KR102134785B1 - Method and device for supporting user to arrange object on optimized position based on machine learning - Google Patents

Abstract

According to the present invention, disclosed are an object arrangement recommendation method for allowing a user to arrange an object in an optimized location based on machine learning and an object arrangement recommendation device using the same. The object arrangement recommendation method includes the steps of: (a) performing, by an object arrangement recommendation device, analysis processing for a plurality of candidate arrangement detailed images and a new arrangement detailed image to calculate an arrangement similarity representing a degree of similarity to the new arrangement detailed image for each of the plurality of candidate arrangement detailed images based on the new arrangement detailed image when the new arrangement detail image is obtained while the plurality of candidate arrangement detail images are registered in a predetermined storage unit; and (b) by the object arrangement recommendation device, (i) with reference to the arrangement similarity, determining a specific candidate arrangement detailed image of which the arrangement similarity satisfies a predetermined similarity condition among the plurality of candidate arrangement detailed images, and (ii) with reference to the power consumption information corresponding to each of the specific candidate arrangement detailed images, determining an optimal candidate arrangement detailed image that satisfies a predetermined power consumption condition, among the specific candidate arrangement detailed images, and providing the corresponding information to a user.

Description

An object arrangement recommendation method and an object arrangement recommendation apparatus using the same for an object arrangement to allow a user to arrange an object in an optimized location based on machine learning.

The present invention relates to an object arrangement recommendation method and an object arrangement recommendation apparatus using the object arrangement for allowing a user to arrange objects in an optimized position based on machine learning.

In general, when a specific user places a rack and a server in a specific space, the size and structure of the specific space are remembered, and the rack and the server are arranged based on the size and structure. It may happen that the power exceeds the reference value. Nevertheless, a certain user has to repeatedly perform the same task until it is difficult to easily know the amount of power consumed by the entire server until the rack and the server are actually placed and operated. There was a problem.

In addition, when a specific user wants to place the rack and the server in the same space in a different space having a similar size and similar structure to a specific space, there is a problem that a specific user must directly remember the arrangement structure of the rack and the server in a specific space Did.

In addition, even if a specific user has placed the rack and the server so that the power consumed by the entire server is small, there is a problem that other users experience a myriad of trial and error because the deployment structure is often unknown.

Therefore, a situation is required to improve the above problems.

The present invention aims to solve all of the above-mentioned problems.

In addition, the present invention, by storing and managing the rack and server arrangement structure and power consumption in conjunction, has a structure similar to the layout structure of the rack and server desired by the user, while the structure of the rack and server desired by the user Another object is to recommend to the user an arrangement structure having a smaller amount of power consumption than when using.

In addition, the present invention, with reference to the size and number of racks, the size, number, type and year of the rack, as well as the location of the rack and server the user wants to place, the arrangement of the rack and the server desired by the user Another object is to recommend to the user an arrangement structure having a smaller amount of power consumption than when using the structure.

In addition, the present invention, the actual power consumption when the user actually deploys the rack and the server according to the layout structure recommended to the user, if the predicted power consumption corresponding to the recommended deployment structure, if the actual consumption power consumption prediction Another object is to recommend the user to a deployment structure having a more efficient power consumption by updating the power value.

In addition, the present invention enables users to deploy racks and servers through a user interface (UI), thereby confirming expected power consumption of a desired deployment structure without a user having to spend a lot of effort, and accordingly Another goal is to help you find a more efficient deployment structure.

The characteristic configuration of the present invention for achieving the object of the present invention as described above and realizing the characteristic effects of the present invention described below is as follows.

According to an aspect of the present invention, in the object arrangement recommendation method for allowing a user to arrange an object in an optimized position based on machine learning, the method includes: (a) a plurality of candidate arrangement detailed images in a predetermined storage unit When the new arrangement detailed image is obtained in the registered state, the object arrangement recommendation apparatus performs image analysis processing on the plurality of candidate arrangement detailed images and the new arrangement detailed image, based on the new arrangement detailed image. Calculating the similarity of arrangements representing the similar degree to the new arrangement detailed image for each of the plurality of candidate arrangement detailed images; And (b) the object arrangement recommendation apparatus determines (i) a specific candidate arrangement detailed image among the plurality of candidate arrangement detailed images with reference to the arrangement similarity, wherein the arrangement similarity satisfies a predetermined similarity condition. And (ii) determining the optimal candidate arrangement detail image that satisfies a predetermined power consumption condition among the specific candidate arrangement detail images with reference to power consumption information corresponding to each of the specific candidate arrangement detail images, and determining the corresponding information. Disclosed is a method comprising providing to a user.

As an example, in the step (a), (a1) a plurality of candidate array data corresponding to each of the plurality of candidate arrangement detailed images, wherein the plurality of candidate array data includes candidate index information indicating each position in the candidate arrangement space. And the candidate index information includes specific candidate index information regarding a specific position in which the object is arranged, among each position in the candidate arrangement space, while-is additionally registered in the predetermined storage unit, the New array data corresponding to the new arrangement detailed image-the new array data includes new index information indicating each position in the new arrangement space, and the new index information is among each position in the new arrangement space, If the object includes specific new index information related to a specific location in which the object is arranged-is additionally obtained, the object arrangement recommendation apparatus refers to the candidate index information and the new index information, so that the object is the candidate arrangement. The number of candidate objects arranged in space and the number of new objects in which the objects are arranged in the new arrangement space are determined, and among the plurality of candidate array data, the number of candidate objects is within a threshold range from the number of new objects Determining present, at least one selection candidate array data; And (a2) the object arrangement recommendation apparatus performs the image analysis processing on a plurality of selection candidate arrangement detailed images and the new arrangement detailed image corresponding to each of the selection candidate array data, and based on the new arrangement detailed image. By calculating each of the plurality of selection candidate arrangement detailed image similarity of the selection arrangement indicating the similar degree to the new arrangement detailed image for each; including, in step (b), the object arrangement recommendation device, (i) ) With reference to the sorting arrangement similarity, among the plurality of sorting candidate arrangement detailed images, a specific sorting candidate arrangement detailed image is determined, wherein the sorting arrangement similarity satisfies the predetermined similarity condition, and (ii) the specific sorting. With reference to the power consumption information corresponding to each candidate arrangement detailed image, among the specific selection candidate arrangement detailed images, an optimal selection candidate arrangement detail image satisfying the predetermined power consumption condition is determined as the optimum candidate arrangement detail image, Disclosed is a method characterized by providing corresponding information to the user.

