Virtual machine optimal configuration recommendation device and server operation system including same

Abstract

The virtual machine optimal configuration recommendation device is characterized in that the virtual machine optimal configuration recommendation device is used for achieving optimal configuration of a virtual machine working in a physical server and comprises a collection module used for collecting information, namely working information, generated when an administrator operates the physical server; the preference evaluation module is used for calculating a preference evaluation value related to the preference of the administrator according to a preset preference degree calculation method based on the work information; and the recommendation module is used for reflecting the preference evaluation value and calculating recommendation information for recommending the virtual machine configuration in the physical server to an administrator.

Description

Technical Field

[0001] The present invention relates to a virtual machine optimal configuration recommendation device for configuring and managing virtual machines in a physical server and a server operating system comprising the device. Background Art

[0002] As internet technology advances, the spatial constraints of computing operations are gradually disappearing, driven by shifts to working from home, for example. This has led to a recent surge in cloud environments. For businesses providing cloud services, managing virtual machines within physical machines is crucial to ensuring uninterrupted service delivery.

[0003] Previously, a recommended solution was to migrate virtual machines from less-used servers to more-used servers in a phased manner based on a first-fit decreasing algorithm to minimize server startups. However, minimizing server startups would overload the remaining servers, leading to instability issues in server stability. Summary of the Invention

[0004] Issues to be addressed

[0005] In order to solve the above-mentioned drawbacks of conventional technologies, the present invention provides a virtual machine optimal configuration recommendation device and a server operation system including the same, which can objectively evaluate and judge the configuration of the virtual machine and, based on this, recommend the optimal virtual machine configuration.

[0006] Problem Solutions

[0007] The technical features of the virtual machine optimal configuration recommendation device described in one embodiment of the present invention are as follows: the virtual machine optimal configuration recommendation device is used to achieve the optimal configuration of the virtual machine working in the physical server, which may include: a collection module, which is used to collect information generated when the administrator operates the physical server, that is, work information; a preference evaluation module, which calculates the preference evaluation value related to the administrator's preference based on the work information and a preset preference calculation method; and a recommendation module, which is used to reflect the preference evaluation value and calculate recommendation information for recommending the virtual machine configuration in the physical server to the administrator.

[0008] Furthermore, the system includes a configuration calculation module configured to calculate a virtual machine configuration for the service server that satisfies preset configuration conditions, i.e., an optimal configuration. The configuration calculation module receives a configuration evaluation value of the optimal configuration evaluated by the configuration evaluation module according to a preset evaluation method. The recommendation module then sorts the optimal configurations according to a preset recommendation method, in order of configuration evaluation values that approximate the preferred evaluation value, and selects the virtual machine configuration to be recommended.

[0009] Furthermore, the preference evaluation value and the configuration evaluation value may be calculated taking into account stability and efficiency aspects.

[0010] Furthermore, the preset preference calculation method may be the following method: based on the work information, the operation mode of the administrator's virtual machine management system is divided, and the preference evaluation value is calculated.

[0011] Furthermore, the preset recommendation method may be as follows: based on the preference evaluation value, the configuration evaluation value is compared, and the configuration of the virtual machine is recommended in a similar order; when the stability and efficiency are similar, the weighted value is applied to the stability to recommend the configuration of the virtual machine.

[0012] Furthermore, the system further comprises an interface module for calculating an interface so as to display the recommendation information generated by the recommendation module to the administrator. The interface module can calculate an interface for displaying the optimal configuration information and the recommendation information together.

[0013] Furthermore, the interface module may change the preference evaluation value within a range from a minimum to a maximum, calculate the preference evaluation value reflecting the change, and recommend an interface for the virtual machine configuration in the physical server.

[0014] Furthermore, when the amount of work information collected is less than a preset amount, the collection module will collect the administrator's personal information. The preset preference calculation method may be a method that considers both the personal information and the work information to calculate the preference evaluation value.

[0015] Furthermore, the system further includes a proficiency determination module that determines the administrator's proficiency in operating the virtual machine management system based on the work information. The preset preference calculation method may be a method that modifies the weighted values of the personal information and the work information based on the operation proficiency to calculate the preference evaluation value.

[0016] The technical features of the method for recommending the optimal configuration of a virtual machine described in one embodiment of the present invention are as follows: the method for recommending the optimal configuration of a virtual machine is implemented through a virtual machine management system to operate virtual machines working in a physical server. The method for recommending the optimal configuration of a virtual machine includes the following steps: collecting information generated when the administrator controls the virtual machine management system, that is, work information, through a collection module; calculating a preference evaluation value related to the administrator's preference based on the work information and a preset preference calculation method through a preference evaluation module; and reflecting the preference evaluation value through a recommendation module to select a virtual machine configuration in the physical server to be recommended to the administrator.

[0017] The technical features of the virtual machine configuration evaluation device described in one embodiment of the present invention are as follows: the configuration evaluation value is an evaluation of the configuration of the virtual machine operating in the physical server. The virtual machine configuration evaluation device for calculating the configuration evaluation value may include: a receiving module, which is used to collect information generated when the virtual machine is started, that is, operation information; a first calculation module, which calculates the value of the virtual machine configuration efficiency based on the operation information, that is, the efficiency evaluation value; a second calculation module, which calculates the value of the virtual machine configuration stability based on the operation information, that is, the stability evaluation value; and a configuration evaluation module, which applies the efficiency evaluation value and the stability evaluation value to calculate the configuration evaluation value.

[0018] Furthermore, the first calculation module may calculate the comprehensive power consumption based on the CPU startup rate of the physical server, thereby calculating the efficiency evaluation value.

[0019] Furthermore, the first calculation module may calculate the comprehensive power by summing up the startup power and the mobile power, wherein the startup power is the power generated when the virtual machine is working, and the mobile power is the power generated when the virtual machine is migrating.

