KR101610377B1 - System and Method for Managing Intellactual Power and Recording Medium - Google Patents

Abstract

The present invention relates to an intelligent power supply control method and system, and a recording medium therefor. More particularly, the present invention relates to a method of controlling power of a server system constituted by N (N> = 1) servers according to traffic for each service function of each server (1) classifying one or more service functions provided in each server into M (M > = 1) service function categories, connecting the M service function category information and N server information classified by each service function (2) calculating at least one reference traffic information corresponding to each of the M service function categories, and maintaining the service quality of the service function at or above a predetermined reference value in each reference traffic Storing (or updating) resource usage information for each reference traffic for each of the plurality of traffic categories in association with each category information, and (3) (1 < = m < = M) in which traffic per service function is less than or equal to a preset reference value by classifying / analyzing traffic for the N servers identified through the traffic detection means provided in the apparatus, (4) checking (or calculating) the resource usage information for each of the m service functions corresponding to the traffic for each of the m service functions through the storage medium, (N '&lt; = N) servers having at least one server (N' &lt; = N), and (5) checking resource usage and resource availability for each server, (1 < = n < = N) servers migrating the resource usage corresponding to the service function among the N 'servers to the resource availability of the other servers; and (6) n By controlling the power to the system, excluding the target server comprises the step of excluding from the server system.

Power control

Description

TECHNICAL FIELD [0001] The present invention relates to an intelligent power control method and system,

1 is a diagram illustrating a configuration of an intelligent power supply control system according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a traffic analysis process for each service function according to an embodiment of the present invention.

3 is a diagram illustrating a process of excluding one or more servers in a server system according to service-function-specific traffic that is confirmed in real time according to an embodiment of the present invention.

4 is a diagram illustrating a process of inputting one or more servers excluded from a server system according to service-function-specific traffic that is confirmed in real time according to an embodiment of the present invention.

5 is a diagram illustrating a process of configuring a scheduling policy for excluding one or more servers in a server system according to an embodiment of the present invention.

6 is a diagram illustrating a process of excluding one or more servers in a system according to a scheduling policy according to an embodiment of the present invention.

7 is a diagram illustrating a process of configuring a scheduling policy for putting one or more servers into a server system according to an embodiment of the present invention.

8 is a diagram illustrating a process of putting one or more servers into a system according to a scheduling policy according to an embodiment of the present invention.

9 is a diagram illustrating a configuration of an intelligent power supply control system according to another embodiment of the present invention.

10 is a diagram illustrating a traffic analysis process for each service function according to another embodiment of the present invention.

11 is a diagram illustrating a process of excluding one or more servers in a server system according to service-function-based traffic that is confirmed in real time according to another embodiment of the present invention.

12 is a diagram illustrating a process of inputting one or more servers excluded from the server system according to service-function-specific traffic that is confirmed in real time according to another embodiment of the present invention.

DESCRIPTION OF THE EMBODIMENTS

100: terminal 105: network connection device

110: traffic detection means 115: intelligent power supply control system

120: power source control means 125: server identification means

130: Scheduling policy processing means 135: Information registration means

140: resource analysis means 145: traffic analysis means

150: Storage medium 155: Scheduling D / B

160: Server

The present invention provides a server system comprising N (N> = 1) servers having one or more service functions classified into M (M> = 1) service function categories, wherein m = 1 < = n < = N) in which the traffic for each service function decreases among the N servers when the traffic corresponding to the number of service functions decreases, M '(1 <= m' <= M) of the M service functions, when n servers among the N servers are excluded from the server system in accordance with the reduction of traffic for each of m service functions, (N '<= n) servers having a service function for increasing the traffic for each service function among the n servers excluded from the server system when the traffic corresponding to the service function increases, 'N' '(n' '<= n') of 'n' servers It controls the source to provide a method and system for input to the server system.

With the continuous development of information and communication technology, it has become essential to maintain the quality of service for the Internet service over a certain period of time as the Internet service is activated. Therefore, most of the servers providing the Internet service are called the Internet Data Center Infrastructure is leased and offered.

The costs that the Internet service providers pay in response to the IDC lease are not only the labor costs such as the server management cost but also the electricity for the power used by the servers and the additional equipment (e.g., air conditioner, network connection device, etc.) Includes fees. Here, it is difficult to reduce the labor cost as a fixed ratio, while the electricity rate can be reduced by controlling the power according to the server usage rate.

However, in the conventional method of controlling the use of the server, the administrator periodically monitors the traffic of each power control target server identified through the network connection device, blocks the power of the server remotely according to the monitoring result, And if the power control target server is provided with two or more plural service functions, it is very difficult to provide a method of controlling the power by controlling the system through the manager as well as the method.

An object of the present invention to solve the above problem is to provide a server system configured with N (N> = 1) servers having one or more service functions classified into M (M> = 1) service function categories (1 < = n < = M) in which the traffic for each service function decreases among the N servers when traffic corresponding to m N in the server system to be excluded from the server system.

Another object of the present invention is to provide a server system comprising N (N > = 1) servers having one or more service functions classified into M (M > = 1) service function categories, (1 < = m < = N) servers are excluded from the server system, if traffic corresponding to m '(1 <= m' <= M) service functions among the M service functions increases, (N '<= n) servers having a service function in which traffic for each service function increases among n servers excluded from the server system, and n' '(n' &lt; = n &quot;) servers and puts them into the server system.

A method for controlling the power of a server system constituted by N (N> = 1) servers according to the present invention according to traffic for each service function of each server is as follows: (1) &Gt; = 1) service function categories, storing the M service function category information and the N server information classified by each service function in a storage medium, and (2) And stores the resource usage information for each reference traffic in association with each category information for maintaining the service quality of the service function at a predetermined reference value or higher in each reference traffic (3) storing (or updating) the traffic for the N servers identified through the traffic detection means provided in the network connection device (1 < = m < = M) traffic for each service function, wherein traffic for each service function is less than or equal to a predetermined reference value; (4) (Or calculating) m resource usage information for each of the m service functions corresponding to the m service functions, and for each of N '(N' <= N) servers having one or more m service functions, (5) checking the resource usage amount corresponding to the service function among the N 'servers through the resource usage amount and the resource availability amount for each server, (1 < = n < = N ') number of system migration target servers migrating to the n number of system exclusion target servers, and (6) It characterized by comprising.

(7) When n (1 <= n <= N) servers out of the N servers are excluded from the server system, the intelligent power supply control method according to the present invention may include: M '(1 <= m' <= M) service function traffic whose service function-specific traffic is equal to or greater than a preset reference value by classifying / analyzing traffic for the (Nn) (N '<= n) servers having at least one m' service function among the n servers through the storage medium, (or calculating) (m "<= m ') resource usage information for each service function; and (9) determining a maximum resource capacity of the m' (N "&lt; = n ") number of system input target servers; and (10) Characterized in that by controlling the power of the n "of the system excluding the target server resolver comprises a step of input on the server system.

A method for controlling the power of a server system constituted by N (N> = 1) servers according to the present invention according to traffic for each service function of each server is as follows: (1) &Gt; = 1) service function categories, storing the M service function category information classified and the N server information for each service function in a storage medium, and (2 ' Category, and when the reference traffic is equal to or less than a preset reference value, n pieces of system exclusion target server information to be excluded from the server system are stored in the storage medium in association with each category information And (3) transmitting traffic for the N servers identified through the traffic detection means provided in the network connection device to the M service functions (1 < = m < = M) traffic for each service function, wherein traffic for each service function is less than or equal to a preset reference value; (4-5) (1 < = n < = N ') system exclusion target servers associated with the reference traffic matched with the traffic for each service function, and (6) And excluded from the server system.

The intelligent power supply control method according to the present invention is characterized in that in the state where n (1 <= n <= N) servers are excluded from the server system by the step (2 " N '(n' <= n) system input target server information to be input to the server system among the n servers in the storage medium in association with each category information; (N) servers among the N servers are excluded from the server system, (7) the traffic for the (Nn) servers identified through the traffic detection means included in the network connection device is divided into (1 < = m &lt; = M) traffic for each service function, wherein traffic for each service function is equal to or greater than a preset reference value; (8-9) The number of the above-identified m ' (N "< = n &quot;) system input target servers associated with the reference traffic matched with the peaks, and (10) controlling power of the n" And a step of injecting the gas into the combustion chamber.

A method for controlling the power of a server system constituted by N (N> = 1) servers according to the present invention according to traffic for each service function of each server is as follows: (1) &Gt; = 1) service function categories, storing the M service function category information and the N server information classified by each service function in a storage medium, and (2) And stores the resource usage information for each reference traffic for maintaining the service quality of the service function at a predetermined reference value or more in each reference traffic with each category information, (3) when n (1 <= n <= N) servers out of the N servers are excluded from the server system, (7) (1 &lt; = m < = &lt; m) &gt;, which is greater than or equal to a preset reference value, by classifying / analyzing traffic for (Nn) servers identified by the traffic detection means provided in M) service functions, and (8) checking the traffic for each of n '(n' <= n) servers having at least one m 'service functions among the n servers through the storage medium (Or calculating) m "(m" <= m ') resource usage amount information for each service function matched with the provided service function; and (9) (N "&lt; = n ") number of system input target servers having a maximum resource capacity equal to or greater than a predetermined resource usage amount, and (10) controlling power of the n" And injecting it into the system. .

Meanwhile, a method of controlling the power of a server system constituted by N (N> = 1) servers according to the present invention according to traffic for each service function of each server is as follows: (1) (M > = 1) service function categories, storing the M service function category information and N server information classified for each service function and storing them in a storage medium; and (2 ' (1 &lt; = n < = N) servers out of the N servers are excluded from the server system, Storing n '(n' <= n) system input target server information to be input to the server system in the storage medium in association with each category information when the predetermined value is equal to or greater than a preset reference value; The N (N) the traffic for the (Nn) servers identified through the traffic detection means provided in the network connection device is classified / analyzed according to the M service function categories Checking m '(1 &lt; = m < = M) traffic for each service function whose traffic per service function is equal to or greater than a preset reference value; (8-9) (N "< = n ') system connection target servers associated with the identified m' number of service function-specific traffic matches with the reference traffic, and (10) And controlling the power of the server system to be input to the server system.

In the intelligent power supply control method according to the present invention, the step (1) includes classifying a plurality of service functions provided in the N servers according to a predetermined service function classification criterion, 1) number of service function categories, and N server information having one or more service functions and M service function category information are interconnected on the basis of respective service functions and stored in a storage medium.

In the intelligent power supply control method according to the present invention, the step (2) may include: detecting traffic for a predetermined period of time for the N servers through the traffic detection unit for each predetermined analysis unit time, Analyzing user log information generated / accumulated by the M function service function for a predetermined period of time by analyzing / analyzing user log information generated by the M / N "out of the n servers excluded from the server system or a reference value to be input into the server system &lt; RTI ID = 0.0 &gt; Setting a maximum weight for the maximum traffic, setting the minimum traffic and the setting Setting q (q &gt; 2) or more traffic intervals using the reference value and the maximum traffic applied with the weight as traffic point reference points and calculating a reference point of each traffic interval as reference traffic, Calculating a resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount to provide a plurality of terminals within a predetermined time limit; And storing (or updating) the reference traffic information and the resource usage information for each reference traffic in association with the category information corresponding to the service function and storing the resource information in the storage medium And the like.

The intelligent power supply control method according to the present invention is characterized by further comprising the steps of: checking traffic for a predetermined period of time for the N servers through the traffic detection unit for each predetermined analysis unit time, and generating user log information Statistically processing the traffic for each of the M service functions for a predetermined period on the basis of at least one of the time zone, day, day of the week, week, and month, and comparing the analyzed traffic with the identified traffic by the analysis unit time, Analyzing M traffic patterns for each service function for a certain period of time and determining traffic decrease period information including at least one of time zone, day, day, week, and month, (4) to (5) or (4-5), the step And a system elimination scheduling policy information including the traffic reduction interval information, the traffic information of the traffic reduction interval and the n number of system exclusion target server information, and transmits the system exclusion scheduling policy information to the scheduling D / When the system exclusion scheduling policy information that excludes n system exclusion target servers from the system in a predetermined traffic decrease interval is stored in the scheduling D / B, Checking whether a traffic reduction period of the system exclusion scheduling policy information has been reached through the scheduling D / B, determining whether a traffic decrease period of the system exclusion scheduling policy information is reached through the N servers Lt; RTI ID = 0.0 &gt; M &lt; / RTI & Checking whether the traffic for each service function identified and analyzed matches the traffic information of the system exclusion scheduling policy information, and if the traffic for each service function matches the traffic information of the system exclusion scheduling policy information, (4) to (4), if the traffic for each service function is not matched with the traffic information of the system exclusion scheduling policy information, controlling the power of n system excluding servers included in the scheduling policy information to be excluded from the server system, And controlling the power of the n system exclusion target servers identified through the step (5) or (4-5) to be excluded from the server system.