As an example, a plurality of candidate arrangement outline images corresponding to each of the plurality of candidate arrangement detailed images are respectively managed as the plurality of candidate array data and are additionally registered in the predetermined storage unit, and a new arrangement outline image is additionally added. When acquired, the object arrangement recommendation device manages the new arrangement outline image as the new array data, respectively, and the plurality of candidate arrangement outline images and the new arrangement outline image are positions related to a specific location where the object is arranged. Disclosed is a method comprising information, wherein the location information corresponds to the specific candidate index information and the specific new index information, respectively.

As an example, the object includes at least one rack object and at least one server object, and in step (a2), the object arrangement recommendation apparatus includes the plurality of selection candidate arrangement detailed images and the new arrangement. The image analysis processing is performed on the detailed image, and among the rack object and the server object, (i) a selection candidate rack object and a selection candidate server object included in each of the plurality of selection candidate arrangement detailed images and (ii) the The selection candidate rack included in the plurality of selection candidate arrangement detailed images for each of the plurality of selection candidate arrangement detailed images by performing the image analysis processing on the new rack object and the new server object included in the new arrangement detail image. Each rack object arrangement degree of similarity indicating information about the object and the information about the new rack object included in the new arrangement detail image, and the selection candidate server object included in the plurality of selection candidate arrangement detail images Each server object arrangement degree of similarity indicating the degree of similarity between information and information about the new server object included in the new arrangement detailed image is calculated, and the plurality of the plurality of objects are referred to with reference to the rack object arrangement degree of similarity and the server object arrangement degree of similarity. Disclosed is a method for calculating the similarity of the selection arrangement for each selection image of the candidate selection arrangement.

As an example, the information about the new rack object included in the new arrangement detailed image includes information on the location where the new rack object is arranged, the number and size of the new rack objects, and information about the new server object. Disclosed is a method characterized by including information on a location where the new server object is arranged, the number and size of the new server object, a model name of the server object, and a manufacturing year.

As an example, (c) the object arrangement recommendation apparatus supports to display an object list corresponding to the first object to the n-th object in a first area of the display area of the user terminal, and allows the user to determine the optimal candidate arrangement details. Disclosed is a method further comprising: supporting the arrangement of at least one of the first object to the n-th object in a second area of the display area of the user terminal by referring to an image.

As an example, the second region includes a plurality of sub-regions, and in step (c), the object arrangement recommendation apparatus refers to the optimum candidate arrangement detailed image, and among the sub-regions, the first object to Disclosed is a method characterized in that the specific object to be arranged is displayed in a specific color.

As an example, the first object to the n-th object are (i) a first rack object to a k-th rack object, wherein k is 1 or more and an integer less than n-and (ii) k+1 server objects to n server objects, and in step (c), the object arrangement recommendation apparatus allows the user to refer to the optimal candidate arrangement detailed image, so that the first rack object to the kth area in the second area Support to arrange at least one specific rack object among the rack objects, and support to arrange at least one specific server object among the k+1 server object or the n-th server object to the specific rack object. A method of characterizing is disclosed.

As an example, in the step (c), in the state in which the object arrangement recommendation device supports displaying the object list corresponding to the first object to the n-th object in the first area, the user First object add button to n-th object add button for rendering one object to the n-th object in the second area-The first object add button to the n-th object add button is displayed in the object list-Medium When a specific object add button is selected by the user, the object arrangement recommendation apparatus arranges a specific object corresponding to the specific object add button among the first object to the n-th object in the second area by the user. Disclosed is a method characterized by supporting the support.

As an example, in the state in which the specific object add button is selected by the user, the specific object is arranged from the first area to the second area through a drag and drop method. This is disclosed.

According to another aspect of the present invention, an object arrangement recommendation apparatus for allowing a user to arrange an object in an optimized location based on machine learning, the apparatus comprising: at least one memory storing instructions; And at least one processor configured to execute the instructions, wherein the processor, when a plurality of candidate arrangement detail images are registered in a predetermined storage unit, when a new arrangement detail image is obtained, (1) the plurality Image analysis processing is performed on the candidate arrangement detail image and the new arrangement detail image, and an arrangement similarity indicating the degree of similarity to the new arrangement detail image for each of the plurality of candidate arrangement detail images based on the new arrangement detail image. A process for calculating each and (2) (i) a specific candidate arrangement detailed image, from which the arrangement similarity satisfies a predetermined similarity condition, is determined from the plurality of candidate arrangement detailed images with reference to the arrangement similarity, and ( ii) With reference to the power consumption information corresponding to each of the specific candidate arrangement detailed images, among the specific candidate arrangement detailed images, an optimal candidate arrangement detailed image that satisfies a predetermined power consumption condition is determined and the corresponding information is provided to the user Disclosed is an object arrangement recommendation apparatus that executes the instructions for performing a process.

As an example, the (1) process may include (1-1) a plurality of candidate array data corresponding to each of the plurality of candidate arrangement detailed images, wherein the plurality of candidate array data are candidate indexes indicating respective positions in the candidate arrangement space. Information, and the candidate index information includes specific candidate index information regarding a specific position in which the object is arranged, among each position in the candidate arrangement space-in a state in which is additionally registered in the predetermined storage unit , New array data corresponding to the new arrangement detailed image-The new array data includes new index information indicating each position in the new arrangement space, and the new index information is each position in the new arrangement space. Among, if the object includes specific new index information regarding a specific position in which the object is arranged-is additionally obtained, the processor refers to the candidate index information and the new index information, so that the object is the candidate arrangement space The number of candidate objects arranged in the number and the number of new objects in which the objects are arranged in the new arrangement space are determined, and among the plurality of candidate array data, the number of candidate objects is within a threshold range from the number of new objects The process of determining at least one selection candidate array data, and (1-2) the processor analyzes the images for a plurality of selection candidate arrangement detailed images and the new arrangement detailed image corresponding to each of the selection candidate array data. And performing a processing to calculate a selection arrangement similarity indicating a similar degree to the new arrangement detail image for each of the plurality of selection candidate arrangement detail images based on the new arrangement detail image, wherein the processor comprises: (2) In the process, (i) with reference to the sorting arrangement similarity, Among the plurality of selection candidate arrangement detailed images, a specific selection candidate arrangement detailed image that satisfies the predetermined similarity condition is determined, and (ii) the amount of power consumption corresponding to each of the specific selection candidate arrangement detailed images. With reference to information, among the specific selection candidate arrangement detailed images, an optimal selection candidate arrangement detailed image that satisfies the predetermined power consumption condition is determined as the optimum candidate arrangement detail image, and the corresponding information is provided to the user Disclosed is an object arrangement recommendation device.