[0020] Furthermore, the first calculation module calculates the startup power by summing up the basic power and the additional power, wherein the basic power is the power generated when the CPU is in an idle state, and the additional power is the power generated when the CPU is started to work for the virtual machine.

[0021] Furthermore, the first calculation module may calculate an approximate basic power amount by dividing the power amount generated when the CPU is fully started by a given value.

[0022] Furthermore, the first calculation module may calculate the additional power based on the difference between the frequency when the CPU is fully started and the frequency when the CPU is in an idle state, taking the ratio of the CPU startup frequencies as the standard.

[0023] Furthermore, the first calculation module may calculate an approximate mobile power amount according to a preset ratio of power amount generated when the CPU is fully started.

[0024] Furthermore, the first calculation module may calculate the mobile power based on a difference between a frequency when the CPU is fully started and a frequency when the CPU is in an idle state, using a frequency ratio calculated according to a preset ratio of the frequency when the CPU is fully started.

[0025] Furthermore, the second calculation module may calculate the stability evaluation value based on the dirty memory ratio and the network link ratio.

[0026] The technical features of the virtual machine configuration evaluation method described in one embodiment of the present invention are as follows: the virtual machine configuration evaluation method applies a virtual machine configuration evaluation device to calculate the configuration evaluation of the virtual machine operating in the physical server, that is, the configuration evaluation value, which may include the following steps: through a receiving module, collecting information generated when the virtual machine is started, that is, the operation information; through a first computing module, based on the operation information, calculating the virtual machine configuration efficiency value, that is, the efficiency evaluation value; through a second computing module, based on the operation information, calculating the virtual machine configuration stability value, that is, the stability evaluation value; and through a configuration evaluation module, applying the efficiency evaluation value and the stability evaluation value to calculate the configuration evaluation value.

[0027] Effects of the Invention

[0028] According to the present invention, the virtual machine optimal configuration recommendation device and the server operating system including the same can maximize the stability of the server.

[0029] Furthermore, server management can be achieved with minimal personnel.

[0030] Furthermore, server management can be time-consuming and minimized.

[0031] However, the effects of the present invention are not limited to the aforementioned effects, and those skilled in the art can clearly understand the effects not mentioned from this specification and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a relationship diagram of the server operation system described in one embodiment of the present invention.

[0033] Figure 2 It is a structural diagram of a server operating system according to an embodiment of the present invention.

[0034] Figure 3 This is a flowchart of a method for recommending optimal configuration of a virtual machine according to an embodiment of the present invention.

[0035] Figure 4 This is a schematic diagram of the weighted values of personal information and work information used in the method for recommending optimal virtual machine configurations according to an embodiment of the present invention.

[0036] Figure 5 An image example of a recommended virtual machine configuration in the method for recommending optimal virtual machine configuration according to an embodiment of the present invention is shown.

[0037] Figure 6 An example of representing virtual machine configurations based on preference evaluation values in a method for recommending optimal virtual machine configurations according to an embodiment of the present invention is shown.

[0038] Figure 7It is a flow chart of a server operation method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] Specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. However, the concept of the present invention is not limited to the embodiments described. Those skilled in the art who understand the concept of the present invention may add, modify, or delete components within the scope of the same concept and easily propose other inventions or other embodiments within the scope of the present invention. Such additions, modifications, and deletions should also be considered to be within the scope of the present invention.

[0040] Furthermore, components having the same function and falling within the scope of the same concept shown in the drawings of the respective embodiments are described using the same reference symbols.

[0041] Figure 1 It is a relationship diagram of the server operation system described in one embodiment of the present invention.

[0042] like Figure 1 As shown, the server operating system 100 described in an embodiment of the present invention may be a system for operating, maintaining, migrating, etc. virtual machines set up on a physical server.

[0043] In a specific example, the server operation system 100 can monitor whether a problem occurs with a virtual machine, and when a problem occurs with a virtual machine, can resolve the problem. Furthermore, the server operation system can optimally configure the virtual machines of the physical server from the perspective of application resources.

[0044] The server operation system 100 can be connected to the physical server 200 and / or the external server 300 via a wired / wireless network. The server operation system can collect and receive all information generated when starting a virtual machine in the physical server, all information generated when starting the physical server, and / or required information from the external server.

[0045] In the present invention, the network can be a core network integrated with a wired public network, a wireless mobile communication network or a portable Internet, and can mean a global open computer network structure that provides a variety of services of the TCP / IP protocol and its upper layers, namely, Hypertext Transfer Protocol (HTTP), Hypertext Transfer Protocol Secure (HTTPS), Telnet, File Transfer Protocol (FTP), Domain Name System (DN S), Simple Mail Transfer Protocol (SMTP), etc. It is not limited to the above examples and in a broad sense means a data communication network that can send and receive data in various forms.

[0046] The physical server of the present invention may include other components of the server environment for executing the server. The server may include all devices in any form.

[0047] In one example, the server as a digital device can be a digital device equipped with a processor, having a memory, and having computing capabilities, such as a laptop, a notebook computer, a desktop computer, a wireless Internet access device, and a mobile phone.

[0048] In one example, the server may be a web server. However, the server is not limited thereto, and a person skilled in the art can generally make various changes to the type of server at an obvious level.

[0049] Figure 2 It is a structural diagram of a server operating system according to an embodiment of the present invention.

[0050] like Figure 2 As shown, the server operating system 100 according to an embodiment of the present invention may include a virtual machine optimal configuration recommendation device 110 , a server management device 120 , and a virtual machine configuration evaluation device 130 .

[0051] Each device is connected via a wired / wireless network to send and receive the required information.

[0052] The server management device 120 can perform all the processes required to operate a physical server.

[0053] The server management device 120 may include: a storage module 121, which stores information required for managing and operating physical servers; a monitoring module 122, which is used to monitor in real time whether virtual machines started in physical servers are started abnormally; and a migration module 123, which enables virtual machines to move between physical servers.