The intelligent power supply control method according to the present invention is characterized by further comprising the steps of: checking traffic for a predetermined period of time for the N servers through the traffic detection unit for each predetermined analysis unit time, and generating user log information Statistically processing the traffic for each of the M service functions for a predetermined period on the basis of at least one of the time zone, day, day of the week, week, and month, and comparing the analyzed traffic with the identified traffic by the analysis unit time, The M traffic service function traffic patterns for each service function for a certain period of time and the traffic increase interval information including at least one of time, day, day, week, and month, (8) to (9) or (8-9), the step The system input scheduling policy information including the traffic increase period information, the traffic information of the traffic increase period, and n "number of system input target server information is established, and the scheduling D / B, the system exclusion scheduling policy information for excluding n system exclusion target servers from the system in a predetermined traffic increase interval of the N servers is stored in the scheduling D / B Checking whether a traffic increase period of the system exclusion scheduling policy information has been reached through the scheduling D / B; and (Nn) checking whether a traffic increase period of the system exclusion scheduling policy information is reached through the scheduling D / Traffic for each of the M service function categories Checking whether the traffic for each service function identified and analyzed matches the traffic information of the system exclusion scheduling policy information, and if the traffic for each service function matches the traffic information of the system exclusion scheduling policy information, (8) to (8), if the traffic for each service function is not matched with the traffic information of the system exclusion scheduling policy information, the control unit controls the power of the "n " system exclusion target servers included in the scheduling policy information, Further comprising the step of controlling the power of the n "system input target servers identified through the step (9) or (8-9) and inputting the control signals to the server system.

In the intelligent power control method according to the present invention, the step (3) may include: checking traffic for a predetermined unit time with respect to the N servers through the traffic detection unit, generating / And comparing the analyzed user log information with the identified traffic for each unit time, and verifying M traffic for each service function during the unit time.

In the intelligent power control method according to the present invention, the step (7) may include: checking traffic for a predetermined unit time with respect to the (Nn) servers through the traffic detection unit, And comparing / analyzing the user log information generated / accumulated by the M function service function with the identified traffic for each unit time to identify M service function traffic for the unit time.

In the intelligent power control method according to the present invention, the step (6) may include: checking whether the n servers normally respond to a ping; if the n servers respond to a ping, Service request from the plurality of terminals to the n servers and checking whether the in-service deactivation is normally performed; and when the in-service deactivation is confirmed, The method of claim 1, further comprising the steps of: determining whether the service function is terminated normally; and shutting down the n servers when the service function for the n servers is terminated, And confirming whether or not it is not responding to the request.

In the intelligent power supply control method according to the present invention, the step (10) includes the steps of: checking whether a server of n "" does not respond to a ping; periodically confirming whether or not the power of the " n "servers responds to the ping; and if the" n &quot; servers respond to the ping, And activating an in-service request from the plurality of terminals to the n servers through the network connection device when the service function is executed, And confirming whether the service activation is normally performed.

On the other hand, a computer-readable recording medium having a program for executing the intelligent power source control method described above is recorded.

A system for controlling the power of a server system constituted by N (N> = 1) servers according to the present invention according to traffic for each service function of each server comprises N server information and a service function (M > = 1) service function category information for each service function; and a storage medium for storing one or more reference traffic information corresponding to the M service function categories and resource usage information for each reference traffic (Or update) the traffic for the N servers identified through the traffic detection means included in the network connection device to the M service function categories Traffic analysis that identifies m (1 <= m <= M) traffic for each service function whose traffic by service function is less than preset reference value (Or estimating) resource usage information for each of m service functions corresponding to the m service function traffic through the storage medium, and calculating N '(N' <= N) servers corresponding to the service function among the N 'servers through the resource usage amount and the resource availability for each server identified by the resource analyzing means, (1 < = n < = N ') of server exclusion target servers for transferring the resource amount of the other server to the resource availability of another server, And power supply control means for excluding the power supply from the power supply control means.

In the intelligent power control system according to the present invention, the traffic analyzing unit classifies / analyzes traffic for (Nn) servers in association with the traffic detecting unit for each of the M service function categories, (1 &lt; = m < = M) traffic functions of a plurality of service functions that are equal to or greater than a set reference value, wherein the resource analyzing unit analyzes m ' (Or calculating) m "(m" <= m ') resource use amount information for each service function matched with a service function provided in n' (n '<= n) servers having at least one service function (N "&lt; = n ') number of system input destination servers having the maximum resource capacity equal to or larger than the resource usage amount of each m" number of service functions among the n' number of servers Identification feature To, and further it characterized in that the power control means, to control the power of the identified n "of the system excluding the target server to input to the server system.

Meanwhile, a system for controlling power of a server system constituted by N (N > = 1) servers according to the present invention according to traffic for each service function of each server, = 1) service function categories, a storage medium for storing the M service function category information and the N server information classified and linked for each service function, and a storage medium for storing one or more Information registration means for calculating n standard traffic information and storing n system elimination target server information to be excluded in the server system when the reference traffic is equal to or less than a predetermined reference value in association with each category information in the storage medium; The traffic for the N servers identified through the traffic detection means provided in the connection device is transmitted to the M services (1 &lt; = m < = M) traffic functions for which the traffic for each service function is less than or equal to a preset reference value; Server verification means for identifying n (1 < = n < = N) system exclusion target servers associated with the reference traffic matched with the function-specific traffic, and controlling the power of the n And power supply control means which is excluded from the server system.

In the intelligent power supply control system according to the present invention, the information registering means may calculate at least one reference traffic information corresponding to each of the M service function categories, and determine n (1 < = n & (N '<= n) system input target server information to be input to the server system when the reference traffic is equal to or greater than a preset reference value in the exclusion state in the server system in association with each category information (Nn), which is confirmed through traffic detection means provided in the network connection apparatus, when n servers among the N servers are excluded from the server system, The traffic for each of the M service function categories is classified and analyzed so that the traffic for each service function is m '(1 &lt; = 1) m '<= M) service function-specific traffic, and the server identification unit identifies the m'-pieces of service function-specific traffic among the n' (N "&lt; = n ") number of system entry target servers associated with the traffic, and the power source control means controls power of the n" To the system.

A system for controlling the power of a server system constituted by N (N> = 1) servers according to the traffic of each service function of each server according to the present invention comprises N server information and service functions of the N servers A storage medium for storing M (M > = 1) service function category information for each service function, and storing one or more reference traffic information corresponding to the M service function categories and resource usage information for each reference traffic, (1 < = n < = N) servers out of the N servers are excluded from the server system, (Nn) servers identified by the detected traffic detection means by the M service function categories and classifies / analyzes the traffic for each of the M service function categories, (M ') of m' (1 &lt; = m '&lt; = M) service functions, resource analyzing means for identifying (or calculating) m "(m" <= m ') resource usage information for each service function matched with a service function provided in (n' <= n) (N "&lt; = n ") number of system input target servers having a maximum resource capacity equal to or larger than the resource usage amount of each m" service function among the servers, And power control means for controlling the power of the server and inputting the power to the server system.

Meanwhile, a system for controlling power of a server system constituted by N (N > = 1) servers according to the present invention according to traffic for each service function of each server, = 1) service function categories, a storage medium for storing the M service function category information and the N server information classified and linked for each service function, and a storage medium for storing one or more (N > = n) of the N servers, when the reference traffic is equal to or greater than a preset reference value in the exclusion state in the server system, (n '<= n) system entry target server information in association with each category information and storing the information in the storage medium; When excluded from the server system, the traffic for the (Nn) servers identified through the traffic detection means included in the network connection device is classified / analyzed according to the M service function categories, (1 &lt; = m &lt; = M) m 'service function traffic of the m' service function among the n 'servers through the storage medium, (N "< = n &quot;) system input target servers associated with the reference traffic, and a power control unit for controlling power of the n" And means for controlling the operation of the apparatus.

In the intelligent power control system according to the present invention, the information registering unit classifies a plurality of service functions provided in the N servers according to a predetermined service function classification criterion, and classifies M (M > = 1) N server information having one or more service functions and M service function category information are interconnected on the basis of respective service functions and stored in a storage medium.

In the intelligent power control system according to the present invention, the traffic analyzing means may identify traffic for a predetermined period of time for the N servers through the traffic detecting means, for each predetermined analysis unit time, And analyzing M traffic for each service function for a predetermined period by comparing / analyzing user log information generated / accumulated with the identified traffic for each analysis unit time, N "out of the n servers excluded from the server system or a reference value for excluding n servers out of the N servers from the server system based on the highest traffic and the lowest traffic among the M service- Set the reference value to be input to the server system, and set the maximum traffic (Q > 2) or more traffic sections set with the minimum traffic and the set reference value as traffic section reference points as reference traffic, Resource usage information for each reference traffic is calculated by applying a predetermined weight to a resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount to be provided to the terminal, (Or updates) the resource usage information for each reference traffic in association with the category information corresponding to the service function and stores the information in the storage medium.

The intelligent power control system according to the present invention may be configured to statistically process M traffic for each service function for a predetermined period identified by the traffic analyzing unit based on at least one of time, day, day, week, and month, A traffic reduction period information including at least one of a time zone, a day, a week, a week, and a month in which traffic for each service function is less than a preset reference value, traffic for the traffic reduction period, Wherein the scheduling policy processing means includes a scheduling policy setting means for setting, in the scheduling D / B, the traffic decrease period information, the traffic information of the traffic decrease period, and the traffic reduction period information, After n pieces of system exclusion target server information are stored in association with each other, And controlling power of n system excluding servers stored in the scheduling D / B when the traffic for each service function checked through the traffic analyzing unit matches the traffic information stored in the scheduling D / B, And if the traffic for each service function identified through the traffic analyzing means does not match the traffic information stored in the scheduling D / B, the power of the n system elimination target servers, which are confirmed in real time through the resource analyzing means and the server checking means, So that the server system is excluded.

The intelligent power control system according to the present invention may be configured to statistically process M traffic for each service function for a predetermined period identified by the traffic analyzing unit based on at least one of time, day, day, week, and month, The server 10 checks the system input target server of the traffic increase section information including at least one of the time zone, the day, the week, the week, and the month in which the traffic for each service function is equal to or greater than the preset reference value, Wherein the scheduling policy processing means includes scheduling policy processing means for scheduling the scheduling policy to include traffic increase interval information, traffic information of the traffic increase interval, and n " After the information is stored in association with each other, when the traffic increase period is reached, When the traffic for each service function to be matched with the traffic information stored in the scheduling D / B is matched, the power of the n "number of system exclusion target servers stored in the scheduling D / B is controlled and put into the server system, Quot; n "number of system exclusion target servers that are confirmed in real time through the resource analyzing means and the server checking means, if the traffic for each service function identified through the resource analyzing means matches the traffic information stored in the scheduling D / To the system.

The operation principle of the preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings and description. It should be understood, however, that the drawings and the following description are provided for illustrative purposes only and are not intended to limit the scope of the present invention. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The terms used below are defined in consideration of the functions of the present invention, which may vary depending on the user, intention or custom of the operator. Therefore, the definition should be based on the contents throughout the present invention.

As a result, the technical idea of the present invention is determined by the claims, and the following embodiments are merely means for effectively explaining the technical idea of the present invention to a person having ordinary skill in the art to which the present invention belongs Only.

1 is a diagram illustrating a configuration of an intelligent power supply control system 115 according to an embodiment of the present invention.

1 shows a server 160 system including N (N> = 1) servers 160 having one or more service functions classified into M (M> = 1) categories, wherein M (1 < = m < = M) of service functions decreases among the N services 160 among the N servers 160, the service function of the service function- (1 &lt; = n &lt; = N) servers 160 according to the resource usage and resource availability of N '(N' <= N) (M ') of the M service functions when n servers 160 of the N servers 160 are excluded from the server 160 according to the reduction of traffic for each of the m service functions, When the traffic corresponding to 1 <= m '<= M service functions increases, n servers 160 excluded from the server 160 system, (N "<= n) according to the resource usage and resource availability of the n '(n' <= n) servers 160 having the service function whose service function- ) Of the server 160 to control the power of the server 160 and input the same to the server 160. Those skilled in the art will be able to refer to FIG. 1 and / (E.g., some of the components may be omitted, or broken down, or combined) with respect to the configuration of the intelligent power supply control system 115, but the present invention is not limited to any of the above- And the technical characteristics thereof are not limited only by the method shown in FIG. Hereinafter, although the roles of the plurality of means constituting the intelligent power supply control system 115 in the method shown in FIG. 1 are different, means having similar functions will be described by the same means, It is clear that it is not limited.

According to the present invention, the intelligent power control system 115 is provided on a server 160 system composed of one or more network connection devices 105 and N servers 160, The server 105 is connected to the plurality of terminals 100 and the N servers 160. The terminal 105 communicates traffic between the plurality of terminals 100 and the N servers 160 in response to the communication connection, And a detection means (110).

The network connection apparatus 105 connects a plurality of terminals 100 requesting a communication connection to the N servers 160 through a communication network on the server 160 system and a communication channel between the N servers 160 (LVS), or a Linux Virtual Server (LVS), which connects communication channels between the plurality of terminals 100 and the N servers 160,

The traffic detecting means 110 included in the network connection apparatus 105 may be configured such that when the network connection apparatus 105 connects and maintains communication channels between the plurality of terminals 100 and the N servers 160 The traffic detected by the traffic detection means 110 is detected by the plurality of terminals 100 and the N servers 160, (For example, IP address to IP address traffic or IP address to IP address for a specific communication port) exchanged between the specific terminal 100 and the specific server 160 among the N servers 160 Address-to-address traffic) amount.

The N servers 160 constituting the server 160 system include a server 160 providing a service function based on a plurality of terminals 100 requesting communication connection through the network connection device 105, , And each server 160 is provided with an application corresponding to one or more service functions.

Here, the service function may be a web service function, a mail service function, a messenger service function, a mini homepage service function, a blog service function, a cafe service function, a knowledge exchange service function, a search service function, an SMS dispatch service function, (Including an internal service function including one or more of a user service function including one or more functions, a development service function, a beta test service function, and an intranet service function) .

