluation value to the selection popularity value, and a placement of the virtual machine to be recommended is selected. The virtual machine optimal placement recommendation device according to claim 1 .

3. The selection rating value and the placement rating value are Calculations are made taking into account stability and efficiency. The virtual machine optimal placement recommendation device according to claim 2 .

4. The predetermined preference calculation method includes: A method for classifying an operation pattern of the virtual machine management system of an administrator based on the operation information and calculating the selection value. The virtual machine optimal placement recommendation device according to claim 3 .

5. The predetermined recommendation method includes: The placement evaluation values ​​are compared based on the preference value, and placement of virtual machines is recommended in a similar order. When the similarity between stability and efficiency is equal, the placement of virtual machines is recommended with a weight given to stability. The virtual machine optimal placement recommendation device according to claim 2 .

6. an interface module that calculates an interface for displaying the recommendation information generated by the recommendation module to an administrator; The interface module includes: Calculate the interface so that the optimal placement information and the recommended information are displayed together. The virtual machine optimal placement recommendation device according to claim 1 .

7. The interface module includes: The preference value can be changed within a range from minimum to maximum, and the changed preference value is reflected in calculating an interface for recommending virtual machine placement on a physical server. The virtual machine optimal placement recommendation device according to claim 6 .

8. The collection module includes: If the predetermined amount of operational information is not collected, personal information of the administrator is collected. The predetermined preference calculation method includes: The method for calculating the selection value by considering both the personal information and the action information. The virtual machine optimal placement recommendation device according to claim 2 .

9. a proficiency determination module for determining a manager's proficiency with the virtual machine management system based on the operation information; The predetermined preference calculation method includes: The method of calculating the selection value by changing the weights to which the personal information and the action information are applied based on the operation proficiency. The virtual machine optimal placement recommendation device according to claim 8 .

10. A virtual machine management method for operating a virtual machine that is run on a physical server and that is realized by a virtual machine management system, comprising: collecting operation information generated while an administrator controls the virtual machine management system by a collection module; A step in which a preference evaluation module calculates a preference evaluation value regarding the manager's preference by a predetermined preference calculation method based on the operation information; and selecting, by a recommendation module, an arrangement of virtual machines on a physical server that reflects the selection value and is recommended to an administrator. A method for recommending optimal placement of virtual machines.

Images