[0054] Furthermore, the server management device 120 may further include an input module 124 for receiving device control input from an administrator, generating a control signal, and transmitting the control signal to the virtual machine optimal configuration recommendation device 110 , the server management device 120 , and the virtual machine configuration evaluation device 130 .

[0055] In addition, the server management device 120 may further include a display module 125, which is used to display the values calculated by the virtual machine optimal configuration recommendation device 110, the server management device 120 and the virtual machine configuration evaluation device 130, and information required for the operation and management of the server management device, such as the status of the virtual machine, the status of the physical server, etc.

[0056] The storage module 121 can store all data required for the server management device 120 to operate.

[0057] In one example, the storage module may include an internal memory and / or an external memory.

[0058] In one example, the internal memory may include at least one of a volatile memory (e.g., DRAM, SRAM, or SDRAM) and a non-volatile memory (e.g., OTPROM (one time programmable ROM), PROM, EPROM, EEPROM, mask ROM, flash ROM, flash memory, hard disk, or solid state drive (SSD)).

[0059] The external memory may include a flash drive, such as a compact flash card (CF), a secure digital card (SD), a Micro-SD, a Mini-SD, an extreme digital card (xD), a multi-media card (MMC), or a memory stick.

[0060] The monitoring module 122 can receive logarithmic data and metric data generated by the virtual machine, and monitor in real time whether there is abnormal operation, including whether the virtual machine is underloaded or overloaded.

[0061] When a problem occurs on a physical server, the migration module 123 may move the virtual machine between physical servers based on a control signal from an administrator.

[0062] The migration module 123 can determine whether migration is necessary or appropriate before migrating the virtual machine, so that the physical server accepts the migration.

[0063] The administrator can input commands and command signals for managing and operating the server operating system 100 through the input module 124 .

[0064] In one example, the input module may include a mouse, keyboard, touch panel, (digital) pen sensor, key or ultrasonic input device, etc. For example, the touch panel may adopt at least one of electrostatic, decompression, ultraviolet or ultrasonic methods.

[0065] Furthermore, the touch panel may further include a control circuit. The touch panel may further include a tactile layer to provide a user with a tactile response. For example, a (digital) pen sensor may include a portion of the touch panel or an additional recognition graphic.

[0066] Also, for example, the key may include a physical key, an optical key, or a keypad.

[0067] Furthermore, the ultrasonic input device can sense the ultrasonic waves generated in the input tool through the microphone and confirm data corresponding to the sensed ultrasonic waves.

[0068] The display module 125 may include all devices capable of displaying images such as a display device, a display screen, a projector, and the like.

[0069] In one example, a display (eg, a display device) may include a panel, a hologram device, a projector, or a control circuit for controlling the same.

[0070] The technical features of the virtual machine optimal configuration recommendation device 110 described in one embodiment of the present invention are as follows: the virtual machine optimal configuration recommendation device 110 enables the virtual machines working in the physical server to achieve the optimal configuration, which may include: a collection module 111, which is used to collect information generated when the administrator operates the physical server, that is, work information; a preference evaluation module 112, which calculates the preference evaluation value related to the administrator's management preference based on the work information according to a preset preference calculation method; and a recommendation module 113, which reflects the preference evaluation value and selects the virtual machine configuration in the physical server to be recommended to the administrator.

[0071] Furthermore, the virtual machine optimal configuration recommendation device 110 may further include a configuration calculation module 114 for calculating an optimal configuration, where the optimal configuration is a configuration of the virtual machine in the physical server that meets a preset configuration condition.

[0072] In addition, the virtual machine optimal configuration recommendation device 110 may further include: an interface module 115, which is used to calculate an interface so as to display the recommendation information generated by the recommendation module 113 to the administrator; and a proficiency judgment module 116, which judges the administrator's proficiency in operating the virtual machine management system based on the work information.

[0073] The collection module 111 can collect information generated when an administrator manages a physical server, that is, work information.

[0074] The work information may include logarithmic information and / or measurement data of the physical server / virtual machine, and may include control data used by the administrator to control the physical server and all information about the current startup status of the physical server and virtual machine when the control data is generated.

[0075] In one example, the work information may include: in the current state of the physical server and the virtual machine, the administrator controls the physical server and the virtual machine and performs several control methods in order to resolve abnormal situations.

[0076] When the amount of work information collected is less than a preset amount, the collection module 111 will collect the administrator's personal information.

[0077] In one example, the preset collection volume may be 1 TB. However, this is not a limitation, and a person skilled in the art can generally make various modifications to the preset collection volume at an obvious level.

[0078] The personal information may include information that is not related to the server operating system 100 and is generated when the administrator controls the electronic device from the outside, as well as the administrator's identity information.

[0079] In view of this, the collection module 111 can establish a network connection with an external server, a public agency server, an open server, etc. to collect the required information.

[0080] For this purpose, pre-approval from the administrator can be obtained.

[0081] In one example, the personal information may include the administrator's navigation path selection information.

[0082] When exploring navigation paths, multiple paths can be recommended, and navigation users can choose a path from the multiple paths based on their own tendencies.

[0083] For this reason, when the information is used by the administrator's navigation application, it is helpful to grasp the administrator's preference trend.

[0084] In one example, personal information may include an administrator's car control information.

[0085] Based on the recent expansion of IOT technology and the trend of electronic instrumentation in automobiles, various control information of automobiles is being collected. An administrator can control the automobile and the collection module 111 collects the control information from the automobile manufacturer's server and / or the communication company's server.

[0086] How a car is controlled can be important information when understanding whether the operator prioritizes stability or efficiency.

[0087] The vehicle control information is integrated with the navigation information in the vehicle, which is important information when understanding the traffic situation and how the administrator controls the vehicle at the site.

[0088] In one example, the personal information may include information about the administrator's age, gender, school he graduated from, major information, and place of birth.

[0089] The older people get, the more efficient they become and the more they value stability.

[0090] Compared with men, women value stability more than efficiency.