As an example, a plurality of candidate arrangement outline images corresponding to each of the plurality of candidate arrangement detailed images are respectively managed as the plurality of candidate array data and are additionally registered in the predetermined storage unit, and a new arrangement outline image is additionally added. When acquired, the processor manages the new arrangement outline image as the new array data, respectively, and the plurality of candidate arrangement outline images and the new arrangement outline image include location information regarding a specific location where the object is arranged. Then, the object arrangement recommendation apparatus is characterized in that the location information corresponds to the specific candidate index information and the specific new index information, respectively.

As an example, the object includes at least one rack object and at least one server object, and the processor, in the (1-2) process, the plurality of selection candidate arrangement detail images and the new arrangement details The image analysis processing is performed on an image, but among the rack object and the server object, (i) a selection candidate rack object and a selection candidate server object included in each of the plurality of selection candidate arrangement detailed images and (ii) the new The screening candidate rack object included in the plurality of screening candidate arrangement detail images for each of the screening candidate arrangement detail images by performing the image analysis processing on the new rack object and the new server object included in the arrangement detail image. Information on each rack object arrangement similarity and information on the selection candidate server object included in the plurality of selection candidate arrangement detailed images indicating the degree of similarity between information on the new arrangement object and the information on the new rack object included in the new arrangement detail image. And each server object arrangement degree of similarity indicating the degree of similarity of information on the new server object included in the new arrangement detailed image, and referring to the rack object arrangement degree of similarity and the server object arrangement degree of similarity. Disclosed is an object arrangement recommendation apparatus, wherein the selection arrangement similarity is calculated for each candidate arrangement detailed image.

As an example, the information about the new rack object included in the new arrangement detailed image includes information on the location where the new rack object is arranged, the number and size of the new rack objects, and information about the new server object. Disclosed is an object arrangement recommendation apparatus characterized in that information on a location where the new server object is arranged, the number and size of the new server object, a model name of the server object, and a manufacturing year is included.

As an example, (3) the processor supports to display a list of objects corresponding to the first object to the n-th object in a first area of the display area of the user terminal, and allows the user to refer to the detailed image of the optimal candidate arrangement Disclosed is an object arrangement recommendation device, further comprising a process for supporting the arrangement of at least one of the first object to the n-th object in a second area of the display area of the user terminal.

As an example, the second region includes a plurality of sub-regions, and the processor, in the process (3), refers to the optimal candidate arrangement detail image, and among the sub-regions, the first object to the n th Disclosed is an object arrangement recommendation apparatus, characterized in that a specific color of an object is to be arranged in a specific color.

As an example, the first object to the n-th object are (i) a first rack object to a k-th rack object, wherein k is 1 or more and an integer less than n-and (ii) k+1 server objects to n server objects, and the processor, in the (3) process, by allowing the user to refer to the optimal candidate arrangement detailed image, in the second area of the first rack object to the k-th rack object Characterized in that it supports to arrange at least one specific rack object, and supports to arrange at least one specific server object among the k+1 server object or the n-th server object to the specific rack object. Disclosed is an object arrangement recommendation device.

As an example, in the process (3), while the processor supports to display the object list corresponding to the first object to the n-th object in the first area, the user may use the first object to Add a specific object from the first object add button to the n-th object add button for rendering the n-th object in the second area-the first object add button to the n-th object add button are displayed in the object list When a button is selected by the user, the processor supports the user to arrange a specific object corresponding to the specific object addition button among the first object or the n-th object to the second area. Disclosed is a device for recommending an arrangement of objects.

As an example, the specific object is an object characterized in that the specific object is arranged in the second area from the first area through a drag and drop (drag and drop) method while the specific object addition button is selected by the user. An arrangement recommendation device is disclosed.

According to the present invention, there are the following effects.

According to the present invention, a rack and server arrangement structure and a power consumption are interlocked to store and manage, thereby having a structure similar to that of a rack and server desired by a user and a structure of rack and server desired by a user. There is an effect of recommending a batch structure having a smaller amount of power consumption to the user.

In addition, the present invention, with reference to the size and number of racks, the size, number, type and year of the rack, as well as the location of the rack and server the user wants to place, the arrangement of the rack and the server desired by the user There is an effect of recommending to the user an arrangement structure having a smaller amount of power consumption than when using the structure.

In addition, the present invention, the actual power consumption when the user actually deploys the rack and the server according to the layout structure recommended to the user, if the predicted power consumption corresponding to the recommended deployment structure, if the actual consumption power consumption prediction By updating the power value, there is an effect of recommending a deployment structure having a more efficient power consumption to the user.

In addition, the present invention enables users to deploy racks and servers through a user interface (UI), thereby confirming expected power consumption of a desired deployment structure without a user having to spend a lot of effort, and accordingly It has the effect of helping to find a more efficient deployment structure.

1 schematically illustrates an object arrangement recommendation apparatus and a user terminal for allowing a user to arrange objects in an optimized location based on machine learning according to an embodiment of the present invention,
2A, 2B, and 2C are object arrangement recommendation methods for allowing a user to arrange objects in an optimized location based on machine learning according to an embodiment of the present invention, new arrangement detailed image, new array data And a schematic image of the new arrangement, respectively,
3A and 3B are object arrangement recommendation methods for allowing a user to arrange an object in an optimized position based on machine learning according to an embodiment of the present invention, a predetermined display through a UI in a display area of a user terminal It schematically shows how information is displayed.