Accordingly, when the N servers 160 are classified according to service functions, the web server 160, the mail server 160, the messenger server 160, the mini homepage server 160, the blog server 160, The server 160, the knowledge exchange server 160, the search server 160, the SMS sending server 160, the discussion server 160, the development server 160, the beta test server 160, and the intranet server 160 Can be classified.

According to the present invention, the physical server 160 constituting the N servers 160 is not provided with only one service function and is provided with at least one server 160. [

For example, the server 1 of the N servers 160 may have a web service function, a mini homepage service function, a cafe service function, a knowledge exchange service function, and a search service function, A service function, a knowledge exchange service function, and a search service function. The server 3 may have a mail service function. The server 4 may include a messenger service function, a web service function, a mail service function, A service function can be provided.

For example, in the case of the server 4, the main service function is a messenger service function related to a chat, a note, and a file exchange using a messenger. However, when providing the web page to the terminal 100 through the messenger service, And further includes the mail service function when the messenger is interlocked with the web mail account, and may further include the SMS dispatch service function when the SMS is sent through the messenger.

If the server 3 is developing a new mail service, the server 3 may have a mail service function, a development service function, and a beta test service function, and the server 4 may also include a messenger service function, a development service function, Function can be provided.

According to the present invention, the N servers 160 are connected to the network 100 through the network connection device 105, Information is generated and stored in the storage means. The user log information includes log generation time information and log data.

According to the present invention, the N servers 160 may determine a maximum resource capacity for CPU resources and memory resources, a process / thread usage amount according to software components, A maximum resource capacity for a session allocation amount is set. In a process of providing a communication connection of the plurality of terminals 100 through the network connection apparatus 105 and one or more service functions, A resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount.

The intelligent power control system 115 that controls the power of the server 160 system configured by the N servers 160 according to the traffic for each service function of each server 160 may be configured to control the N servers Category information corresponding to M (M > = 1) service function categories classified according to a predetermined service function classification criterion and service information that can be provided through a server 160 system constituted by the service server 160, The storage medium 150 may include a storage medium 150 for storing information of N servers 160 connected to each other on the basis of respective service functions or may be linked through a communication network. One or more reference traffic information corresponding to each of the M service function categories, and at least one reference traffic information for maintaining service quality of the service function in each reference traffic at a predetermined reference value or more It is preferable that the resource usage information for each reference traffic is stored in association with each category information. In FIG. 1, the storage medium 150 is shown as a single medium for convenience, but the present invention is not limited thereto.

The category information corresponding to the M (M > = 1) service function categories is information obtained by classifying the respective services provided by the N servers 160 constituting the server 160 system according to functions, One service function category may be matched with any one service function provided by the N servers 160. [

According to an embodiment of the present invention, the category information may include a category code or an identifier for any one of the service functions provided by the N servers 160.

The N server 160 information includes identification information for identifying a physical server 160 constituting the server 160 system and one or more services corresponding to one or more service functions provided in the server 160 And function information.

Herein, the identification information of the server 160 includes a name of the server 160, a server 160 address, a server 160, and the like, which are assigned to the respective servers 160 to identify / identify the N servers 160, It is preferable to include an identifier.

Preferably, the service function information of the server 160 information includes a service function classification code or an identifier for identifying a service function of one director provided in each server 160, It is preferable to match with the information.

The reference traffic information is used to compare traffic of the N servers 160 confirmed through the traffic detection unit 110 with user log information generated / accumulated by the N servers 160 (Q > 2) or more traffic segments including the lowest traffic and the highest traffic (e.g., the traffic with the weight predetermined to the highest traffic) The traffic is used as a reference value for excluding n servers 160 among the N servers 160 constituting the server 160 system from the server 160 system, N 'number of servers 160 included in the n' number of servers 160 having the service function among the n number of the servers 160 excluded from the system are used as reference values to be inputted into the server 160 system.

The resource usage information for each reference traffic may be transmitted to a plurality of terminals 100 requesting a service function generating the reference traffic for a unit time in which the reference traffic is maintained, The amount of memory usage, the amount of process / thread usage, and the amount of session allocation of each server 160 for maintaining the service (QoS) (= providing the service function within a predetermined time limit) .

According to an embodiment of the present invention, the resource usage information for each reference traffic is calculated based on a traffic change amount (or an increase / decrease ratio) for each service function for a predetermined period of time based on the average of the month, (Or highest) service change amount (or increase / decrease ratio) for each of the reference traffic, calculates a weight to be applied to the resource usage amount information for each reference traffic, and stores the calculated weight as the CPU usage amount, the memory usage amount, And a value that is applied to the first and second images.

For example, if the average (or maximum) traffic change amount (or increase / decrease ratio) per service function in the month, week, day, and time zone to which the resource usage information for each reference traffic is calculated is 10% The resource usage information for each reference traffic is provided to each of the plurality of terminals 100 corresponding to the reference traffic to maintain the quality of service (QoS) for the service function (= providing the service function within a predetermined time limit) 160, and the CPU usage, the memory usage, the process / thread usage, and the session allocation amount of 10%.

According to the embodiment of the present invention, the CPU usage amount of the resource usage amount information per reference traffic may include the CPU operation ratio of each server 160, but preferably the CPU usage amount Speed (or computation speed).

Also, the memory usage amount of the reference traffic-specific resource usage information may include a memory usage ratio of each server 160, but preferably includes a memory capacity for maintaining the QoS Do.

According to the present invention, the intelligent power control system 115 may include traffic reduction period information for one or more service function-specific traffic, traffic information of the traffic reduction period, and n system elimination target server 160 information System scheduling policy information including system exclusion scheduling policy information, traffic increase interval information for traffic per one or more service function, traffic information of the traffic increase interval, and n "system input target server 160 information, And a scheduling D / B 155 for storing policy information, or interworking through a communication network.

The traffic reduction period information may include traffic for a certain period of time for the N servers 160 identified by the traffic detection means 110 for a certain period of time, A traffic pattern for each service function calculated by comparing / analyzing the log information is statistically processed based on at least one of a predetermined time zone, day, day, week, and month, Day, week, month, and month that is less than a preset reference value.

According to the embodiment of the present invention, the traffic reduction period information does not directly specify a specific period such as the time zone, day, day of week, week, month, The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

For example, if there is a pattern in which the traffic corresponding to the mail service function drops below a predetermined reference value on the day of rain, the traffic reduction section information can not directly specify the rainfall date, And setting the traffic reduction period to the traffic reduction period.

Preferably, the traffic information of the traffic reduction period includes a traffic amount and a traffic increase / decrease ratio that decrease within a predetermined range during the traffic decrease period, which is decreased to less than the set reference value in the traffic decrease period.

The n system elimination target server 160 information in the traffic reduction interval includes resource usage information for each service function corresponding to the reduced traffic in the traffic decrease interval and information on the amount of traffic corresponding to the reduced traffic among the N servers 160 (For example, an amount obtained by subtracting the resource usage from the maximum resource capacity of N 'servers 160) for the N' servers 160 having service functions It is preferable to include information on the number of servers 160.

The traffic increase period information may include traffic for a certain period of time for the N servers 160 that are confirmed through the traffic detection unit 110 for a certain period of time, A traffic pattern for each service function calculated by comparing / analyzing the log information is statistically processed based on at least one of a predetermined time zone, day, day, week, and month, And includes section information corresponding to at least one of a time zone, a day, a day of the week, a week, and a month, which is equal to or greater than a preset reference value.

According to the embodiment of the present invention, the traffic increase section information does not directly specify a specific section such as the time zone, day, day of week, week, and month, and increases the traffic pattern by service function The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

For example, if there is a pattern in which a traffic corresponding to the discussion service function increases to a predetermined reference value or more on the day of a certain group gathering, the traffic increase section information can not directly designate the day of the meeting And information for setting the meeting date and time as the traffic increase section if the group meeting is predicted.

The traffic information of the traffic increase section may include a traffic amount and a traffic increase / decrease ratio which are increased within a predetermined range during the traffic increase period by increasing to a value exceeding the set reference value in the traffic increase section.

The n pieces of system exclusion target server 160 information of the traffic increase period include information on resource usage amount for each service function corresponding to the traffic increased in the traffic increase period and correspondence to the increased traffic among the N servers 160 (For example, an amount obtained by subtracting the resource usage from the maximum resource capacity of N 'servers 160) for the N' servers 160 having service functions It is preferable to include information on the number of servers 160.

1, the intelligent power control system 115 for excluding n servers 160 among the N servers 160 from the server 160 may include N servers 160 constituting the server 160 system, M service function category information classified into M (M > = 1) service function categories and N server 160 information for each service function in the storage medium 150 And one or more reference traffic information calculated corresponding to each of the M service function categories and resource usage information for each reference traffic for maintaining the service quality of the service function at a predetermined reference value or higher in each reference traffic, Information registration means 135 for storing (or updating) the information on the storage medium 150 in association with the category information of the network connection device 105 and the traffic detection means 110 provided in the network connection device 105 The traffic for the N servers 160 is classified / analyzed according to the M service function categories to check traffic for each of the M service functions, and if the traffic among the M services for each service function is less than a preset reference value and traffic analysis means 145 for confirming m (1 <= m <= M) traffic for each service function.

The information registering unit 135 receives service function classification criteria for one or more service functions provided in the N servers 160 through the administrator terminal 100 managing the intelligent power control system 115 (M> = 1) service function categories for each service function by classifying a plurality of service functions provided in the N servers 160 according to the service function classification criteria, (M > = 1) configured by classifying a plurality of service functions provided in the N servers 160 according to a predetermined service function classification criterion, at least one service function included in the N servers 160, N service 160 category information having at least one service function, and M service function category information configured at each service function, And is stored in the storage medium 150. [0033] FIG.

The traffic analyzing means 145 periodically checks traffic for a predetermined unit time for the N servers 160 through the traffic detecting means 110 and generates traffic by the N servers 160 And comparing / analyzing accumulated user log information with the identified traffic for each unit time, and verifying M traffic for each service function during the unit time.

In case that the n servers 160 having the service function with the service function-specific traffic of the service function among the N servers 160 are checked in real time and excluded from the server 160 system, The means 145 identifies m (1 < = m < = M) traffic for each service function whose traffic is less than or equal to a preset reference value among the traffic for each of the M service functions identified.

In order to store (or periodically update) the reference traffic information and the resource usage information for each reference traffic in the storage medium 150, the traffic analyzing unit 145 analyzes the N traffic information (160) for each predetermined unit time, and comparing / analyzing user log information generated / accumulated by the N servers (160) with the identified traffic for each analysis unit time And analyzes the traffic for each M service functions for a predetermined period.

When the traffic for each M service functions is analyzed for the predetermined period, the information registering unit 135 registers the N servers 160 based on the highest traffic and the lowest traffic among the M service function traffic for the predetermined period, N servers 160 having n service servers 160 among the n servers 160 excluded from the server 160 system or reference values for excluding n servers 160 from the server 160 Quot; n "number of servers 160 to the server 160 system.

When the reference value is set, the information registering unit 135 applies a predetermined weight to the maximum traffic, sets q (q (q)) as the traffic section reference point by using the minimum traffic, the set reference value, > 2) traffic sections are set, and reference points of each traffic section are calculated as reference traffic.

Here, the weight applied to the maximum traffic may be a power control according to the service function-specific traffic even if the traffic per service function exceeds the maximum traffic according to the amount of traffic change (or increase / decrease ratio) And it may be omitted according to the intention of the person skilled in the art.

Wherein the traffic interval is obtained by dividing the interval between the maximum traffic and the maximum traffic by q (q> 2) or more with the lowest traffic as the lowest reference point and the highest traffic with the weight applied as the highest reference point, The reference point of the section for dividing the traffic section may include the set reference value. That is, the traffic interval does not necessarily have to be equal intervals, and the lowest traffic, the reference value, and the highest traffic are periodically updated according to traffic for each service function for a certain period analyzed through the traffic analyzing unit 145.

When the reference traffic is calculated, the information registering unit 135 registers a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session usage amount to provide the service function to a plurality of terminals 100 within a predetermined time limit, The resource usage information including at least one of the quota amounts is calculated and the resource usage information for each reference traffic is calculated by applying a predetermined weight to the calculated resource usage amount.

Herein, the weight applied to the resource usage may be a weighted average of the amount of traffic per service function exceeding the maximum traffic according to the amount of traffic change (or increase / decrease ratio) (For example, booting of the server 160 and loading of the service function) for putting the service function into the server 160 system in advance within a predetermined time limit, Value. &Lt; / RTI &gt;

The information registering unit 135 stores the calculated reference traffic information and the resource usage information for each reference traffic in association with the category information corresponding to the service function in the storage medium 150, The reference traffic and the resource usage information for each reference traffic are periodically updated according to the traffic analysis for each service function of the base station.

1, an intelligent power control system 115 that excludes n servers 160 among the N servers 160 from a server 160 is connected to the m services via the storage medium 150, (Or estimates) m resource usage information for each service function corresponding to the function-specific traffic, and determines (or calculates) the resource usage information for each service function for the N '(N' <= N) (160) through the resource utilization amount of each server (160) and the resource availability of each server (160), and a resource analysis unit (140) (1 <= n <= N ') server elimination target servers 160 for transferring the resource usage corresponding to the service function to the resources available for the other servers 160 (125), and controlling the power of the n system elimination target servers (160) 0) system. The power supply control means 120 includes:

In case that the n servers 160 having the service function with the service function-specific traffic of the service function among the N servers 160 are checked in real time and excluded from the server 160 system, When the m-number of service function-specific traffic with a service function-specific traffic value equal to or less than a predetermined reference value is identified by the means 145, the resource analysis means 140 transmits the identified m number of service function- ), And confirms m pieces of resource usage information for each service function associated with the identified reference traffic information.