[0091] The university atmosphere will have an impact on students, so the emphasis on efficiency and stability will vary depending on the school you graduate from.

[0092] The major you study will have an impact on the students, so the emphasis on efficiency and stability will vary depending on the major.

[0093] The birthplace of an urban area, a rural area, a fishing village, etc. will have an impact on the residents, and therefore the emphasis on efficiency and stability will be different.

[0094] For example, people who live in urban areas tend to value stability more and prioritize efficiency more. On the other hand, people who live in rural / fishing areas tend to value efficiency more and prioritize stability more.

[0095] The preference evaluation module 112 may calculate a preference evaluation value related to the administrator's management preference according to a preset preference calculation method.

[0096] The preference evaluation module 112 may store a mode classification model, which divides the work information into preset operation modes.

[0097] The preference evaluation module 112 may perform deep learning based on the classification information of the work information corresponding to the historical work information to generate a pattern classification model.

[0098] Deep learning may utilize a back propagation algorithm, which utilizes labeled data of an output layer to update weights of a neural network. However, this is not a limitation.

[0099] Moreover, the deep neural network and back propagation algorithm are the same as those known in conventional technology, and therefore, their detailed description is omitted here.

[0100] In one example, the preference evaluation value may be composed of efficiency and stability values, with the unit being %, where the sum of efficiency and stability is 100.

[0101] However, this is not limiting here, and generally, technicians can make various modifications to the form of the preference evaluation value at an obvious level.

[0102] The operating models can be divided into different modes from the perspective of stability and efficiency.

[0103] For example, the first operating mode can be a mode with 0% stability and 100% efficiency, and the second operating mode can be a mode with 2% stability and 98% efficiency. In this way, several operating modes can be divided based on all stability and efficiency levels.

[0104] The preset preference calculation method may be a method of calculating the preference evaluation value by dividing the operation mode of the virtual machine management system of the administrator based on the work information.

[0105] The preference evaluation value can be calculated by considering the stability and efficiency aspects.

[0106] The preference evaluation module 112 can be used to effectively analyze the management tendency of the administrator.

[0107] The preset preference calculation method may be a method of calculating the preference evaluation value by considering both the personal information and the work information.

[0108] When the amount of collected work information is less than a predetermined amount, the preference evaluation module 112 may use the personal information to calculate a preference evaluation value.

[0109] The preference evaluation module 112 may store a personal computing model, which may calculate efficiency and stability based on personal information.

[0110] The preference evaluation module 112 performs deep learning on the personal information corresponding to the historical personal information based on the values divided into efficiency and stability, and calculates a personal calculation model.

[0111] The deep learning algorithm utilizes well-known technologies, and a detailed description thereof may be omitted here.

[0112] The preference evaluation module 112 may perform a weighted average of the first preference evaluation value calculated by the pattern classification model and the second preference evaluation value calculated by the personal calculation model to finally calculate the preference evaluation value.

[0113] The preference evaluation module 112 may modify the weighted values of the first preference evaluation value and the second preference evaluation value according to the amount of work information collected, until the amount of work information collected reaches a preset amount, and calculate the preference evaluation value.

[0114] In one example, as the amount of work information collected increases, the weighted value may be applied to the first preference evaluation value to calculate an average and a final preference evaluation value may be calculated compared to the second preference evaluation value.

[0115] When the job information is too little, the first preference evaluation value may not fully reflect the administrator's tendency and preference.

[0116] To remedy this, personal information that has accumulated a considerable amount of data may be used to apply a more weighted value to the second preference evaluation value to calculate a final preference evaluation value.

[0117] As the amount of work information gradually increases, the weight of the first preference evaluation value is increased at a first ratio, and the second preference evaluation value calculated using the personal information is decreased.

[0118] The preset preference calculation method may be a method for calculating the preference evaluation value by modifying the weighted values of the personal information and the work information based on the operation proficiency.

[0119] In other words, the preset preference calculation method can be as follows: when the amount of work information collected is greater than a given benchmark value, based on the operation proficiency, the ratio at which the weighted value of the second preference evaluation value decreases as the weighted value of the first preference evaluation value increases is used as a ratio different from the first ratio, and the weighted value is changed.

[0120] In a specific example, when the amount of work information collected is greater than a given benchmark value (change benchmark value) and the administrator's operational proficiency is greater than a given proficiency, the preference evaluation module uses the ratio of the decrease in the weighted value of the second preference evaluation value as the weighted value of the first preference evaluation value increases as a ratio greater than the first ratio, that is, the second ratio, to calculate the final preference evaluation value.

[0121] Furthermore, when the amount of work information collected is greater than a given benchmark value and the administrator's operational proficiency is lower than a given proficiency, the preference evaluation module can calculate the final preference evaluation value by using the ratio at which the weighted value of the second preference evaluation value decreases as the weighted value of the first preference evaluation value increases as a ratio less than the first ratio, that is, the third ratio.

[0122] The reason is that when the administrator's proficiency is low, the consistency of the work information is low, which leads to lower information reliability. Conversely, when the administrator's proficiency is high, the consistency of the work information is high, which leads to higher information reliability.

[0123] When the proficiency determination module 116 cannot calculate the operation proficiency, the preference evaluation module may maintain the first ratio and calculate the final preference evaluation value.

[0124] The proficiency determination module 116 can calculate the administrator's proficiency in operating the server operating system 100 based on the work information.

[0125] To this end, the proficiency determination module 116 may store a proficiency determination model.

[0126] The proficiency determination module 116 may generate a proficiency determination model by performing deep learning on the operation methods of the physical servers and virtual machines and their evaluation scores based on the historical status of the physical servers and virtual machines.

[0127] The specific algorithm of deep learning is an application of well-known technology, and its detailed description will be omitted here.

[0128] The proficiency judgment model can calculate the administrator's operational proficiency.

[0129] However, the proficiency judgment model only needs the control information required to judge the operation proficiency and the status information of the physical server and virtual machine at the time of control in the work information. Therefore, even if there is a lot of work information, it cannot be assumed that the information required to judge the operation proficiency is also a lot.