For a detailed description of the present invention, which will be described later, reference is made to the accompanying drawings that illustrate, by way of example, specific embodiments in which the invention may be practiced. These examples are described in detail enough to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the invention are different, but need not be mutually exclusive. For example, certain shapes, structures, and properties described herein may be implemented in other embodiments without departing from the spirit and scope of the invention in relation to one embodiment. In addition, it should be understood that the location or placement of individual components within each disclosed embodiment can be changed without departing from the spirit and scope of the invention. Therefore, the following detailed description is not intended to be taken in a limiting sense, and the scope of the present invention, if appropriately described, is limited only by the appended claims, along with all ranges equivalent to those claimed. Like reference numerals in the drawings refer to the same or similar functions throughout the several aspects.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to enable those skilled in the art to easily implement the present invention. It will be readily understood by those skilled in the art that the method of the present invention will be implemented utilizing a combination of computer hardware and software.

1 schematically illustrates an object arrangement recommendation apparatus 100 and a user terminal 200 for allowing a user to arrange an object in an optimized position based on machine learning according to an embodiment of the present invention. It is a drawing.

First, the object arrangement recommendation apparatus 100 may include a memory 110 and a processor 120.

Here, the memory 110 of the object arrangement recommendation apparatus 100 may store instructions. Specifically, instructions are computer software for allowing the processor 120 of the object arrangement recommendation apparatus 100 to function in a specific manner. , Computer readable or computer readable memory capable of directing a computer or other programmable data processing equipment. It is also possible for the instructions to carry out processes for executing the functions shown in the drawing.

In addition, the processor 120 of the object arrangement recommendation apparatus 100 may include hardware configurations such as a micro processing unit (MPU) or a central processing unit (CPU), a cache memory, and a data bus. have. Also, it may further include a software configuration of an operating system and an application performing a specific purpose.

On the other hand, the user terminal 200 may also include a memory 210 and a processor 220, respectively, as in the object arrangement recommendation apparatus 100, and a description thereof is provided for the memory 110 and the object arrangement recommendation apparatus 100. Since it is similar to the description of the processor 120, a detailed description is omitted.

Next, the overall process for the object arrangement recommendation apparatus 100 to allow the user to arrange objects in an optimized position based on machine learning will be described.

First, when a plurality of candidate arrangement detailed images are registered in a predetermined storage unit, when the new arrangement detailed image 310 is obtained, the object arrangement recommendation apparatus 100 displays a plurality of candidate arrangement detailed images and a new arrangement detailed image. By performing image analysis processing on 310, it is possible to calculate arrangement similarity, which indicates the similarity to the new arrangement detailed image 310 for each of the plurality of candidate arrangement detailed images, based on the new arrangement detailed image 310. .

Here, the new arrangement detailed image 310 is an image obtained from the user terminal 200 and may correspond to an image in which at least one object is arranged in the new arrangement space.

As an example, as illustrated in FIG. 2A, the new arrangement detailed image 310 may correspond to an image showing a view of at least one rack and at least one server arranged in the new arrangement space in a predetermined direction. have.

The new arrangement detailed image 310 may correspond to an image obtained by arranging a rack object and a server object through a UI to be described later, but is not limited thereto, and the rack arranged on the actual space and It may be an image containing a view of the server from a predetermined direction.

On the other hand, the candidate arrangement detailed image is an image previously registered in a predetermined storage unit, and the description thereof is similar to the description of the new arrangement detailed image 310, so a detailed description is omitted.

In addition, when the object arrangement recommendation apparatus 100 performs image analysis processing on a plurality of candidate arrangement detailed images and a new arrangement detailed image 310, image analysis processing based on machine learning may be performed, but is not limited thereto. It is not possible to perform general image analysis processing.

For example, as shown in FIG. 2A, the new arrangement detailed image 310 may represent a state in which a rack object is arranged at a specific location in the new arrangement space, and each rack object, that is, the rack object 3 1 server object, 2 server objects in rack object 4, 4 server objects in rack object 5, 2 server objects in rack object 6, 1 server object in rack object 7, rack object in rack 8 It may indicate that one server object is arranged.

If, in the predetermined storage unit, when a state in which 10 candidate arrangement detailed images representing a state in which rack objects and server objects are arranged in a space of various sizes is registered, the object arrangement recommendation device 100, the new arrangement details On the basis of the image 310, for each of the 10 candidate arrangement detailed images, the arrangement similarity indicating the similarity to the new arrangement detailed image 310 may be calculated.

For example, based on the new arrangement detail image 310, the object arrangement recommendation apparatus 100 may have a similarity of 92% for the candidate candidate 1 detailed arrangement image, a similarity of 95% for the candidate 2 detailed arrangement image, and candidate 3 The arrangement detailed image has 89% similarity, and the candidate #10 arrangement detailed image has 85% similarity, so that the arrangement similarity can be calculated.

In addition, the object arrangement recommendation apparatus 100 may determine a specific candidate arrangement detailed image in which a plurality of candidate arrangement detailed images satisfy a predetermined degree of similarity condition, with reference to the arrangement similarity.

Here, the predetermined similarity condition may mean that the arrangement similarity of candidate arrangement detailed images is a reference value, for example, 90% or more. In this case, among the 10 candidate arrangement detailed images mentioned above, a specific candidate arrangement detailed image that satisfies a predetermined similarity condition may include a candidate first arrangement detail image, a second candidate arrangement detail image, and the like. However, the predetermined similarity condition is not limited to this, and for example, it may mean that it is included in the top 3 when judging the order of arrangement similarity.

Then, the object arrangement recommendation apparatus 100 determines the optimal candidate arrangement detail image that satisfies a predetermined power consumption condition among the specific candidate arrangement detail images with reference to power consumption information corresponding to each specific candidate arrangement detail image. And provide the information to the user.

Here, the predetermined power consumption condition is the power consumption amount of the reference value, corresponding to each specific candidate arrangement detailed image, similar to the predetermined similarity condition. For example, it may mean that it is 5000W or less, but is not limited thereto. For example, the predetermined power consumption condition may mean that the power consumption is included in the top five when judging in the order of low power consumption.