At this time, it is rare that the identified m traffic for each service function and the reference traffic corresponding to each service function coincide with each other. Therefore, the resource analyzing unit 140 checks the traffic interval matched with the reference traffic information including the identified m service-specific traffic from the storage medium 150, It is preferable to calculate the m-th resource usage information for each service function corresponding to the identified intermediate value among the resource usage information for each service function associated with the reference traffic.

The resource analyzing unit 140 reads information of N servers 160 stored in the storage medium 150 and stores N '(N') 's having at least one m service functions among the N servers 160, (= N) servers 160 and confirms the resource usage amount of each server 160 and the resource availability amount of each server 160 for the N '

The resource usage amount of each server 160 and the maximum resource capacity of each server 160 can be checked by a resource manager included in each server 160. The resource availability of each server 160 can be checked by the server 160 ) Of the server 160. In this case,

When the resource usage amount of each server 160 and the resource availability amount of each server 160 for the N 'servers 160 are confirmed, the server 160 checking means 125 confirms each of the servers 160 The resource usage amount corresponding to the service function among the N '' servers 160 is transferred to the available resource amount of the other server 160 according to the resource usage amount and the resource availability by the server 160 n (1 < = n &lt; = N ') system elimination target servers 160.

According to an embodiment of the present invention, the resource redistribution policy may include a priority to transfer the resource usage corresponding to the service function among the N 'servers 160 having the m service functions to another server 160 (Or order) of the server 160. In this case, the server 160 checking means 125 checks the resource usage amount of each server 160 and the resource availability amount per server 160 The resource allocation information for transferring the resource usage of the server 160 to which the priority is set to the resource availability of the other server 160 having no priority or lower priority according to the resource redistribution information, It is desirable to identify the n number of system exclusion target servers 160 that have been eliminated.

Alternatively, the resource redistribution policy may be configured such that the resource usage corresponding to the service function among the N 'servers 160 having the m service functions is allocated to one or more servers 160 In this case, the server 160 checking unit 125 may determine whether or not the server 160 has the priority set according to the resource usage amount for each server 160 and the resource availability for each server 160 And allocates the resource usage information of the other server 160 having the lower priority or the lower priority to the resource allocation information of the other server 160, It is desirable to check the user interface 160.

According to another embodiment of the present invention, the resource redistribution policy may be configured such that a resource usage of a server 160 having only a service function corresponding to a subset of the m service functions among the N ' The server 160 includes policy information for transferring resources to the available capacity of the server 160 having a service function corresponding to the combination of the m service functions. In this case, ) Of the server 160 having only a subset of the m service functions among the N 'servers 160 according to the resource usage of each server 160 and the resource availability of each server 160 The resource redistribution information for transferring the usage amount to the resource available amount of the server 160 having the redundancy, and for calculating the resource redistribution information based on the resource redistribution information, To check (160) it is preferred.

According to another embodiment of the present invention, the resource redistribution policy may be configured such that the resource usage of the server 160 having the smallest resource capacity among the N 'servers 160 having the m service functions, In this case, the server 160 checking unit 125 may determine that the maximum resource capacity of the large server 160 is greater than the maximum resource capacity of the N 'servers 160 It is preferable to identify n system elimination target servers 160 that transfer all of the resource usage among the small servers 160 to the server 160 having the largest resource capacity.

Alternatively, the resource redistribution policy may be configured such that the resource usage of the server 160 having the smallest resource capacity among the N 'servers 160 having the m service functions is used as the resource availability of the server 160 having the large resource usage In this case, the server 160 checking means 125 determines whether the available resource amount is smaller than the maximum available resource amount of the N 'servers 160 and the available resource amount It is preferable to check n system elimination target servers 160 that transfer all of the resource usage among the servers 160 to the server 160 having a large resource capacity.

Those skilled in the art will be able to refer to / modify the resource redistribution policy described above to further deduce various types of additional resource redistribution policies, I will clarify the inclusion.

According to another embodiment of the present invention, the resource analyzing unit 140 reads the N servers 160 information stored in the storage medium 150 and stores the m service functions among the N servers 160 (N '<= N) servers 160 having at least one N' number of servers 160, checking the resource usage amount of each of the N 'servers 160 by the server 160, It is possible to confirm the resource availability for each server 160 that subtracts the resource usage for each server 160 from the resource usage for each service function and the maximum resource capacity for each server 160.

According to an embodiment of the present invention, when a resource manager provided in each server 160 provides resource usage information for each process / thread, the resource usage amount for each service function is managed by a resource manager Lt; / RTI &gt;

According to another embodiment of the present invention, a unit resource usage rate (or a unit resource usage rate) per unit time for each service function (e.g., corresponding to a processor / thread included in a corresponding service function in the process of developing each service function) , The resource usage amount for each service function is determined by checking the resource usage amount for each server 160 for each unit time for each server 160, (Or reads) the user log information for the unit time or the type and number of processes / threads generated in the server 160 corresponding to each service function during the unit time, (Or unit resource utilization rate) and the identified service function type It may be estimated by comparing / analyzing the product of the number of available service function and the server 160 by resource usage.

When the resource analysis unit 140 analyzes the resource usage by each server 160 and the resource usage by the service function and the resource availability by the server 160, the server 160 checking unit 125 checks The resource usage amount corresponding to the service function among the N '' servers 160 according to the resource redistribution policy predetermined through the resource usage by the server 160, the resource usage by the service function, And the n system elimination target servers 160 to be transferred to the resource availability of the server 160 are confirmed.

According to an embodiment of the present invention, the resource redistribution policy may be configured such that a priority for transferring resource usage for each service function among the N 'servers 160 having the m service functions to another server 160 is set In this case, the checking unit 125 of the server 160 confirms the resource usage amount of each service 160 of each server 160 and the resource availability of each server 160 Resource allocation information for transferring the resource usage amount of each service function of the server 160 for which the priority is set to the resource availability of another server 160 having no priority or low priority according to the amount of resource, It is desirable to identify the n number of system exclusion target servers 160 to which all the usage amounts have been transferred.

Alternatively, the resource redistribution policy may include information on at least one server 160 having priority to receive the resource usage for each service function among the N 'servers 160 having the m service functions from another server 160 In this case, the server 160 checking means 125 determines whether the priority is set according to the resource usage amount of each service function of each server 160 and the resource availability amount of each server 160 The resource redistribution information for transferring the resource usage amount of each service function of the other server 160 having no or low priority to the resource amount of the server 160 is calculated and the resource redistribution information for transferring the resource usage amount of n It is desirable to confirm the system exclusion target server 160.

According to another embodiment of the present invention, the resource redistribution policy may include a resource redistribution policy for each service function of the server 160 having only a service function corresponding to a subset of the m service functions among the N ' The server 160 includes policy information for transferring the amount of usage to the resource available amount of the server 160 having the same service function but further including a service function corresponding to the combination of the m service functions. In this case, The means 125 may include only a subset of the m service functions among the N 'servers 160 according to the resource usage of each service function of the server 160 and the resource availability of each server 160 The server 160 calculates resource redistribution information for transferring the resource usage by the service function of the server 160 to the resource availability of the server 160 having the round-off, and, based on the resource redistribution information, To make all the n previous systems excluding the destination server 160, the amount is preferred.

According to another embodiment of the present invention, the resource redistribution policy may be configured to maximize the resource usage of each service function of the server 160 having the smallest resource capacity among the N 'servers 160 having the m service functions, The server 160 checking means 125 may check the maximum value of the resource capacity of the server 160 in accordance with the maximum resource capacity of the N ' It is desirable to identify n system exclusion target servers 160 that transfer all of the resource usage among the servers 160 having a small resource capacity to the server 160 having the maximum resource capacity.

Alternatively, in the resource redistribution policy, the resource usage amount of the service 160 of the server 160 having the smaller resource capacity among the N 'servers 160 having the m service functions is set to the resource In this case, the server 160 checking means 125 may determine whether the resources (resources) of the N ' servers 160 have been changed based on the maximum resource capacity and the resource availability, It is preferable to check n system elimination target servers 160 that transfer all of the resource usage from the server 160 having a small capacity to the server 160 having a large resource capacity.

Those skilled in the art will be able to refer to / modify the resource redistribution policy described above to further deduce various types of additional resource redistribution policies, I will clarify the inclusion.

When the n server elimination target servers 160 are confirmed by the server 160 confirmation means 125, the power source control means 120 controls the power of the n system elimination target servers 160 And is excluded from the server 160 system.

According to an embodiment of the present invention, the power control unit 120 controls power of the n system elimination target servers 160 to be excluded from the server 160 system, (160) normally responds to the ping.

If the n servers 160 respond to the ping, the power control unit 120 receives the paging service requested from the plurality of terminals 100 through the network connection unit 105 to the n servers 160 deactivates the in-service request and verifies that the in-service deactivation is operating normally.

If the in-service disabling is confirmed, the power control unit 120 terminates the service functions provided in the n servers 160, checks whether the service function is normally terminated, The power of the n servers 160 is shut off and then the n servers 160 are checked whether they are not responsive to the ping.

For example, if the network connection apparatus 105 is an L4 switch, if the power control unit 120 succeeds in establishing a telnet connection through predetermined L4 connection information (for example, IP, account, password, etc.) Connect to the L4 switch through the authentication phase (but for stability, one L4 switch allows 10 simultaneous configuration telnet connections), and when connected to the L4 switch, the configuration of the L4 switch .

- conf t

- mod con <moduleNo>

- serverfarm <farmname>

- real [ip]

- no in service

- end

- wr

- exit

Thereafter, the power source control means 120 receives the &quot; slb real | [ip] &quot;, and if the status value is confirmed as &quot; OUTOFSERVICE &quot;, the in service for the L4 switch is regarded as inactivated.

Alternatively, if the network connection apparatus 105 is an LVS (Linux Virtual Server), the power control unit 120 updates an active option of the LVS conf file and reloads the one or more servers, Exclude from the server system.

That is, the power control unit 120 checks whether the actual ip is active through the LVS. If it is in the active state, the power control unit 120 backs up the corresponding config file, sets the actual ip active state to 0, and applies the change through the command of "/etc/init.d/pulse reload" , And if one of the above processes fails, the backup file is restored.

1, the intelligent power control system 115 for excluding n servers 160 among the N servers 160 from the server 160 may be configured to determine a schedule Week, month, and month of the service function for each of the M service function functions during a time period, a day, a week, a week, and a month And a scheduling policy for associating and storing n number of system exclusion target servers 160 with respect to the traffic of each of the traffic decrease periods and the respective traffic reduction periods and associated with the scheduling D / B 155 Wherein the scheduling policy processing means 130 includes a scheduling D / B 155 and a scheduling policy processing means 130. The scheduling policy processing means 130 includes a scheduling D / The traffic information of the service function identified through the traffic analyzing unit 145 is transmitted to the scheduling D / B 155 (step &lt; RTI ID = 0.0 &gt; The control unit 145 controls the power of the n system elimination target servers 160 stored in the scheduling D / B 155 to be excluded from the server 160 system, (N), which is confirmed in real time through the resource analyzing unit 140 and the server 160 checking unit 125, if the traffic for each service function, which is confirmed through the checking unit 125, does not match the traffic information stored in the scheduling D / The power of the system exclusion target server 160 is controlled to be excluded from the server 160 system.

When the traffic for each M service functions for a certain period identified through the traffic analysis unit 145 is confirmed, the scheduling policy processing unit 130 determines whether the N servers 160 ), And compares / analyzes the user log information generated / accumulated by the N servers (160) with the identified traffic for each analysis unit time to determine a schedule Statistically processed traffic patterns of M service functions for a certain period of time based on at least one of time zone, day, day, week, and month, A time, a day, a day, and a time when traffic per one service is reduced to a preset reference value or less or the traffic per service function is increased to a predetermined reference value or more, Check at least one day, week, or month.

According to the embodiment of the present invention, the scheduling policy processing unit 130 checks the traffic for a predetermined period of time for the N servers 160 through the traffic detection unit 110 for each predetermined analysis unit time , User log information generated / accumulated by the N servers (160) is compared / analyzed with at least one of external information such as weather information, temperature information, humidity information, and event information for each analysis unit time, The traffic for each service function is statistically processed on the basis of at least one of weather, temperature, humidity, and event, and traffic patterns for each of the M service functions for the statistically processed period are analyzed, Temperature, humidity, or event in which the traffic per service function increases above a predetermined reference value Check the above.

According to the method of the present invention, the external condition information may further include an external condition for determining the traffic for each service function in addition to the weather information, the temperature information, the humidity information, and the event information, The present invention is not limited thereto.

If at least one of the time zone, day, day of week, week, and month in which the traffic for each service function decreases to a preset reference value or less is detected, the scheduling policy processing unit 130 updates the M- The resource analyzing unit 140 and the server 160 checking unit 125 may check the traffic decrease period information including at least one of the time zone, day, day, week, (N) system exclusion target servers 160 for each traffic reduction period, and the system including the traffic decrease period information, traffic information of the traffic decrease period, and n system exclusion target server 160 information And configures exclusion scheduling policy information and stores it in the scheduling D / B 155.