[0130] The proficiency judgment model can calculate the operation proficiency only when the amount of data required to judge the operation proficiency is greater than the given benchmark amount.

[0131] The amount of data required to determine the operation proficiency may be smaller than the change reference value.

[0132] Its purpose is to implement the above algorithm efficiently.

[0133] The configuration calculation module 114 can calculate the optimal configuration of the virtual machine.

[0134] To this end, the configuration calculation module 114 may predict the future workload of the virtual machine and, based on the predicted workload, calculate the configuration of the virtual machine in the physical server that minimizes the preset objective function.

[0135] The method for configuring the calculation module 114 to calculate the optimal configuration of the virtual machine may be a well-known technique, and a detailed description thereof will be omitted here.

[0136] The configuration calculation module 114 may request the configuration evaluation module 134 described below to evaluate the calculated optimal configuration.

[0137] The configuration calculation module 114 may receive a configuration evaluation value for evaluating the optimal configuration through an evaluation method preset by the configuration evaluation module 134 .

[0138] The configuration calculation module 114 may transmit the configuration evaluation value and the optimal configuration transmitted by the configuration evaluation module 134 to the recommendation module 113 .

[0139] The configuration evaluation value can be in a form that takes into account stability and efficiency.

[0140] In one example, expressing stability and efficiency as percentages with an overall value of 100 allows for easy comparison of configuration evaluations and (ultimate) preference evaluations with each other.

[0141] The recommendation module 113 may classify the optimal configurations according to a preset recommendation method in the order of the configuration evaluation values that are similar to the preference evaluation values, and select the configuration of the virtual machine to be recommended.

[0142] In order to reflect the administrator's preference in the optimal configuration and recommend the virtual machine configuration, the recommendation module 113 may compare the preference evaluation value and the configuration evaluation value with each other.

[0143] The recommendation module 113 may calculate recommendation information for recommending an optimal configuration recommendation order in the order of configuration evaluation values having the most similar preference evaluation values among the optimal configurations.

[0144] The preset recommendation method can be as follows: based on the preference evaluation value, the configuration evaluation values are compared, and the configuration of the virtual machine is recommended in a similar order. When the approximate degree of stability and efficiency is the same, a weighted value is applied to the stability to recommend the configuration of the virtual machine.

[0145] In other words, in the configuration of a virtual machine, when the distance between the preference evaluation value and the configuration evaluation value is the same, more emphasis will be placed on stability, and the virtual machine configuration with higher stability will be recommended in a higher order.

[0146] The interface module 115 can calculate the interface so as to display the optimal configuration information and the recommended information together.

[0147] The display module 125 displays the interface calculated by the interface module 115 , and the administrator can easily view the information required to manage and operate the server operation system 100 .

[0148] The technical features of the virtual machine configuration evaluation device 130 described in one embodiment of the present invention are as follows: the configuration evaluation value is an evaluation of the configuration of the virtual machine operating in the physical server. The virtual machine configuration evaluation device 130 for calculating the configuration evaluation value may include: a receiving module 131, which is used to collect information generated when the virtual machine is started, that is, operation information; a first calculation module 132, which calculates the value of the virtual machine configuration efficiency based on the operation information, that is, the efficiency evaluation value; a second calculation module 133, which calculates the value of the virtual machine configuration stability based on the operation information, that is, the stability evaluation value; and a configuration evaluation module 134, which applies the efficiency evaluation value and the stability evaluation value to calculate the configuration evaluation value.

[0149] The receiving module 131 can collect information generated when the virtual machine is started, that is, operation information.

[0150] Operational information may include hardware working information of the physical server based on the virtual machine.

[0151] In one example, the operation information may include CPU startup information, memory working information, network information, etc. The above example does not limit the present invention. Generally, technicians can make various modifications to the specific example of the operation information at an obvious level.

[0152] The first calculation module can calculate an efficiency evaluation value.

[0153] The first calculation module 132 may calculate the comprehensive power consumption based on the CPU activation rate of the physical server, thereby calculating the efficiency evaluation value.

[0154] In one example, the greater the total power, the greater the efficiency evaluation value becomes, and the smaller the total power, the smaller the efficiency evaluation value becomes.

[0155] In one example, the CPU activation rate, as a unit of CPU frequency, may be expressed in MHZ.

[0156] The first calculation module 132 can calculate the comprehensive power by summing up the startup power and the mobile power, where the startup power is the power generated by the physical server when the virtual machine is working, and the mobile power is the power generated when the virtual machine migrates.

[0157] E total =E host +E migration (Formula 1)

[0158] Among them, E total Means comprehensive electricity (MW), E host Refers to starting power (MW), E migration It refers to the amount of mobile electricity (MW).

[0159] That is, the total power is calculated by summing up the power generated by the migration and the power generated by the physical server due to the virtual machine configuration after the migration.

[0160] The first calculation module 132 calculates the startup power by summing up the power generated when the CPU is in an idle state, that is, the basic power and the additional power generated when the virtual machine is working and the CPU is started.

[0161] Furthermore, the first calculation module 132 divides the power consumption generated when the CPU is fully started by a given value to calculate an approximate basic power consumption.

[0162] Furthermore, the first calculation module 132 may calculate the additional power based on the difference between the frequency when the CPU is fully started and the frequency when the CPU is in an idle state, taking the ratio of the CPU startup frequencies as a standard.

[0163]

[0164] Among them, TDP model It can mean the amount of power generated by a physical server when the CPU startup rate is 100%. Can mean basic power.

[0165] Directly measuring various test data and the power generated by physical servers, the result shows that the more physical servers are started, the closer the basic power consumption is to TDP. model half.

[0166] in, Can mean additional power.

[0167] and, It can mean the average of all CPU activation rates. The additional power is calculated based on the difference between the CPU's fully driven frequency and the CPU's idle frequency, using the CPU's activation frequency ratio.