As described above, the object arrangement recommendation apparatus 100 determines a specific candidate arrangement detailed image similar to the new arrangement detailed image 310 obtained from the user through a predetermined similarity condition, and among them, an optimal candidate arrangement that satisfies the power consumption condition. By determining the detailed image and providing the information to the user, the user is provided with information on the same or similar placement structure as the user desires, but provides information on the placement structure that consumes power more efficiently. Can receive

Meanwhile, new array data may be additionally acquired along with the new arrangement detailed image 310.

That is, while a plurality of candidate array data corresponding to each of the plurality of candidate arrangement detailed images are additionally registered in a predetermined storage unit, new array data 320 corresponding to the new arrangement detailed image 310 is additionally acquired. When the object arrangement recommendation apparatus 100 refers to the candidate index information and the new index information, the number of candidate objects, which is the number of objects arranged in the candidate arrangement space, and the number of new objects, the number of objects arranged in the new arrangement space Can judge.

Here, the new array data 320 may include new index information indicating each position in the new arrangement space, and the new index information is a specific new one of each position in the new arrangement space, regarding a specific position in which the object is arranged. It may include index information.

Similarly, the candidate array data may also include candidate index information indicating each position in the candidate arrangement space, and the candidate index information is specific candidate index information regarding a specific position in which an object is arranged, among each position in the candidate arrangement space. It may include.

As an example, as illustrated in FIG. 2B, the new array data 320 may include new index information indicating each of positions 0 to 11 in the new arrangement space.

For convenience of description, a shadow is displayed in the area corresponding to the corresponding index on the new array data 320 to indicate that the objects are arranged at positions 2, 5, 6, 9 and 11. The data 320 may include specific new index information on positions 2, 5, 6, 9, and 11 in which the object is arranged. In addition, since the new array data 320 illustrated in FIG. 2B represents a state in which five objects are arranged in the new arrangement space, the number of new objects is 5.

In addition, the object arrangement recommendation apparatus 100 may determine at least one selection candidate array data in which the number of candidate objects is within a threshold range from the number of new objects among the plurality of candidate array data.

For example, the new arrangement space and the candidate arrangement space may have the same space size and structure, and 50 candidate array data for the candidate arrangement space may be additionally registered in a predetermined storage unit. Of these, 20 candidate array data are candidate array data having two candidate objects, the other 20 candidate array data are candidate array data having 9 candidate objects, and the remaining 10 candidate array data are candidate object numbers. It may be four candidate array data.

Here, if the difference between the number of candidate objects and the number of new objects is within one threshold range, among 50 candidate array data mentioned above, candidate array data having four candidate objects may be determined as selection candidate array data.

As described above, the object arrangement recommendation apparatus 100 must perform image analysis processing to compare whether the candidate arrangement detailed images and the new arrangement detailed images 310 are similar. Among a plurality of candidate array data, the candidate object Through the process of determining the selection candidate array data whose number is within the threshold range from the number of new objects, the target of image analysis processing is reduced, and thus the total number of times of performing image analysis processing can be significantly reduced.

Meanwhile, the classification process may be applied to the size of the arrangement space in which objects are arranged. That is, among the plurality of candidate array data, at least one spatial selection candidate array data in which the size of the candidate arrangement space exists within a threshold range from the size of the new arrangement space may be determined.

Then, the object arrangement recommendation apparatus 100 performs image analysis processing on the plurality of selection candidate arrangement detailed images and the new arrangement detailed image 310 corresponding to each of the selection candidate array data, and thus the new arrangement detailed image 310 Based on the criteria, it is possible to calculate the degree of similarity of the sorting arrangement, which represents the degree of similarity to the new arrangement detail image 310 for each of the plurality of sorting candidate arrangement detailed images.

Then, the object arrangement recommendation apparatus 100 may determine a specific selection candidate arrangement detailed image, among the plurality of selection candidate arrangement detailed images, satisfying a predetermined similarity condition, with reference to the selection arrangement similarity. .

Then, with reference to the power consumption information corresponding to each specific selection candidate arrangement detailed image, among the specific selection candidate arrangement detailed images, an optimal selection candidate arrangement detail image that satisfies a predetermined power consumption condition is determined as the optimum candidate arrangement detail image And provide the information to the user. In this regard, since a similar description has been given above, further detailed description is omitted.

Meanwhile, a plurality of candidate arrangement outline images corresponding to each of the plurality of candidate arrangement detailed images are managed as a plurality of candidate array data, respectively, and may be additionally registered in a predetermined storage unit. In addition, when the new arrangement outline image 330 is additionally obtained, the object arrangement recommendation apparatus 100 may manage the new arrangement outline image 330 as new array data 320, respectively.

Here, the new arrangement outline image 330 may correspond to an image in which a user can visually check where an object is arranged.

FIG. 2C schematically illustrates the new arrangement outline image 330, and the two-dimensional new arrangement outline image 330 obtained through the UI from a user is a new one-dimensional array in the order of numbers corresponding to respective positions. Data 320 may be managed.

As can be seen in FIG. 2C, the new arrangement outline image 330 may include location information regarding a specific location in which the object is arranged.

For convenience, the regions corresponding to No. 2, No. 5, No. 6, No. 9, and No. 11 of the new arrangement outline image 330 are shaded, which means that the object is arranged at the corresponding location. That is, the new arrangement outline image 330 may include location information on positions 2, 5, 6, 9, and 11 in which the object is arranged, and the location information is included in specific new index information described above. Can be countered.

In addition, the description of the plurality of candidate arrangement outline images is similar to the description of the new arrangement outline image 330, so a detailed description is omitted.

Meanwhile, the object may include at least one rack object and at least one server object, and the object arrangement recommendation apparatus 100 performs image analysis processing on a plurality of selection candidate arrangement detailed images and new arrangement detailed images 310. However, among the rack objects and server objects, (i) a selection candidate rack object and a selection candidate server object included in each of the plurality of selection candidate arrangement detailed images, and (ii) a new image included in the new arrangement detail image 310. By performing image analysis processing on the rack object and the new server object, for each of the plurality of selection candidate arrangement detailed images, information on the selection candidate rack object included in the plurality of selection candidate arrangement detailed images and the new arrangement detailed image 310. Information about the selected candidate server object included in each rack object arrangement similarity and a plurality of selected candidate arrangement detailed images and the new server included in the new arrangement detailed image 310 indicating the similarity of information about the new rack object included Each server object arrangement degree of similarity indicating the degree of similarity of information about an object may be calculated, and the selection arrangement degree of similarity may be calculated for each detailed image of a plurality of selection candidate arrangements by referring to the rack object arrangement degree of similarity and the server object arrangement degree of similarity.