If at least one of the weather, temperature, humidity, and the event in which the traffic for each service function decreases to a predetermined reference value or less is detected, the scheduling policy processing unit 130 updates the M- A traffic reduction section for dynamically setting a traffic reduction period including at least one of weather, temperature, humidity, and events less than or equal to a predetermined value, and traffic in the traffic reduction period, The n number of system exclusion target servers 160 for each traffic reduction interval through the checking means 125 and the traffic information of the traffic decrease period and the traffic information of the n system exclusion target servers 160, And configures and stores the system exclusion scheduling policy information in the scheduling D / B 155.

If the system exclusion scheduling policy information for excluding n system elimination target servers 160 from the N servers 160 in the predetermined traffic reduction period is stored in the scheduling D / B 155, The policy processing unit 130 confirms whether the traffic reduction period of the system exclusion scheduling policy information has been reached through the scheduling D / B 155 and is confirmed through the traffic analysis unit 145 when the traffic reduction period arrives And checks whether the traffic for each service function matches the traffic information of the system exclusion scheduling policy information.

If the traffic for each service function is matched with the traffic information of the system exclusion scheduling policy information, the scheduling policy processing unit 130 determines power of n system exclusion target servers 160 included in the system exclusion scheduling policy information If the traffic for each service function does not match the traffic information of the system exclusion scheduling policy information, the resource analyzing unit 140 and the server 160 checking unit 125 are checked The n system elimination target servers 160 are checked and the power of the n system elimination target servers 160 checked is excluded from the server 160 system.

Referring to FIG. 1, when n (1 <= n <= N) servers 160 among the N servers 160 are excluded from the server 160 system, n The intelligent power supply control system 115 for inputting the plurality of servers 160 from the server 160 system is provided with Nn servers 160 that are confirmed through the traffic detection means 110 provided in the network connection apparatus 105 ) Traffic analyzing means for classifying / analyzing the traffic for each of the M service function categories to identify m '(1 <= m' <= M) traffic for each service function whose traffic for each service function is equal to or greater than a preset reference value 145).

When n (1 <= n <= N) servers 160 of the N servers 160 are excluded from the server 160 system, the traffic analyzing means 145 determines that the traffic detection means 110 (Nn) number of servers 160 and transmits the user log information generated / accumulated by the (Nn) servers 160 to the Nn number of servers 160 And comparing the analyzed traffic with the traffic for each of the M service functions for the unit time.

If the traffic for each of the M service functions for the unit time is identified, the traffic analyzing unit 145 calculates m '(1 < = m '<= M) traffic for each service function.

According to the embodiment of the present invention, the m 'pieces of traffic for each service function include a boot time at which n' 'servers 160 having m' service functions are booted and a boot time at which the booted server 160 The service quality of the service-specific traffic is confirmed to be maintained in the waiting-for-waiting time in consideration of an input waiting time including an application load time at which the service functions are loaded, So that the quality of service of the traffic for each service function is maintained in the waiting time for the service.

Referring to FIG. 1, an intelligent power control system 115 for inputting n "servers 160 of the n servers 160 from a server 160 is connected to the n M "(m" <= m ') matched with a service function provided in n' (n '<= n) servers 160 having at least one m' A resource analyzing unit 140 for identifying (or calculating) resource usage information for each service function, and a resource analyzing unit 140 for analyzing (or calculating) resource usage information for each of n ' (n &gt; " n) " number of system entry target servers 160 by checking the power of the n & And a power supply control means (120) for supplying power to the system.

When the m 'pieces of traffic for each service function are identified through the traffic analyzing unit 145, the resource analyzing unit 140 analyzes m' of the n servers 160 among the n servers 160 excluded from the server 160 system, (N '<= n) servers 160 having at least one service function, and the service function matching the service function provided in the n' servers 160 among the m ' (m &lt; m &gt; &lt; m &quot;) service functions, compares the identified m "service function traffic with each reference traffic stored in the storage medium 150, Check (or calculate) the usage information.

When the m 'number of resource usage information for each service function is confirmed (or estimated), the server 160 checking unit 125 checks the maximum resource capacity for the n' servers 160, Quot; n "(n" = " n &quot;) input into the system according to the resource input information is calculated by putting the capacity into the server 160 system for each service function, n 'number of system input target servers 160 are confirmed.

When the n "number of system entry target servers 160 are confirmed by the server 160 confirmation means 125, the power source control means 120 determines the power of the n" number of system exclusion target servers 160 And inputs it to the server 160 system.

According to the embodiment of the present invention, the power source control means 120 controls the power of the n "number of system input target servers 160 to be input from the server 160 system, (160) does not respond to a ping.

If the n "number of servers 160 do not respond to the ping, the power control unit 120 periodically checks whether the power of the n" servers 160 is responsive to the ping, When the "number of servers 160 responds to the ping, the service functions provided in the n servers 160 are executed in a predetermined order, and it is confirmed whether the service function is normally executed.

If the service function is executed, the power control unit 120 receives an in-service request from the plurality of terminals 100 to the n servers 160 through the network connection unit 105, And confirms that the in-service activation is normally operating.

For example, if the network connection apparatus 105 is an L4 switch, if the power control unit 120 succeeds in establishing a telnet connection through predetermined L4 connection information (for example, IP, account, password, etc.) Connect to the L4 switch through the authentication phase (but for stability, one L4 switch allows 10 simultaneous configuration telnet connections), and when connected to the L4 switch, the configuration of the L4 switch .

- conf t

- mod con <moduleNo>

- serverfarm <farmname>

- real [ip]

- in service

- end

- wr

- exit

Thereafter, the power source control means 120 receives the &quot; slb real | [ip] &quot;, and if the status value is confirmed as &quot; INSERVICE &quot;, it is regarded that the service for the L4 switch is activated.

Alternatively, if the network connection apparatus 105 is an LVS (Linux Virtual Server), the power control unit 120 updates an active option of the LVS conf file and reloads the one or more servers, Server system.

That is, the power control unit 120 checks whether the actual ip is active through the LVS. If it is not active, the power control unit 120 backs up the corresponding config file, sets the actual ip active state to 1, and applies the change through the command "/etc/init.d/pulse reload" If one of the above processes fails, the backup file is restored.

Referring to FIG. 1, an intelligent power control system 115 that inputs n "number of servers 160 among the n servers 160 in the server 160 system is checked through the traffic analysis unit 145 Day, week, month, and the like, based on at least one of the M, M, and C service functions for a predetermined period of time, The scheduling policy processing means 130 for checking the system input target server 160 of the traffic increase period of the traffic increase interval and storing it in the scheduling D / B 155 in association with each other, The scheduling policy processing means 130 may include the traffic increase period information, the traffic information of the traffic increase period, and the n "number of systems After the information on the destination server 160 is stored in association with the traffic information, traffic for each service function identified through the traffic analysis unit 145 is matched with the traffic information stored in the scheduling D / Of the system exclusion target server 160 stored in the scheduling D / B 155 and inputs the control information to the server 160 system to control the power of the n & N 'number of system exclusion target servers (not shown) that are confirmed in real time through the resource analyzing means 140 and the server 160 checking means 125 when the traffic information of the corresponding system is not matched with the traffic information stored in the scheduling D / 160 to the server 160 system.

Alternatively, if at least one of a time zone, a day, a week, a week, and a month in which the traffic for each service function increases to a predetermined reference value or more is detected, the scheduling policy processing unit 130 transmits the M service- The traffic analyzing unit 140 and the server 160 check the traffic increase period information including at least one of time, day, day, week, and month, which is equal to or greater than a preset reference value, Quot; system entry target server 160 for each traffic increase interval through the network increase request 125, traffic information of the traffic increase interval information and the traffic increase interval, and n " And stores it in the scheduling D / B 155. The scheduling D /

If at least one of weather, temperature, humidity, and event that the traffic for each service function increases to a predetermined reference value or more is detected, the scheduling policy processing unit 130 updates the traffic for each of the M service functions according to a predetermined criterion The traffic analyzing unit 140 and the server 160 determine traffic increase period information that dynamically sets a traffic increase period including at least one of weather, temperature, humidity, N "number of system input target servers 160 for each traffic increase period through the checking means 125, traffic information of the traffic increase period information and the traffic increase interval, and n" 160), and stores the information in the scheduling D / B 155.

If the system exclusion scheduling policy information that excludes n system exclusion target servers 160 for a predetermined traffic increase period from the N servers 160 is stored in the scheduling D / B 155, the scheduling policy The processing unit 130 checks whether the traffic increase period of the system exclusion scheduling policy information is reached through the scheduling D / B 155, and when the traffic increase period is reached, Function-based traffic matches the traffic information of the system exclusion scheduling policy information.

If the traffic for each service function is matched with the traffic information of the system exclusion scheduling policy information, the scheduling policy processing unit 130 notifies the power of the n "system exclusion target servers 160 included in the system exclusion scheduling policy information , And inputs the traffic to the server 160 system. If the traffic for each service function is not matched with the traffic information of the system exclusion scheduling policy information, the traffic analysis unit 145 transmits n " Checks the server 160, and controls the power of the n "number of system input target servers 160 to be input in the server 160 system.

FIG. 2 is a diagram illustrating a traffic analysis process for each service function according to an embodiment of the present invention.

2, the intelligent power control system 115 shown in FIG. 1 analyzes the traffic for a predetermined period of time for the N servers 160 for a predetermined period of time, One or more reference traffic including a reference value to be excluded from the server 160 in the server 160 and resource usage information for each reference traffic and storing the information in the storage medium 150, Those skilled in the art will be able to refer to and / or modify the FIG. 2 to understand various methods of analyzing traffic for each service function (for example, some steps may be omitted, However, it should be understood that the present invention includes all of the above-described methods of practicing the invention, Jing is not limited.

Referring to FIG. 2, the intelligent power control system 115 classifies a plurality of service functions provided in the N servers 160 according to a predetermined service function classification criterion by information setting of the administrator terminal 100 , M (M > = 1) service function categories for each service function, and N server 160 information having at least one service function and M service function category information, (200), the traffic identified by the traffic detection unit 110 of the network connection device 105 is analyzed for each unit time, and the traffic for each of the M service functions is confirmed (step 205).

While the traffic for each of the M service functions is confirmed, the intelligent power control system 115 analyzes M service-function-based traffic for a certain period of time and transmits n servers 160 among the N servers 160 The reference traffic for the q (q> 2) or more traffic sections including the reference value to be excluded from the server 160 system and the lowest traffic and the highest traffic as reference points is calculated 210 and the service traffic By applying a predetermined weight to resource usage including at least one of CPU usage, memory usage, process / thread usage, and session allocation for providing to a plurality of terminals 100 within a predetermined time limit, (215).

If the reference traffic and the resource usage for each reference traffic are calculated 220, the intelligent power control system 115 associates the reference traffic information and the resource usage information for each reference traffic with the category information corresponding to the service function (Or periodically updates) it in the storage medium 150 (225).

FIG. 3 is a diagram illustrating a process of excluding one or more servers 160 from a server 160 according to service-function-based traffic that is confirmed in real time according to an embodiment of the present invention.

In more detail, FIG. 3 illustrates a case where traffic corresponding to m (1 < = m < = M) service functions among the M service functions is decreased according to traffic analysis of M service functions according to the process shown in FIG. (N '<= N) servers 160 having service functions in which the traffic for each service function among the N servers 160 is reduced to a predetermined reference traffic or less, And controlling the power of n (1 <= n <= N) servers 160 to be excluded from the server 160 system. Those skilled in the art will appreciate that, , It may be possible to refer to and / or modify the FIG. 3 to various implementations of the exclusion process of the server 160 (e.g., omitting some steps or changing the order) Include all practices that are inferred W is made, the technical features are not limited only with the exemplary method shown in the figure 3.

Referring to FIG. 3, when the traffic for each of the M service functions is analyzed according to the process shown in FIG. 2, the intelligent power control system 115 analyzes M service function traffic for a predetermined unit time, (300) the traffic for each of the m service functions whose traffic decreases below the set reference value, and transmits resource usage information for each reference traffic connected to the reference traffic corresponding to the m service function traffic from the storage medium 150 (305).

In addition, the intelligent power control system 115 may check resource usage and resource availability for each server 160 for N 'servers 160 having at least one of the m service functions (or for each server 160) The resource usage information for each of the N '' servers 160 and the resources available for each server 160, and the resource availability of each server 160, The system exclusion target server 160 is confirmed (315).

If the n system elimination target servers 160 are confirmed 320, the intelligent power supply control system 115 controls the power of the n system elimination target servers 160, (325).

4 is a diagram illustrating a process of inputting one or more servers 160 excluded from the server 160 system according to service-function-specific traffic that is confirmed in real time according to an embodiment of the present invention.

4, n servers 160 among the N servers 160 according to the traffic analysis of M service functions according to the traffic analysis of M service functions according to the process shown in FIG. 2 (1 <= m '<= M) service functions among the M service functions are excluded from the server 160 system, n (N '<= n) servers 160 having the service function of which the service-specific traffic increases by more than a predetermined reference traffic among n number of servers 160 and n' n "<= n ') power of the server 160 to the server 160 system. As a person skilled in the art, And / or modifying the server 160, It is to be understood that the invention may be practiced otherwise than as specifically described herein, but it is to be understood that the invention may be practiced otherwise than as specifically described herein, Its technical characteristics are not limited.

Referring to FIG. 4, when traffic for each of the M service functions is analyzed according to the process shown in FIG. 2, the intelligent power control system 115 analyzes M service function traffic for a predetermined unit time, (400), and checks n 'number of servers 160 having m' service functions from the storage medium 150 (405 ), And checks resource usage information for each m 'service functions matched with service functions provided in the n' servers 160 (410).