[0168] The CPU startup frequency may refer to the CPU frequency generated when a virtual machine such as an application program of a physical server performs a purposeful operation.

[0169] That is, base freq -idle freq Can mean CPU startup frequency.

[0170] And, base freq It can mean the frequency generated when the CPU is started at 100%, idle freq It can mean that the physical machine is not doing any work for you, and you are only increasing the CPU frequency when the power is turned on.

[0171] Furthermore, in Formula 1-2, It can mean the additional power generated when the CPU is running at 100%.

[0172] The first calculation module 132 can calculate an approximate mobile power according to a preset ratio based on the power generated when the CPU is fully driven.

[0173] Specifically, the first calculation module 132 may calculate the mobile power based on the difference between the frequency when the CPU is fully started and the frequency when the CPU is in an idle state, using a frequency ratio calculated according to a preset ratio of the frequency when the CPU is fully started.

[0174]

[0175] Among them, W x Can mean a preset ratio.

[0176] You can change the preset ratio based on the performance of the physical machine, the size of the migrated virtual machine, the type of virtual machine, etc.

[0177] In one example, the higher the performance of the physical machine, the smaller the preset ratio becomes.

[0178] In one example, the larger the size of the migrated virtual machine is, the larger the preset ratio will be.

[0179] In one example, the predetermined ratio may be 0.4 (40%).

[0180] However, without being limited thereto, a person skilled in the art can generally make various modifications to the preset ratio at an obvious level.

[0181] Furthermore, in formula 1-3, It can mean the additional power generated when the CPU is running at 100%.

[0182] E can be calculated according to different physical servers total .

[0183] The second calculation module 133 may calculate the stability evaluation value based on the dirty memory ratio and the network link ratio.

[0184] As a specific example, the second calculation module 133 compares the memory of the virtual machine, the dirty memory ratio of the virtual machine, and the network connection rate to calculate the time required for migration.

[0185] The second computing module 133 can compare the size of the physical server and the size of the virtual machine, and evaluate and classify the required migration time into five steps (Safety, Stable, Normal, Warning, Danger) based on the relative stability value.

[0186] Among them, the stable evaluation value of the Safety level can be greater than the Danger level.

[0187] The larger the VM's memory, the larger the VM's dirty memory ratio, and the smaller the network link rate, the smaller the stability assessment value.

[0188] On the contrary, the smaller the virtual machine's memory, the smaller the virtual machine's dirty memory ratio, and the larger the network link rate, the larger the stability evaluation value will be.

[0189] The configuration evaluation module 134 may calculate a configuration evaluation value by applying the efficiency evaluation value and the stability evaluation value.

[0190] The configuration evaluation module 134 can calculate the efficiency evaluation value and stability evaluation value of the physical machine at the time of evaluation, compare the efficiency evaluation value and stability evaluation value when changing the virtual machine configuration with the optimal configuration sent by the virtual machine optimal configuration recommendation device 110, and calculate the final configuration evaluation value.

[0191] The configuration evaluation module 134 calculates the efficiency evaluation value and stability evaluation value of the physical machine at the time of evaluation, compares the efficiency evaluation value and stability evaluation value when changing the virtual machine skimming with the modified configuration sent by the server management device 120, and calculates the final configuration evaluation value.

[0192] When the efficiency evaluation value of the configuration to be changed (optimal configuration or modified configuration) is greater than the efficiency evaluation value of the current configuration, the efficiency will be higher.

[0193] On the contrary, if the efficiency evaluation value of the configuration to be changed is smaller than the efficiency evaluation value of the current configuration, the efficiency will become smaller.

[0194] When the stability evaluation value of the configuration to be changed is greater than the stability evaluation value of the current configuration, the stability will be greater.

[0195] On the contrary, if the stability evaluation value of the configuration to be changed is smaller than the stability evaluation value of the current configuration, the stability will become smaller.

[0196] The configuration evaluation value may be composed of efficiency and stability values, with the unit being %, and the sum of efficiency and stability being 100.

[0197] To this end, the values of the efficiency evaluation value and the configuration evaluation value may be processed.

[0198] In one example, using a preset processing table, the difference between the stability evaluation value of each current configuration and the stability evaluation value of the configuration to be changed is divided into the difference between the efficiency evaluation value of the current configuration and the efficiency evaluation value of the configuration to be changed, and efficiency and stability are pre-specified in the classification of the difference between each efficiency evaluation value of the current configuration and the efficiency evaluation value of the configuration to be changed.

[0199] However, without being limited thereto, a person skilled in the art can generally make various modifications to the method of processing efficiency evaluation value and configuration evaluation value at an obvious level.

[0200] While fundamental frequencies are easy to detect, idle frequencies are not. This is because, in actual field conditions, measurements are not always performed with the server powered on and not in operation.

[0201] Figure 3 This is a flowchart of a method for recommending optimal configuration of a virtual machine according to an embodiment of the present invention.

[0202] Hereinafter, detailed description will be omitted without repeating the above contents.

[0203] like Figure 3 As shown, the technical features of the optimal configuration recommendation method described in one embodiment of the present invention are as follows: the optimal configuration recommendation method for virtual machines is implemented through a virtual machine management system to operate virtual machines working in physical servers. The optimal configuration recommendation method for virtual machines includes the following steps: collecting information generated when the administrator controls the virtual machine management system, that is, work information, through a collection module; calculating a preference evaluation value related to the administrator's preference based on the work information and a preset preference calculation method through a preference evaluation module; and reflecting the preference evaluation value through a recommendation module to select a virtual machine configuration in the physical server to be recommended to the administrator.

[0204] You can collect work information through the collection module.

[0205] When the amount of work information collected is less than the preset amount, the administrator's personal information may be collected to determine proficiency, and the personal information and proficiency may be used to calculate the preference evaluation value.

[0206] When the amount of work information collected is equal to a preset amount of collection, the preference evaluation index can be calculated based on only the work information.