Here, the information about the new rack object may be information indicating where the new rack object is arranged, and the like. In more detail, the information on the new rack object may include information on the location where the new rack object is arranged, and the number and size of new rack objects. In addition, when the new rack object is a shape of a rectangular parallelepiped as a whole, information about a specific open surface among six surfaces, that is, a specific open surface for installing a server object, may be included in the information about the new rack object. Can.

Meanwhile, the information on the new server object may include information on the location where the new server object is arranged, the number and size of the new server objects, the model name of the server object, and the year of manufacture.

Here, the location where the new server object is arranged is not limited to only the coordinates where the new server object is located.

For example, a server having a structure in which heat is generated from the server is cooled by intake vents at the front of the server, exhaust ports at the rear of the server, and circulation of air from the front to the rear. When arranging a plurality of servers having the above structure, if another server is arranged such that the front part of the other server is located near the rear part of the specific server among the plurality of servers, the other server intakes hot air discharged by the specific server as it is. Because it can be difficult to cool the heat normally.

Accordingly, when there are a plurality of servers, since a difference in power consumption may occur depending on the directions of the corresponding servers, the information about the new server object may also include direction information of the new server object.

On the other hand, the object arrangement recommendation apparatus 100 may support displaying the object list corresponding to the first object to the n-th object in the first area 410 of the display area 400 of the user terminal 200.

As an example, referring to FIG. 3A, the object arrangement recommendation apparatus 100 includes a first rack object 410-1 and a first rack object 410 in the first area 410 of the display area 400 of the user terminal 200. 2 rack object 410-2, ..., k-th rack object 410-k, k+1 server object 410-(k+1), k+2 server object 410-(k +2)), ..., it may support to display the object list corresponding to the n-th server object (410-n). Here, k is an integer greater than or equal to 1 and less than n.

In addition, the user may arrange at least one of the first object to the n-th object in the second area 420 of the display area 400 of the user terminal 200 by referring to the detailed image of the optimal candidate arrangement.

Specifically, in the state in which the object arrangement recommendation apparatus 100 supports to display the object list corresponding to the first object to the n-th object in the first area 410, the user may implement the first object to the n-th object. When a specific object addition button is selected by a user from a first object addition button to an n-th object addition button to render the second area 420, the object arrangement recommendation device 100 allows the user to A specific object corresponding to a specific object addition button among n-th objects may be arranged to be arranged in the second area 420.

Here, the first object adding button to the n-th object adding button may be displayed separately from the first object to the n-th object appearing on the object list, but is not limited thereto, as shown in FIGS. 3A and 3B. The first to nth objects themselves appearing on the object list may correspond to an object addition button.

Meanwhile, as illustrated in FIGS. 3A and 3B, the specific object is second from the first area 410 through a drag and drop method while a specific object add button is clicked by the user. It may be arranged in the region 420.

Here, the second region 420 may include a plurality of detailed regions, and the object arrangement recommendation apparatus 100 arranges the first to nth objects among the detailed regions with reference to the optimal candidate arrangement detailed image. It is possible to display a specific detail area to be displayed in a specific color.

For example, as illustrated in FIG. 3A, a specific detail region to be arranged in the kth rack object 410-k may be displayed in gray shades. However, this is only to assist the user to easily grasp a specific detail area to arrange the k-th rack object 410-k, and the user must place the k-th rack object 410-k in a specific detail area. It is not done.

In addition, even if an area corresponding to the k-th rack object 410-k among the first areas 410 is displayed in a specific color, the user does not necessarily have to select only the k-th rack object 410-k, The second rack object 410-2 may also be selected.

On the other hand, the object arrangement recommendation apparatus 100 is not limited to allowing the user to arrange only the rack object, and as shown in FIG. 3B, the k on the specific rack object arranged in the second area 420 It is also possible to support +1 server object 410-(k+1) to be arranged as a specific server object.

At this time, whether the object arrangement recommendation apparatus 100 arranges the k+1 server object 410-(k+1) in the rack object arranged in the second area 420 is the optimal arrangement. In order to notify the user, as shown in FIG. 3B, a separate display may be performed on a specific rack object among rack objects arranged in the second area 420.

However, this is only a part of various embodiments, and in the first area 410 of the display area 400 of the user terminal, only the object list corresponding to the rack object may be displayed. In this case, in order for the user to add the k+1 server object 410-(k+1) to a specific rack object, the user clicks a specific rack object arranged in the second area 420 to perform the k+1 server. It is also possible to add an object 410-(k+1).

The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and can be recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention or may be known and available to those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROMs, DVDs, and magneto-optical media such as floptical disks. media), and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

In the above, the present invention has been described by specific embodiments, such as specific components, and limited embodiments and drawings, which are provided to help the overall understanding of the present invention, but the present invention is not limited to the above embodiments , Those skilled in the art to which the present invention pertains can make various modifications and variations from these descriptions.