In addition, the intelligent power control system 115 identifies n "system entry target servers 160 having a maximum resource capacity of more than m"

If the n "system input target server 160 is confirmed 420, the intelligent power supply control system 115 controls power of the n" ) System (425).

5 is a diagram illustrating a process of configuring a scheduling policy for excluding one or more servers 160 in a server 160 according to an embodiment of the present invention.

In more detail, FIG. 5 illustrates a traffic reduction period in which traffic decreases below a predetermined reference value according to the traffic analysis of M service functions according to the process shown in FIG. 2 or FIG. 10, The system exclusion scheduling policy information including n pieces of system exclusion target server 160 information is constructed and stored in the storage medium 150. As a person skilled in the art, , It is possible to refer to and / or modify FIG. 5 to infer various implementations of the scheduling policy configuration process (e.g., omitting some steps or changing the order) All the methods of implementation are included, and the technical features thereof are not limited only by the method shown in FIG.

Referring to FIG. 5, when the traffic for each of the M service functions is analyzed according to the process shown in FIG. 2, the intelligent power control system 115 calculates M traffic for each service function for a predetermined period, , Day of the week, week, month, and so on to determine a traffic reduction period including at least one of time, day, day, week, and month in which the traffic for each service function decreases below a predetermined reference value (500), and n system elimination target servers 160 for the traffic reduction period are calculated (505).

According to the embodiment of the present invention, the traffic reduction period information does not directly specify a specific period such as the time zone, day, day of week, week, month, The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

If n system exclusion target servers 160 for the traffic reduction period are estimated 510, the intelligent power control system 115 determines whether the traffic reduction period information, the traffic of the traffic reduction period, A system exclusion scheduling policy including system exclusion target server 160 information is configured (515), and the configured system exclusion scheduling policy information is stored (520) in the scheduling D / B 155.

FIG. 6 is a diagram illustrating a process of excluding one or more servers 160 in a system according to a scheduling policy according to an embodiment of the present invention.

6, when the system exclusion scheduling policy is configured through the process shown in FIG. 5, n server exclusion target servers 160 are excluded from the server 160 according to the system exclusion scheduling policy Those skilled in the art will be able to refer and / or modify FIG. 6 to illustrate various implementations of the server 160 exclusion process (e.g., some steps omitted The present invention is not limited to the above-described embodiments, but may be embodied in many other specific forms without departing from the spirit or essential characteristics thereof.

Referring to FIG. 6, when the system exclusion scheduling policy is configured through the process shown in FIG. 5, the intelligent power control system 115 reads the system exclusion scheduling policy information and checks whether the system exclusion scheduling policy information reaches the traffic increase period 600).

If the traffic increase period is reached (605), the intelligent power control system 115 analyzes M service-specific traffic for a unit time, identifies m service-function-based traffic that is less than a preset reference value, ), And checks whether the traffic of each of the m service functions and the traffic of the traffic increase period is matched (615).

If the traffic matches (620), the intelligent power control system 115 controls the power of n system elimination target servers 160 included in the system exclusion scheduling policy information and excludes it from the system (625) .

FIG. 7 is a diagram illustrating a process of configuring a scheduling policy for putting one or more servers 160 into a server 160 according to an embodiment of the present invention.

In more detail, FIG. 7 illustrates a traffic increase period in which traffic increases to a predetermined reference value or more according to traffic analysis of M service functions according to the process shown in FIG. 2 or FIG. 10, The system input scheduling policy information including the n "system input target server 160 information is stored in the storage medium 150, It is possible to refer to and / or modify the FIG. 7 to various implementations of the scheduling policy configuration process (e.g., omitting some steps or changing the order). However, And the technical features thereof are not limited only by the method shown in FIG. 7 .

Referring to FIG. 7, when the traffic for each of the M service functions is analyzed according to the process shown in FIG. 2, the intelligent power control system 115 calculates M traffic for each service function for a predetermined period, , Weekday, week, and month, and checks the traffic increase period including at least one of time, day, day, week, and month in which the traffic for each service function increases to a predetermined reference value or more (700) and calculates n "number of system input target servers 160 for the traffic increase period (705).

According to the embodiment of the present invention, the traffic increase section information does not directly specify a specific section such as the time zone, day, day of week, week, and month, and increases the traffic pattern by service function The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

If n "system input target servers 160 for the traffic increase period are calculated 710, the intelligent power control system 115 calculates the traffic increase interval information, the traffic of the traffic increase interval, and the estimated n (715) a system input scheduling policy including information on the number of system input target servers (160), and stores the configured system input scheduling policy information in the scheduling D / B 155 (720).

FIG. 8 is a diagram illustrating a process of putting one or more servers 160 into a system according to a scheduling policy according to an embodiment of the present invention.

8, when the system input scheduling policy is configured through the process shown in FIG. 7, n "system input target servers 160 are inputted to the server 160 system according to the system input scheduling policy 8 is a flowchart illustrating a process of inputting the server 160 according to an exemplary embodiment of the present invention. Referring to FIG. 8, The present invention is not limited to the above-described embodiments, but may be practiced otherwise than as specifically described herein.

Referring to FIG. 8, when the system input scheduling policy is configured through the process shown in FIG. 7, the intelligent power control system 115 reads the system input scheduling policy information and determines whether it reaches the traffic increase period 800).

If the traffic increase period is reached (805), the intelligent power control system 115 analyzes the traffic for each of the M service functions for a unit time to check m ' 810) and compares the traffic for each of the m 'service function functions with the traffic for the traffic increase interval to check if it matches (815).

If the traffic matches (820), the intelligent power control system 115 controls the power of the n "number of system input target servers 160 included in the system input scheduling policy information and inputs it to the system ).

FIG. 9 is a diagram illustrating a configuration of an intelligent power control system 915 according to another embodiment of the present invention.

9 shows a server 955 system including N (N> = 1) servers 955 having one or more service functions classified into M (M> = 1) categories, wherein M (1 < = m < = M) service functions among the N servers 955 of the N servers 955 decreases, The power of n (1 &lt; = n &lt; = N) servers 955 out of N '(N' <= N) servers 955 is controlled to be excluded from the server 955 system, M '(1 <= m' <= M) among the M service functions, when n servers 955 out of the N servers 955 are excluded from the server 955 system according to function- When the traffic corresponding to the number of service functions increases, the traffic for each service function among the n servers 955 excluded from the server 955 system (N "<= n) number of servers 955 included in n '(n' <= n) servers 955 having service functions increased beyond the set reference traffic, 910, and the like, to those skilled in the art to which the present invention pertains, reference is made to and / or modified to illustrate the various aspects of the intelligent power control system 915 configuration It will be appreciated that the method of implementation (e.g., some of the components may be omitted, or broken down, or combined) may be inferred, but the present invention encompasses all such contemplated embodiments, The technical characteristics thereof are not limited. Hereinafter, although the roles of the plurality of means constituting the intelligent power supply control system 915 in the embodiment shown in FIG. 9 are different, the means having similar functions will be described by the same means, It is clear that it is not limited.

According to the present invention, the intelligent power control system 915 is provided on a server 955 system composed of one or more network connection devices 905 and N servers 955, The server 905 connects the communication between the plurality of terminals 900 and the N servers 955 and transmits traffic for detecting traffic between the plurality of terminals 900 and the N servers 955 corresponding to the communication connection, And a detection means (910).

The network connection device 905 is connected to a plurality of terminals 900 requesting communication connection to the N servers 955 through a communication network on the server 955 system and a communication channel between the N servers 955 (LVS), or a Linux Virtual Server (LVS), which connects communication channels between the plurality of terminals 900 and the N servers 955.

The traffic detection unit 910 included in the network connection device 905 may be configured to detect a traffic connection between the plurality of terminals 900 and the N servers 955 when the network connection device 905 connects / Wherein traffic detected by the traffic detection unit (910) is detected by the plurality of terminals (900), and the traffic detected by the traffic detection unit (910) (For example, IP address to IP address traffic, or IP address to IP address for a specific communication port) for a packet exchanged between the specific terminal 900 and the specific server 955 among the N servers 955 Address-to-address traffic) amount.

N servers 955 constituting the server 955 system include a server 955 providing a service function based on a plurality of terminals 900 requesting communication connection through the network connection device 905, , And each server 955 is provided with an application corresponding to one or more service functions.

Here, the service function may be a web service function, a mail service function, a messenger service function, a mini homepage service function, a blog service function, a cafe service function, a knowledge exchange service function, a search service function, an SMS dispatch service function, (Including an internal service function including one or more of a user service function including one or more functions, a development service function, a beta test service function, and an intranet service function) .

Accordingly, if the N servers 955 are classified according to each service function, the web server 955, the mail server 955, the instant messenger server 955, the mini homepage server 955, the blog server 955, Such as knowledge exchange server 955, search server 955, SMS sending server 955, discussion server 955, development server 955, beta test server 955, intranet server 955 Can be classified.

According to the present invention, the physical server 955 constituting the N servers 955 is characterized by having only one service function and one or more servers 955.

For example, the server 1 of the N servers 955 may have a web service function, a mini homepage service function, a cafe service function, a knowledge exchange service function, and a search service function, A service function, a knowledge exchange service function, and a search service function. The server 3 may have a mail service function. The server 4 may include a messenger service function, a web service function, a mail service function, A service function can be provided.

For example, in the case of the server 4, the main service function is a messenger service function related to a chat, a note, and a file exchange using a messenger. However, when providing the web page to the terminal 900 through the messenger service, And further includes the mail service function when the messenger is interlocked with the web mail account, and may further include the SMS dispatch service function when the SMS is sent through the messenger.

If the server 3 is developing a new mail service, the server 3 may have a mail service function, a development service function, and a beta test service function, and the server 4 may also include a messenger service function, a development service function, Function can be provided.

According to the present invention, the N servers 955 are connected to a plurality of terminals 900 through a network connection device 905, Information is generated and stored in the storage means. The user log information includes log generation time information and log data.

According to the present invention, the N servers 955 may be configured to have a maximum resource capacity for a CPU resource and a memory resource according to a hardware component, a process / thread usage amount according to a software component, A maximum resource capacity for a session allocation amount is set. In a process of providing a communication connection of a plurality of terminals 900 through the network connection device 905 and one or more service functions, A function of real-time detection of a resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount to be provided.

The intelligent power control system 915 for controlling the power of the server 955 system constituted by the N servers 955 according to the traffic for each service function of each server 955, 955), category information corresponding to M (M > = 1) service function categories classified according to a predetermined service function classification standard, and at least one service function And a storage medium 945 for storing information of N servers 955 on the basis of respective service functions. The storage medium 945 is connected to the M One or more reference traffic information corresponding to each of the service function categories and n number of reference traffic to be excluded from the server 955 system when the reference traffic is equal to or less than a preset reference value, (1 <= n <= N) servers 955 among the N servers 955 and the server 955 in the server 955 system exclude the reference traffic from the server 955, It is preferable that n '(n' <= n) pieces of system input target server 955 information to be input from the server 955 system are stored in association with each category information. In FIG. 9, the storage medium 945 is shown as a single medium for convenience, but the present invention is not limited thereto.

The category information corresponding to the M (M > = 1) service function categories is information obtained by classifying the respective services provided by the N servers 955 constituting the server 955 according to functions, One service function category is preferably matched with any one service function provided by the N servers 955. [

According to an embodiment of the present invention, the category information may include a classification code or an identifier for a service function provided by the N servers 955.

The N servers 955 information includes identification information for identifying a physical server 955 constituting the server 955 system and at least one service corresponding to one or more service functions provided in the server 955 And function information.

The identification information of the server 955 includes the name of the server 955, the address of the server 955, the address of the server 955, and the name of the server 955 allocated to the respective servers 955 to identify / identify the N servers 955. [ It is preferable to include an identifier.

The service function information of the server 955 information preferably includes a service function classification code or an identifier for identifying a service function of one director provided in each server 955, It is preferable to match with the information.

The reference traffic information is used to compare traffic of a predetermined period with respect to the N servers 955 confirmed through the traffic detection means 910 to user log information generated / accumulated by the N servers 955 / Q (q> 2) or more traffic segments including the lowest traffic and the highest traffic (for example, the traffic for which the predetermined weight is applied to the highest traffic) and the lowest traffic for each service function The traffic 955 may be used as a reference value for excluding n servers 955 out of the N servers 955 constituting the server 955 system from the server 955 system, N "servers 955 included in n '' servers 955 having service functions among the n servers 955 excluded from the server 955 are used as reference values to be inputted into the server 955 system.

The n pieces of system exclusion target server 955 information includes resource usage information for each service function corresponding to the reduced traffic in the traffic decreasing interval and a service function corresponding to the reduced traffic among the N servers 955 N servers 955 (hereinafter, referred to as &quot; servers &quot;) 955 calculated on the basis of the resource usage for the N 'servers 955 and the resource availability (e.g., the amount of resource usage minus the maximum resource capacity of N' ) Information.

The n pieces of system exclusion target server 955 information includes resource usage information for each service function corresponding to the increased traffic in the traffic increase period and service function corresponding to the increased traffic among the N servers 955 N servers 955 (hereinafter, referred to as &quot; servers &quot;) 955 calculated on the basis of the resource usage for the N 'servers 955 and the resource availability (e.g., the amount of resource usage minus the maximum resource capacity of N' ) Information.