[0207] The preference evaluation value may be continuously updated until the administrator requests a recommended virtual machine configuration through the input module.

[0208] When the administrator requests a recommendation for a virtual machine configuration, the optimal configuration can be calculated by the configuration calculation module, and the configuration calculation module can request the virtual machine configuration evaluation device to perform configuration evaluation on the optimal configuration.

[0209] The virtual machine configuration evaluation device can compare the current virtual machine configuration with the virtual machine configuration when it is configured to the optimal configuration, and calculate a configuration evaluation value.

[0210] The recommendation module can compare the preference evaluation values and the configuration evaluation values with each other, calculate the optimal configuration in a similar order, and the one that is suitable for the administrator's operation trend and mode, and recommend the virtual machine configuration to the administrator through the interface module and the display module.

[0211] Figure 4 This is a schematic diagram of the weighted values of personal information and work information used in the method for recommending optimal virtual machine configurations according to an embodiment of the present invention.

[0212] like Figure 4 As shown, for example, when the operation information on the X-axis is applicable to 100%, it means that the collection amount of operation information is equivalent to the preset collection amount.

[0213] When no job information is collected, only personal information is used and the preference evaluation value is calculated using 100%.

[0214] As the amount of work information gradually increases, the weight of the second preference evaluation value calculated using the personal information decreases according to the first ratio for the increased amount.

[0215] However, the present invention is not limited thereto, and a person skilled in the art can generally make various modifications to the first ratio at an obvious level.

[0216] Figure 4 (a) shows that when the operation proficiency is higher than a given proficiency, Figure 4 As shown in (a), the given reference value X10 of the work information collection amount may be half of the preset collection amount.

[0217] However, without being limited thereto, a person skilled in the art can generally make various modifications to a given reference value at an obvious level.

[0218] When the amount of work information collected is less than the given benchmark value X10, the first preference evaluation value and the second preference evaluation value are adjusted according to the first ratio B11 to calculate the final preference evaluation value. However, when the amount of work information collected is greater than the given benchmark value, the first preference evaluation value and the second preference evaluation value can be adjusted according to the second ratio B12 to calculate the final preference evaluation value.

[0219] Therefore, for example, even if the amount of work information collected is only 75% of the preset amount, the first preference evaluation value may be applied as 100%, and the final preference evaluation value may be calculated based only on the first preference evaluation value.

[0220] Figure 4 (b) shows that when the operation proficiency is less than a given proficiency, Figure 4 As shown in (b), the given reference value X10 of the work information collection amount may be half of the preset collection amount.

[0221] However, without being limited thereto, a person skilled in the art can generally make various modifications to a given reference value at an obvious level.

[0222] When the amount of work information collected is less than the given benchmark value X10, the first preference evaluation value and the second preference evaluation value are adjusted according to the first ratio B11 to calculate the final preference evaluation value. However, when the amount of work information collected is greater than the given benchmark value, the first preference evaluation value and the second preference evaluation value can be adjusted according to the third ratio B13 to calculate the final preference evaluation value.

[0223] Therefore, for example, even if the collection amount of work information is equivalent to 100% of the preset collection amount, the first preference evaluation value is applied at 75% and the second preference evaluation value is applied at 25% to calculate the final preference evaluation value.

[0224] Figure 5 An image example of a recommended virtual machine configuration in the method for recommending optimal virtual machine configuration according to an embodiment of the present invention is shown.

[0225] like Figure 5 As shown, the recommendation module recommends the optimal configuration and the recommended configuration reflecting the administrator's preferences, and the interface module can display the recommended information to the administrator.

[0226] The optimal configuration is taken as the first, second and third options. When the objective function is complex, three solutions can be calculated based on the options that minimize each objective function.

[0227] Among them, the preference evaluation value can be stability 40% and efficiency 60%, the configuration evaluation value of the second scheme can be stability 41% and efficiency 59%, the configuration evaluation value of the third scheme can be stability 45% and efficiency 55%, and the configuration evaluation value of the first scheme can be stability 35% and efficiency 65%.

[0228] The recommendation module may recommend the second option with the configuration evaluation value closest to the preference evaluation value as the first priority. When the distance between the preference evaluation value and the configuration evaluation value is equal, the recommendation module may place more emphasis on stability and recommend the third option as the second priority. The first option of the next priority may be recommended as the third priority.

[0229] Figure 6 An example of representing virtual machine configurations based on preference evaluation values in a method for recommending optimal virtual machine configurations according to an embodiment of the present invention is shown.

[0230] The interface module changes the administrator's preference evaluation value. As the preference evaluation value changes, an interface for re-recommending the optimal configuration can be provided in the order of the configuration evaluation values that are similar to the preference evaluation value based on a preset recommendation method.

[0231] The interface module can change the preference evaluation value within a range from a minimum to a maximum, calculate the preference evaluation value reflecting the change, and recommend an interface for the virtual machine configuration in the physical server.

[0232] Figure 6 A graphical interface showing how to change a calculated preference evaluation value within a minimum and maximum range.

[0233] The administrator can change the efficiency and stability (T11) of the preference evaluation value within a range of 0 to 100. The recommendation module can recalculate the recommendation information of the recommended order in the optimal configuration in real time according to the preset recommendation method based on the changed preference evaluation value.

[0234] Thus, if there is doubt about the initial preference evaluation value, the administrator can make changes and recommend the optimal virtual machine configuration.

[0235] Figure 7 It is a flow chart of a server operation method according to an embodiment of the present invention.

[0236] like Figure 7 As shown, the monitoring module can monitor the virtual machine in real time and monitor whether the virtual machine and / or physical machine are working abnormally.

[0237] When the monitoring module determines that the physical machine and / or virtual machine is working abnormally, the migration module can determine whether migration is required to solve the problem.

[0238] In one example, abnormal operation may be an undersized / oversized virtual machine problem.