Therefore, the spirit of the present invention is not limited to the above-described embodiment, and should not be determined, and all claims that are equally or equivalently modified with the claims as described below fall within the scope of the spirit of the present invention. Would say

Claims (20)

In the object arrangement recommendation method for allowing a user to arrange an object in an optimized position based on machine learning,
(a) When a plurality of candidate arrangement detailed images are registered in a predetermined storage unit, when a new arrangement detailed image is obtained, the object arrangement recommendation apparatus images the plurality of candidate arrangement detailed images and the new arrangement detailed image Performing analysis processing to calculate arrangement similarity, each of which represents a similar degree to the new arrangement detail image for each of the plurality of candidate arrangement detail images, based on the new arrangement detail image; And
(b) the object arrangement recommendation apparatus determines (i) a specific candidate arrangement detailed image, with reference to the arrangement similarity, among the plurality of candidate arrangement detailed images, the arrangement similarity satisfies a predetermined similarity condition, , (ii) With reference to the power consumption information corresponding to each of the specific candidate arrangement detailed images, among the specific candidate arrangement detailed images, an optimal candidate arrangement detailed image that satisfies a predetermined power consumption condition is determined and the corresponding information is used by the user Providing them;
Including,
Step (a) is,
(a1) A plurality of candidate array data corresponding to each of the plurality of candidate arrangement detailed images, wherein the plurality of candidate array data includes candidate index information indicating respective positions in the candidate arrangement space, and the candidate index information is the candidate Of each location in the arrangement space, the object includes specific candidate index information regarding a specific location in which the object is arranged-in a state in which additional registration in the predetermined storage unit is performed, a new array corresponding to the new arrangement detailed image ( array) data-the new array data includes new index information indicating each position in the new arrangement space, and the new index information is specific to a specific position in which the object is arranged, among each position in the new arrangement space. If new index information is included-is additionally obtained, the object arrangement recommendation apparatus refers to the candidate index information and the new index information, and the number of candidate objects, which is the number of the objects arranged in the candidate arrangement space, and At least one sorting candidate array data in which the number of new objects, which is the number of the objects arranged in the new arrangement space, is determined, and among the plurality of candidate array data, the number of candidate objects is within a threshold range from the number of new objects. Determining; And
(a2) The object arrangement recommendation apparatus performs the image analysis processing on a plurality of selection candidate arrangement detailed images and the new arrangement detailed image corresponding to each of the selection candidate array data, based on the new arrangement detailed image. Calculating the similarity of the selection arrangements, each of which represents a similar degree to the new arrangement detail image for each of the plurality of selection candidate arrangement detail images;
Including,
In step (b),
The object arrangement recommendation apparatus, (i) with reference to the selected arrangement similarity, among the plurality of selected candidate arrangement detailed images, the selected arrangement similarity satisfies the predetermined similarity condition, and selects a specific selected candidate arrangement detailed image. Determining, and (ii) among the specific selection candidate arrangement detailed images, an optimum selection candidate arrangement detail image satisfying the predetermined power consumption condition, with reference to the power consumption information corresponding to each of the specific selection candidate arrangement detail images. Determine as the detailed image of the optimal candidate arrangement and provide the information to the user,
The object includes at least one rack object and at least one server object,
In step (a2),
The object arrangement recommendation apparatus performs the image analysis processing on the plurality of selection candidate arrangement detailed images and the new arrangement detail image, wherein (i) the plurality of selection candidate arrangement detailed images of the rack object and the server object The plurality of selection candidate arrangements are performed by performing the image analysis processing on the selection candidate rack object and the selection candidate server object included in each and (ii) the new rack object and the new server object included in the new arrangement detailed image. For each detailed image, the degree of similarity of each rack object arrangement indicating the degree of similarity between information on the selected candidate rack object included in the plurality of selection candidate arrangement detailed images and information on the new rack object included in the new arrangement detailed image. And each server object arrangement similarity indicating the degree of similarity between the information on the selection candidate server object included in the plurality of selection candidate arrangement detail images and the information on the new server object included in the new arrangement detail image. And calculating the selection arrangement similarity for each of the plurality of selection candidate arrangement detailed images with reference to the rack object arrangement similarity and the server object arrangement similarity. delete According to claim 1,
The plurality of candidate arrangement outline images corresponding to each of the plurality of candidate arrangement detailed images are respectively managed as the plurality of candidate array data and are additionally registered in the predetermined storage,
When a new arrangement outline image is additionally acquired, the object arrangement recommendation apparatus manages the new arrangement outline image as the new array data, respectively.
The plurality of candidate arrangement outline images and the new arrangement outline image include location information regarding a specific location where the object is arranged,
The location information corresponds to the specific candidate index information and the specific new index information, respectively. delete According to claim 1,
The information on the new rack object included in the detailed image of the new arrangement includes information on the location where the new rack object is arranged, the number and size of the new rack objects, and the information on the new server object includes the new A method characterized in that the server object is arranged, information about the number and size of the new server object, the model name of the server object and the year of manufacture. According to claim 1,
(c) The object arrangement recommendation apparatus supports to display an object list corresponding to the first object to the n-th object in a first area of the display area of the user terminal, and allows the user to refer to the detailed image of the best candidate arrangement Supporting a second area of the display area of the user terminal to arrange at least one of the first object to the n-th object;
Method further comprising a. The method of claim 6,
The second region includes a plurality of detailed regions,
In step (c),
The object arrangement recommending apparatus, wherein the first candidate to the n-th object, the specific region to be arranged is displayed in a specific color among the detailed regions with reference to the optimal candidate arrangement detailed image. The method of claim 6,
The first object to the n-th object are (i) a first rack object to a k-th rack object, where k is 1 or more and an integer less than n-and (ii) k+1 server objects to n-th server objects Including,
In step (c),
The object arrangement recommendation apparatus enables the user to arrange the specific rack object of at least one of the first rack object or the k-th rack object in the second area by causing the user to refer to the detailed image of the optimal candidate arrangement. And supporting the arrangement of at least one specific server object from the k+1 server object to the n-th server object to the specific rack object. The method of claim 6,
In step (c),
In the state in which the object arrangement recommendation apparatus supports displaying the object list corresponding to the first object to the n-th object in the first area, the user may remove the first object or the n-th object from the first object. The first object add button to add n-th object button for rendering in the second area-the first object add button to the n-th object add button is displayed in the object list-a specific object add button is selected by the user When the object arrangement recommendation device is provided, the method for supporting the user to arrange a specific object corresponding to the specific object addition button among the first to nth objects in the second area . The method of claim 9,
The specific object is arranged in the second area from the first area through a drag and drop method while the specific object addition button is selected by the user. In the object arrangement recommendation device for allowing a user to arrange an object in an optimized position based on machine learning,