According to an embodiment of the present invention, the intelligent power control system 915 may include a system that includes traffic reduction period information for one or more service function traffic, traffic information of the traffic reduction period, and n system excluding target servers 955 One or more scheduling policy information among the scheduling policy information for exclusion, the system increase scheduling policy information including the traffic increase interval information for one or more service function-specific traffic, the traffic information of the traffic increase interval, and n & And a scheduling D / B 950 for storing information, or interworking through a communication network.

The traffic reduction period information may include traffic for a certain period of time for the N servers 955 identified through the traffic detection unit 910 for a certain period of time, A traffic pattern for each service function calculated by comparing / analyzing the log information is statistically processed based on at least one of a predetermined time zone, day, day, week, and month, Day, week, month, and month that is less than a preset reference value.

According to the embodiment of the present invention, the traffic reduction period information does not directly specify a specific period such as the time zone, day, day of week, week, month, The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

For example, if there is a pattern in which the traffic corresponding to the mail service function drops below a predetermined reference value on the day of rain, the traffic reduction section information can not directly specify the rainfall date, And setting the traffic reduction period to the traffic reduction period.

Preferably, the traffic information of the traffic reduction section includes a traffic amount and a traffic increase / decrease ratio which are maintained within a predetermined range during the traffic decrease period, which is decreased to less than the set reference value in the traffic decrease section.

The n number of system exclusion target servers 955 in the traffic reduction period include information on resource usage by service function corresponding to the reduced traffic in the traffic reduction period and information on the amount of traffic corresponding to the reduced traffic among the N servers 955 (For example, an amount obtained by subtracting the resource use amount from the maximum resource capacity of N 'servers 955) for N' servers 955 having service functions to service It is preferable to include information on the number of servers 955.

The traffic increase period information may include traffic for a certain period of time for the N servers 955 that are confirmed through the traffic detection unit 910 for a certain period of time by the N servers 955, A traffic pattern for each service function calculated by comparing / analyzing the log information is statistically processed based on at least one of a predetermined time zone, day, day, week, and month, And includes section information corresponding to at least one of a time zone, a day, a day of the week, a week, and a month, which is equal to or greater than a preset reference value.

According to the embodiment of the present invention, the traffic increase section information does not directly specify a specific section such as the time zone, day, day of week, week, and month, and increases the traffic pattern by service function The weather information, the temperature information, the humidity information, and the event information in which the pattern is generated.

For example, if there is a pattern in which a traffic corresponding to the discussion service function increases to a predetermined reference value or more on the day of a certain group gathering, the traffic increase section information can not directly designate the day of the meeting And information for setting the meeting date and time as the traffic increase section if the group meeting is predicted.

The traffic information of the traffic increase section may include a traffic amount and a traffic increase / decrease ratio which are increased within a predetermined range during the traffic increase period by increasing to a value exceeding the set reference value in the traffic increase section.

The n pieces of system exclusion target server 955 information of the traffic increase period include information on resource usage amount by service function corresponding to the traffic increased in the traffic increase period and information on the increase of traffic among the N servers 955 (For example, an amount obtained by subtracting the resource use amount from the maximum resource capacity of N 'servers 955) for N' servers 955 having service functions to service It is preferable to include information on the number of servers 955.

9, an intelligent power control system 915 that excludes n servers 955 out of the N servers 955 from the server 955 system includes N servers 955 constituting the server 955 system, M service function category information classified into M (M > = 1) service function categories and N servers 955 information for each service function and stores the at least one service function in the storage medium 945 And one or more reference traffic information calculated correspondingly to the M service function categories and n system excluding target servers 955 to be excluded from the server 955 when the reference traffic is equal to or less than a predetermined reference value (Hereinafter referred to as &quot; traffic information &quot;) stored in the storage medium 945 in association with each category information, The traffic for each of the M service function functions is classified and analyzed according to the M service function categories of m (1 and &lt; = m &lt; = M) traffic function analyzing means 940 for checking traffic for each service function.

The information registering unit 935 receives service function classification criteria for one or more service functions provided in the N servers 955 through the administrator terminal 900 managing the intelligent power control system 915 (M> = 1) service function categories for each service function by classifying a plurality of service functions provided in the N servers 955 according to the service function classification standard, (M > = 1) configured by classifying a plurality of service functions provided in the N servers 955 according to a predetermined service function classification criterion, at least one service function provided in the N servers 955, N service 955 information having at least one service function, and M service function category information configured at each service function, The interconnect is characterized by storing in the storage medium 945.

The traffic analyzing unit 940 periodically checks the traffic for a predetermined unit time for the N servers 955 through the traffic detecting unit 910 and generates the traffic by the N servers 955 And comparing / analyzing accumulated user log information with the identified traffic for each unit time, and verifying M traffic for each service function during the unit time.

In case that the n servers 955 are checked in real time for n servers 955 having a service function in which the traffic for each service function is less than a predetermined reference value and excluded from the server 955, Means 940 is characterized in that m (1 < = m < = M) traffic for each service function whose traffic is less than or equal to a predetermined reference value among the traffic for each of the M service functions identified.

The traffic analyzing unit 940 analyzes the traffic information of the server 910 through the traffic detecting unit 910 to store (or periodically update) the reference traffic information and the n system excluding target servers 955 information in the storage medium 945. [ N pieces of traffic for a certain period of time for the server 955 for each predetermined analysis unit time and transmits user log information generated / accumulated by the N servers 955 to the identified traffic And analyzing M traffic for each service function for a predetermined period.

If the traffic for each M service functions for the predetermined period is analyzed, the information registering unit 935 registers the N servers 955 based on the highest traffic and the lowest traffic among the M service function traffic for the predetermined period, N servers 955 among the n servers 955 having the service function among the n servers 955 excluded from the server 955 system or reference values excluding the n servers 955 from the server 955 Quot; n "servers 955 to the server 955 system.

When the reference value is set, the information registering unit 935 applies a predetermined weight to the maximum traffic, sets q (q (q)) as the traffic interval reference point by using the minimum traffic, the set reference value, > 2) traffic sections are set, and reference points of each traffic section are calculated as reference traffic.

Here, the weight applied to the maximum traffic may be a power control according to the service function-specific traffic even if the traffic per service function exceeds the maximum traffic according to the amount of traffic change (or increase / decrease ratio) And it may be omitted according to the intention of the person skilled in the art.

Wherein the traffic interval is obtained by dividing the interval between the maximum traffic and the maximum traffic by q (q> 2) or more with the lowest traffic as the lowest reference point and the highest traffic with the weight applied as the highest reference point, The reference point of the section for dividing the traffic section may include the set reference value. That is, the traffic interval does not necessarily have to be equal intervals, and the lowest traffic, the reference value, and the highest traffic are periodically updated according to traffic for each service function for a predetermined period analyzed through the traffic analyzing unit 940.

When the reference traffic having a preset reference value or less is calculated, the information registering unit 935 registers the n services with the service quality for providing the service function to the plurality of terminals 900 within the predetermined time limit, And the exclusion target server 955 information.

The information registering unit 935 stores the calculated reference traffic information and the n system excluding target servers 955 information in association with the category information corresponding to the service function in the storage medium 945, The reference traffic and the information of the n system exclusion target servers 955 are periodically updated according to the traffic analysis of each service function of the means 940.

Alternatively, if n (1 <= n <= N) number of servers 955 among the N servers 955 are calculated in the server 955 system in the excluded state, The means 935 is a means for providing n (1 < = n < = N) number of servers 955 among the N servers 955 having a service function corresponding to the reference traffic in the excluded state in the server 955 system N 'number of system input target servers 955 that maintain the service quality for providing the service functions within a predetermined time limit from the one or more servers 955 to the plurality of terminals 900. [

The information registering unit 935 stores the calculated reference traffic information and n 'number of system entry target servers 955 in association with the category information corresponding to the service function in the storage medium 945, The reference traffic and the n 'number of system input target server 955 information are periodically updated according to the traffic analysis of each service function of the means 940.

9, an intelligent power supply control system 915 that excludes n servers 955 among the N servers 955 from the server 955 system receives the identified m (n) from the storage medium 945, A server 955 confirmation means 925 for confirming n (1 <= n <= N) system exclusion target servers 955 associated with the reference traffic matched with the traffic for each service function, And a power control unit 920 for controlling the power of the server 955 to exclude it from the server 955 system.

In case that the n servers 955 are checked in real time for n servers 955 having a service function in which the traffic for each service function is less than a predetermined reference value and excluded from the server 955, When the m-number of service function-specific traffic with the service function-specific traffic equal to or less than the preset reference value is confirmed by the means 940, the server 955 confirmation means 925 confirms the m (1 < = n < = N) number of system exclusion target servers 955 associated with the reference traffic matched with the function-specific traffic.

When the n system elimination target servers 955 are confirmed by the server 955 confirmation means 925, the power source control means 920 controls the power of the n system elimination target servers 955 And is excluded from the server 955 system.

According to an embodiment of the present invention, the power source control means 920 controls the power of the n system elimination target servers 955 and outputs the determined n And checks whether the server 955 normally responds to the ping.

If the n servers 955 respond to the ping, the power control unit 920 receives the ping request from the plurality of terminals 900 through the network connection device 905 to the n servers 955, deactivates the in-service request and verifies that the in-service deactivation is operating normally.

If the in-service disabling is confirmed, the power control unit 920 terminates the service function provided in the n servers 955, checks whether the service function is normally terminated, The power of the n servers 955 is shut off, and then the n servers 955 check whether they are not responsive to the ping.

9, an intelligent power control system 915 that excludes n servers 955 out of the N servers 955 from the server 955 system includes a scheduler Week, month, and month of the service function for each of the M service function functions during a time period, a day, a week, a week, and a month And a scheduling policy for associating and storing n number of system exclusion target servers 955 with respect to the traffic of each of the traffic reduction periods and each of the traffic reduction periods, to the scheduling D / B 950 The scheduling policy processing unit 930 may further include a scheduling policy setting unit 930 for setting a scheduling policy to the scheduling D / Traffic information for each service function identified through the traffic analyzing unit 940 is transmitted to the scheduling D / B unit 950 (step &lt; RTI ID = 0.0 &gt; The control unit 940 controls the power of the n system elimination target servers 955 stored in the scheduling D / B 950 to be excluded from the server 955 system, The n number of system exclusion target servers 955 that are confirmed in real time by the server 955 checking means 925 are notified if the traffic for each service function identified through the checking step 925 does not match the traffic information stored in the scheduling D / So that the server 955 system is excluded.

When the traffic for each M service functions for a certain period identified through the traffic analyzing unit 940 is confirmed, the scheduling policy processing unit 930 determines whether the N servers 955 ), And compares / analyzes the user log information generated / accumulated by the N servers 955 with the identified traffic for each analysis unit time to determine a schedule The traffic pattern for each M service function for a certain period of time is statistically processed based on at least one of time zone, day, day of week, week, and month, A time zone, a day, and a time when the traffic for each function is reduced to a preset reference value or the traffic for each service function is increased to a predetermined reference value or more, Check at least one day, week, or month.

According to the embodiment of the present invention, the scheduling policy processing unit 930 checks the traffic for a predetermined period of time for the N servers 955 for each predetermined analysis unit time through the traffic detection unit 910 , User log information generated / accumulated by the N servers 955 is compared / analyzed with at least one of external condition information such as weather information, temperature information, humidity information, and event information for each analysis unit time, and M The traffic for each service function is statistically processed on the basis of at least one of weather, temperature, humidity, and event, and traffic patterns for each of the M service functions for the statistically processed period are analyzed, Temperature, humidity, or event in which the traffic per service function increases above a predetermined reference value Check the above.

According to the method of the present invention, the external condition information may further include an external condition for determining the traffic for each service function in addition to the weather information, the temperature information, the humidity information, and the event information, The present invention is not limited thereto.

If at least one of a time zone, a day, a week, a week, and a month in which the traffic for each service function decreases to a predetermined reference value or less is detected, the scheduling policy processing unit 930 sets the traffic for each of M service functions The traffic reduction period information including at least one of a time zone, a day, a week, a week, and a month that is less than a reference value and the traffic of the traffic reduction period are checked, The system exclusion scheduling policy information including the traffic reduction period information, traffic information of the traffic reduction period, and n system excluding target servers 955 is configured And stores it in the scheduling D / B 950.

If at least one of weather, temperature, humidity, and event that the traffic for each service function decreases to a predetermined reference value or more is detected, the scheduling policy processing unit 930 updates the M- A traffic decrease section that dynamically sets a traffic decrease section including at least one of weather, temperature, humidity, and events that are less than or equal to a predetermined value, and traffic of the traffic decrease section, The system exclusion scheduling unit 955 checks n system exclusion target servers 955 for each traffic reduction interval, and performs system exclusion scheduling including the traffic decrease interval information, traffic information of the traffic decrease interval, and n system exclusion target servers 955 And stores the policy information in the scheduling D / B 950.

If the system exclusion scheduling policy information that excludes n system exclusion target servers 955 for a predetermined traffic decrease period among the N servers 955 is stored in the scheduling D / B 950, the scheduling policy The processing means 930 checks whether the traffic reduction period of the system exclusion scheduling policy information has been reached through the scheduling D / B 950, and when the traffic reduction period arrives, Function-based traffic matches the traffic information of the system exclusion scheduling policy information.

If the traffic for each service function is matched with the traffic information of the system exclusion scheduling policy information, the scheduling policy processing unit 930 notifies the power of the n system exclusion target servers 955 included in the system exclusion scheduling policy information If the traffic for each service function is not matched with the traffic information of the system exclusion scheduling policy information, the server 955 confirm means 925 confirms that n system exclusion target servers (955) and controls the power of the n system elimination target servers (955) to be excluded from the server (955) system.