[0239] However, without being limited thereto, a technician can generally make various modifications to the abnormal operation at an obvious level.

[0240] When the migration module determines that migration is necessary, in order to solve the problem, the migration module can use the physical machine to which the virtual machine is migrated to calculate the expected virtual machine configuration.

[0241] The migration module transmits the virtual machine configuration to be changed to the virtual machine configuration evaluation device and requests calculation of configuration evaluation values in terms of stability and efficiency.

[0242] The configuration evaluation value may be obtained by calculating the current virtual machine configuration and the future virtual machine configuration to be migrated.

[0243] The migration module determines whether migration is ultimately required based on the configuration evaluation value. When the configuration evaluation value meets the given evaluation criteria, the migration module can perform the migration.

[0244] In one example, the given evaluation criterion may be a condition that stability or efficiency is greater than a given benchmark (20%).

[0245] However, without being limited thereto, a person skilled in the art can generally make various modifications to a given evaluation criterion at an obvious level.

[0246] The technical features of the virtual machine configuration evaluation method described in one embodiment of the present invention are as follows: the virtual machine configuration evaluation method applies a virtual machine configuration evaluation device to calculate the configuration evaluation of the virtual machine operating in the physical server, that is, the configuration evaluation value, which may include the following steps: through a receiving module, collecting information generated when the virtual machine is started, that is, the operation information; through a first computing module, based on the operation information, calculating the virtual machine configuration efficiency value, that is, the efficiency evaluation value; through a second computing module, based on the operation information, calculating the virtual machine configuration stability value, that is, the stability evaluation value; and through a configuration evaluation module, applying the efficiency evaluation value and the stability evaluation value to calculate the configuration evaluation value.

[0247] When the virtual machine optimal configuration recommendation device and the server management device send a request, the virtual machine configuration evaluation device can evaluate the configuration of the virtual machine from the aspects of stability and efficiency according to the situation and transmit the configuration evaluation value.

[0248] To this end, necessary information may be received from a server management device and / or a virtual machine optimal configuration recommendation device to facilitate evaluation.

[0249] In order to make the drawings more clearly illustrate the technical concept of the present invention, structures that have no relevance or have reduced relevance to the technical concept of the present invention are simply expressed or omitted.

[0250] The above content describes the composition and features of the present invention based on the embodiments of the present invention. However, the present invention is not limited thereto. It is obvious to those skilled in the art that various changes or modifications can be made within the concept and scope of the present invention. Therefore, it should be clear that the above changes or modifications fall within the scope of the claims attached hereto.

Claims

1. A device for recommending optimal configuration of a virtual machine, characterized by: The virtual machine optimal configuration recommendation device is used to achieve the optimal configuration of the virtual machine working in the physical server, which includes: A collection module is used to collect information generated when an administrator operates a physical server, that is, work information; a preference evaluation module, which calculates a preference evaluation value related to the administrator's preference based on the work information and according to a preset preference calculation method; and The recommendation module is used to reflect the preference evaluation value and calculate recommendation information for recommending the virtual machine configuration in the physical server to the administrator.

2. The device for recommending optimal virtual machine configuration according to claim 1, wherein: It further includes a configuration calculation module, which is used to calculate the virtual machine configuration of the service server that meets the preset configuration conditions, that is, the optimal configuration, The configuration calculation module receives the configuration evaluation value of the optimal configuration evaluated by the configuration evaluation module according to a preset evaluation method, The recommendation module divides the optimal configurations according to a preset recommendation method in the order of the configuration evaluation values that are similar to the preference evaluation values, and selects the virtual machine configuration to be recommended.

3. The device for recommending optimal virtual machine configuration according to claim 2, wherein: The preference evaluation value and the configuration evaluation value may be calculated taking into account stability and efficiency aspects.

4. The device for recommending optimal virtual machine configuration according to claim 3, wherein: The preset preference calculation method is as follows: based on the work information, the operation mode of the administrator's virtual machine management system is divided, and the preference evaluation value is calculated.

5. The device for recommending optimal virtual machine configuration according to claim 2, wherein: The preset recommendation method is as follows: based on the preference evaluation value, the configuration evaluation value is compared, and the configuration of the virtual machine is recommended in a similar order. When the stability and efficiency are similar, the weighted value is applied to the stability to recommend the configuration of the virtual machine.

6. The device for recommending optimal virtual machine configuration according to claim 1, wherein: It further includes an interface module, which is used to calculate an interface so as to display the recommendation information generated by the recommendation module to the administrator. The interface module can calculate an interface for displaying the optimal configuration information and the recommended information together.

7. The device for recommending optimal virtual machine configuration according to claim 6, wherein: The interface module can change the preference evaluation value within a range from a minimum to a maximum, calculate the preference evaluation value reflecting the change, and recommend an interface for the virtual machine configuration in the physical server.

8. The device for recommending optimal virtual machine configuration according to claim 2, wherein: When the amount of work information collected is less than the preset amount, the collection module will collect the administrator's personal information. The preset preference calculation method is a method of calculating the preference evaluation value by taking the personal information and the work information into consideration.

9. The device for recommending optimal virtual machine configuration according to claim 8, wherein: It further includes a proficiency judgment module, which judges the administrator's proficiency in operating the virtual machine management system based on the work information, The preset preference calculation method is as follows: based on the operation proficiency, the weighted values of the personal information and the work information are modified to calculate the preference evaluation value.

10. A method for recommending optimal virtual machine configuration, characterized by: The virtual machine optimal configuration recommendation method is implemented through a virtual machine management system to operate virtual machines working on a physical server. The virtual machine optimal configuration recommendation method includes the following steps: The collection module collects information generated when the administrator controls the virtual machine management system, that is, work information; Calculating, by the preference evaluation module, a preference evaluation value related to the administrator's preference based on the work information and according to a preset preference calculation method; and The recommendation module reflects the preference evaluation value and selects a virtual machine configuration in the physical server to be recommended to the administrator.

Images