At least one memory for storing instructions; And
Including at least one processor configured to execute the instructions,
The processor, when a plurality of candidate arrangement detailed images are registered in a predetermined storage unit and a new arrangement detailed image is obtained, (1) image analysis processing of the plurality of candidate arrangement detailed images and the new arrangement detailed image A process of calculating arrangement similarity indicating the degree of similarity to the new arrangement detailed image for each of the plurality of candidate arrangement detailed images based on the new arrangement detailed image, and (2) (i) referring to the arrangement similarity As a result, among the plurality of candidate arrangement detailed images, a specific candidate arrangement detailed image that satisfies a predetermined similarity condition is determined, and (ii) power consumption information corresponding to each of the specific candidate arrangement detailed images is determined. For reference, among the specific candidate arrangement detailed images, the instructions for determining an optimal candidate arrangement detailed image that satisfies a predetermined power consumption condition and performing a process of providing corresponding information to a user are executed.
The (1) process is,
(1-1) A plurality of candidate array data corresponding to each of the plurality of candidate arrangement detailed images-The plurality of candidate array data includes candidate index information indicating respective positions in the candidate arrangement space, and the candidate index information is Among each position in the candidate arrangement space, specific object index information related to a specific position in which the object is arranged includes-in a state in which is additionally registered in the predetermined storage unit, a new image corresponding to the new arrangement detail image Array data-The new array data includes new index information indicating each position in the new arrangement space, and the new index information is located at a specific position in which the object is arranged among each position in the new arrangement space. If specific acquisition of new index information regarding-is additionally obtained, the processor refers to the candidate index information and the new index information, and the number of candidate objects that are the number of the objects arranged in the candidate arrangement space and the The number of new objects, which is the number of objects arranged in the new arrangement space, is determined, and among the plurality of candidate array data, at least one sorting candidate array data in which the number of candidate objects is within a threshold range from the number of new objects is selected. The process of determining and (1-2) the processor performs the image analysis processing on a plurality of selection candidate arrangement detailed images and the new arrangement detailed image corresponding to each of the selection candidate array data, so that the new arrangement detail image And a process of calculating the degree of similarity of the sorting arrangement representing the degree of similarity to the detail image of the new arrangement for each of the plurality of selection candidate arrangement detailed images on the basis of,
The processor,
In the above (2) process,
(i) with reference to the sorting arrangement similarity, among the plurality of sorting candidate arrangement detailed images, a specific sorting candidate arrangement detailed image is determined, wherein the sorting arrangement similarity satisfies the predetermined similarity condition, and (ii) the With reference to the power consumption information corresponding to each specific selection candidate arrangement detailed image, among the specific selection candidate arrangement detailed images, the optimum selection candidate arrangement detailed image satisfying the predetermined power consumption condition is used as the optimum candidate arrangement detailed image. Determine and provide the information to the user,
The object includes at least one rack object and at least one server object,
The processor,
In the above (1-2) process,
The image analysis processing is performed on the plurality of selection candidate detail images and the new arrangement detail image, and among the rack objects and the server objects, (i) the selection candidates included in each of the plurality of selection candidate arrangement detail images The image analysis processing of the rack object and the candidate server object and (ii) the new rack object and the new server object included in the new arrangement detailed image is performed to perform the plurality of selection candidate arrangement detailed images for each of the plurality of candidate objects. Each rack object arrangement similarity and the plurality of selection candidate arrangements indicate the degree of similarity between the information about the selection candidate rack object included in the selection candidate arrangement detail image and the information about the new rack object included in the new arrangement detail image. Each server object arrangement degree of similarity indicating the degree of similarity between the information on the selection candidate server object included in the detailed image and the information on the new server object included in the new arrangement detailed image is calculated, and the rack object arrangement similarity and The object arrangement recommendation apparatus, characterized in that the selection arrangement similarity is calculated for each of the plurality of selection candidate arrangement detailed images with reference to the server object arrangement similarity. delete The method of claim 11,
The plurality of candidate arrangement outline images corresponding to each of the plurality of candidate arrangement detailed images are respectively managed as the plurality of candidate array data and are additionally registered in the predetermined storage,
When a new arrangement outline image is additionally acquired, the processor manages each of the new arrangement outline images as the new array data,
The plurality of candidate arrangement outline images and the new arrangement outline image include location information regarding a specific location where the object is arranged,
And the location information corresponds to the specific candidate index information and the specific new index information, respectively. delete The method of claim 11,
The information on the new rack object included in the detailed image of the new arrangement includes information on the location where the new rack object is arranged, the number and size of the new rack objects, and the information on the new server object includes the new An object arrangement recommendation apparatus comprising information on a location where a server object is arranged, the number and size of the new server objects, a model name of the server object, and a manufacturing year. The method of claim 11,
(3) The processor supports to display an object list corresponding to the first object to the n-th object in the first area of the display area of the user terminal, and allows the user to refer to the detailed image of the optimal candidate arrangement. A process of supporting at least one of the first object to the n-th object to be arranged in a second area of the display area of the user terminal;
Object arrangement recommendation device further comprising a. The method of claim 16,
The second region includes a plurality of detailed regions,
The processor,
In the above (3) process,
An object arrangement recommendation apparatus characterized in that a specific detail region in which the first object or the n-th object is to be arranged is displayed in a specific color among the detail regions with reference to the optimal candidate arrangement detail image. The method of claim 16,
The first object to the n-th object are (i) a first rack object to a k-th rack object, where k is 1 or more and an integer less than n-and (ii) k+1 server objects to n-th server objects Including,
The processor,
In the above (3) process,
Supporting the user to refer to the detailed image of the optimal candidate arrangement, to assist in arranging at least one specific rack object among the first rack object or the k-th rack object in the second area, and the specific rack object In the object arrangement recommendation device, characterized in that to support the arrangement of at least one specific server object from the k+1 server object to the n-th server object. The method of claim 16,
In the above (3) process,
In the state in which the processor supports displaying the object list corresponding to the first object to the n-th object in the first area, the user may place the first object or the n-th object in the second area. When a specific object add button is selected by the user, the first object add button to add n-th object button for rendering-the first object add button to the n-th object add button is displayed in the object list. The processor, the object arrangement recommendation device, characterized in that to support the user to arrange a specific object corresponding to the specific object addition button from the first object to the n-th object in the second area. The method of claim 19,
The specific object, the object arrangement recommendation device, characterized in that arranged in the second region from the first region through a drag and drop (drag and drop) method, while the specific object add button is selected by the user .

Images