Referring to FIG. 9, when n (1 <= n <= N) servers 955 out of the N servers 955 are excluded from the server 955 system, n The intelligent power supply control system 915 for putting a plurality of servers 955 in the server 955 system is provided with (Nn) servers 955 (Nn) which are confirmed through the traffic detection means 910 provided in the network connection device 905 ) Traffic analyzing means for classifying / analyzing the traffic for each of the M service function categories to identify m '(1 <= m' <= M) traffic for each service function whose traffic for each service function is equal to or greater than a preset reference value 940).

When n (1 <= n <= N) servers 955 out of the N servers 955 are excluded from the server 955 system, the traffic analyzing means 940 determines whether the traffic detection means 910 (Nn) servers 955, and the user log information generated / accumulated by the (Nn) servers 955 is checked for the unit time And comparing the analyzed traffic with the traffic for each of the M service functions for the unit time.

If the traffic for each of M service functions for the unit time is identified, the traffic analyzing unit 940 calculates m '(1 < = m '<= M) traffic for each service function.

According to an embodiment of the present invention, the m 'service-specific traffic includes a boot time at which n "number of servers 955 having m' service functions are booted and a boot time at which the booted server 955 The service quality of the service-specific traffic is confirmed to be maintained in the waiting-for-waiting time in consideration of an input waiting time including an application load time at which the service functions are loaded, So that the quality of service of the traffic for each service function is maintained in the waiting time for the service.

Referring to FIG. 9, an intelligent power control system 915 which inputs n "number of servers 955 out of the n servers 955 in the server 955 system is connected to the n ' Confirmation of the server 955 for confirming n "(n" <= n ') system input destination servers 955 associated with the identified reference traffic matched with the identified m' service function traffic among the 95 servers 955 And a power control means 920 for controlling the power of the n "number of system exclusion target servers 955 and putting the power into the server 955 system.

When the traffic for each m 'service function is confirmed through the traffic analyzing unit 940, the server 955 checking unit 925 confirms the traffic of each of the n' servers 955 through the storage medium 945 (N "&lt; = n)) system input destination servers 955 associated with the identified reference traffic matched with the m 'service function-specific traffic.

When the n "number of system entry target servers 955 are confirmed by the server 955 confirmation means 925, the power source control means 920 checks the power of the n" number of system exclusion target servers 955 And inputs it to the server 955 system.

According to an embodiment of the present invention, the power source control means 920 controls the power of the n "system input target servers 955 and inputs them to the n" (955) does not respond to the ping.

If the n "number of servers 955 do not respond to the ping, the power control unit 920 periodically checks whether the power of the n" servers 955 is in response to the ping, "Servers 955 responds to the ping, the service functions provided in the n servers 955 are executed in a predetermined order, and it is confirmed whether the service function is normally executed.

If the service function is executed, the power control unit 920 receives an in-service request from the plurality of terminals 900 to the n servers 955 through the network connection unit 905, And confirms that the in-service activation is normally operating.

Referring to FIG. 9, an intelligent power control system 915 that inputs n "number of servers 955 out of the n servers 955 in the server 955 system checks Day, week, month, and the like, based on at least one of the M, M, and C service functions for a predetermined period of time, , And a scheduling policy processing means 930 for checking and connecting to the scheduling D / B 950 the system input target server 955 of the traffic increase period of the traffic increase period, The scheduling policy processing means 930 may include the traffic increase period information, the traffic information of the traffic increase period, and the n "number of systems &lt; RTI ID = 0.0 &gt; After the information on the destination server 955 is stored in association with the traffic information, traffic for each service function identified through the traffic analysis unit 940 is matched with the traffic information stored in the scheduling D / B 950, The control unit 910 controls the power of the n "number of system exclusion target servers 955 stored in the scheduling D / B 950 to put them into the server 955 system, The server 955 controls the power of the n "number of system exclusion target servers 955 that are confirmed in real time by the server 955 checking means 925, if the corresponding traffic does not match the traffic information stored in the scheduling D / To the server 955 system.

Alternatively, if at least one of a time zone, a day, a week, a week, and a month in which the traffic for each service function increases to a predetermined reference value or more is detected, the scheduling policy processing unit 930 transmits the M service- The server 955 checks the traffic increase section information including at least one of time, day, day, week, and month that is equal to or greater than a preset reference value and the traffic of the traffic increase section, System loading target server 955 for the traffic increase period, and performs system input scheduling including the traffic increase interval information, traffic information of the traffic increase interval, and n "system input target server 955 information And stores the policy information in the scheduling D / B 950.

If at least one of weather, temperature, humidity, and event that the traffic for each service function increases to a predetermined reference value or more is detected, the scheduling policy processing unit 930 determines whether the M- The traffic increase section periodically sets the traffic increase section including the at least one of weather, temperature, humidity, and event more than the predetermined value, and traffic of the traffic increase section, and transmits the traffic increase section information through the server 955 confirmation means 925 (N ") system entrance target server 955 for each traffic increase section, and includes traffic increase section information, traffic information of the traffic increase section, and n & And sets up the input scheduling policy information and stores it in the scheduling D / B 950.

If the system exclusion scheduling policy information that excludes n system exclusion target servers 955 for a predetermined traffic increase period from the N servers 955 is stored in the scheduling D / B 950, the scheduling policy The processing means 930 checks whether the traffic increase period of the system exclusion scheduling policy information is reached through the scheduling D / B 950, and when the traffic increase period is reached, Is matched with the traffic information of the system exclusion scheduling policy information.

If the traffic for each service function matches the traffic information of the system exclusion scheduling policy information, the scheduling policy processing unit 930 notifies the power of the n "system exclusion target servers 955 included in the system exclusion scheduling policy information The control unit 940 controls the traffic analyzing unit 940 so that n "system input targets (i.e., n " system input targets) are not included in the system exclusion scheduling policy information, Checks the server 955, and controls the power of the n "number of system input target servers 955 to be processed in the server 955 system.

FIG. 10 is a diagram illustrating a traffic analysis process for each service function according to another embodiment of the present invention.

In more detail, FIG. 10 is a flowchart illustrating an operation of analyzing traffic for a predetermined period of time for the N servers 955 in the intelligent power control system 915 shown in FIG. 1 for a predetermined period of time, The number of servers 955 to the server 955 is set to n "servers 955 out of n servers 955 excluded from the system and a reference value for excluding n servers 955 from the system 955, (N) servers to be excluded from the server 955 in each reference traffic, and n servers (N) of the N servers 955, (955) confirms n 'number of system input target server (955) information to be input to the server (955) system in each reference traffic in the exclusion state in the server (955) system and stores the information in the storage medium (945) As shown in FIG. Those skilled in the art will be able to refer to and / or modify the FIG. 10 to understand various methods of analyzing the traffic for each service function (for example, some steps may be omitted, However, the present invention includes all of the above-mentioned embodiments, and the technical features of the present invention are not limited by the method shown in FIG.

Referring to FIG. 10, the intelligent power control system 915 classifies a plurality of service functions provided in the N servers 955 according to a predetermined service function classification criterion by information setting of an administrator terminal 900 M (M > = 1) service function categories for each service function, and information of N servers 955 having one or more service functions and M service function category information, (1000), and the traffic identified by the traffic detection unit 910 of the network connection device 905 is analyzed on a unit time basis to confirm traffic for each of the M service functions (1005).

While the traffic for each of the M service functions is confirmed, the intelligent power control system 915 analyzes traffic for each of the M service functions for a predetermined period to calculate n "number of servers out of n servers 955 excluded from the system (Q > 2) or more traffic segments including a reference value for inputting the reference traffic value 955 to the system of the server 955 and a minimum traffic value and a maximum traffic value as reference points 1010, N system excluding target servers 955 to be excluded from the server 955 system and n servers 955 among the N servers 955 are not included in each reference traffic in the excluded state in the server 955 system (N) 'system entry target server 955 information to be input to the server 955 (1015).

If the reference traffic, the n system exclusion target server 955 information, and the n 'system insertion target server 955 information are calculated 1020, the intelligent power control system 915 compares the reference traffic information with n (Or periodically updates) the information on the system exclusion target server 955 and the information on n 'number of system input target servers 955 in the storage medium 945 in association with the category information corresponding to the service function (1025).

FIG. 11 is a diagram illustrating a process of excluding one or more servers 955 in a server 955 system according to service function-based traffic that is confirmed in real time according to another embodiment of the present invention.

FIG. 11 is a flowchart illustrating a method of reducing traffic corresponding to m (1 <= m <= M) service functions among the M service functions according to traffic analysis of M service functions according to the process shown in FIG. (1 <= n <n) among N '(N' <= N) servers 955 having service functions in which the traffic for each service function among the N servers 955 is reduced to a predetermined reference traffic or less, = N) servers 955 and excludes them from the server 955 system. Those skilled in the art can refer to FIG. 11 and / or FIG. Or variations may be made to various implementations of the server 955 exclusion process (e.g., omissions or alterations of some steps), but the present invention includes all such contemplated implementations In FIG. 11, The technical features thereof are not limited only by the method of implementation.

Referring to FIG. 11, when the traffic for each of the M service functions is analyzed according to the process shown in FIG. 10, the intelligent power control system 915 analyzes M service function traffic for a predetermined unit time, The m number of service function-based traffic whose traffic is less than or equal to the set reference value is checked 1100 and the n number of system exclusion target servers 955 connected to the reference traffic corresponding to the m number of service function traffic from the storage medium 945 (1105).

If the n system elimination target servers 955 are confirmed 1110, the intelligent power supply control system 915 controls power of the n system elimination target servers 955, (1115).

FIG. 12 is a diagram illustrating a process of inputting one or more servers 955 excluded from the server 955 system according to service function-based traffic that is confirmed in real time according to another embodiment of the present invention.

12, n servers 955 among the N servers 955 are classified according to the m traffic function-based traffic analysis according to the M service function analyzes according to the process shown in FIG. 10 When excluded from the server 955 system, if traffic corresponding to m '(1 <= m' <= M) service functions among the M service functions increases, n (N "<= n) included in n '(n' <= n) servers 955 having service functions whose service function- ) Of the server 955 and controlling the power of the server 955 and inputting the power of the server 955 to the server 955. Those skilled in the art can refer to FIG. 12 and / Various methods for the process of inputting the server 955 (e.g., It is to be understood that the invention may be practiced otherwise than as specifically described herein, but it is to be understood that the invention is not to be limited to the specific embodiments thereof, Or more.

Referring to FIG. 12, when traffic for each of the M service functions is analyzed according to the process shown in FIG. 10, the intelligent power control system 915 analyzes M service function traffic for a predetermined unit time, Checks m 'pieces of traffic for each service function in which the traffic increases to a predetermined reference value or more (1200), checks n' servers 955 having m 'service functions from the storage medium 945 (1205 ), N "(n" <= n ') system input destination servers 955 associated with the reference traffic matched with the identified m' service function traffic among the n 'number of servers 955 (1210).

If the n "number of system input target servers 955 are confirmed (1215), the intelligent power supply control system 915 controls power of the n" number of system input destination servers 955, (1220).

According to the present invention, the power of a server system constituted by N (N> = 1) servers having one or more service functions classified into M (M> = 1) service function categories is divided into M The server system is advantageous in that the power consumption of servers constituting the server system is controlled and efficiently reduced according to the amount of traffic for each service function.

Claims (25)

A method of controlling power of a server system including N (N > = 1) servers according to traffic for each service function of each server, (1) one or more service functions provided in each server are classified into M (M > = 1) service function categories, and the M service function category information and N server information are classified by each service function Storing in a medium; (2) calculating at least one reference traffic information corresponding to each of the M service function categories, and for each reference traffic, resource usage information for each reference traffic for maintaining the service quality of the service function at a predetermined reference value or more Storing (or updating) in a storage medium in association with each category information; (3) Classifying / analyzing the traffic for the N servers identified through the traffic detection means provided in the network connection device according to the M service function categories, and if traffic for each service function is m (1 < m < = M) traffic for each service function; (4) identifying (or calculating) m resource usage amount information for each m service functions corresponding to the m service function traffic through the storage medium, and calculating N "(N" = N) servers and resource availability for each server; (5) transferring a resource usage amount corresponding to the service function among the N "number of servers to a resource availability amount of another server through the resource usage amount and the resource availability amount for each server identified as above (1 < = n & ") &Lt; / RTI &gt; And (6) controlling the power of the n system elimination target servers to be excluded in the server system. The method according to claim 1, (7) If n (1 < = n < = N) servers out of the N servers are excluded from the server system, traffic for the (Nn) (1 &lt; = m < = M) traffic for each service function having traffic according to the service function equal to or greater than a preset reference value; (M "&lt; m &gt; &lt; m &gt; &lt; m & = m ') of resource usage information for each service function; (9) identifying n "(n" &lt; = n ') system input destination servers having the maximum resource capacity of at least m " (10) controlling the power of the n "number of system exclusion target servers, and inputting the control result into the server system. A method of controlling power of a server system including N (N > = 1) servers according to traffic for each service function of each server, (1) one or more service functions provided in each server are classified into M (M > = 1) service function categories, and the M service function category information and N server information are classified by each service function Storing in a medium; (2 ") one or more reference traffic information corresponding to each of the M service function categories, and when the reference traffic is equal to or less than a preset reference value, n system elimination target server information to be excluded from the server system, Storing information in the storage medium in association with information; (3) Classifying / analyzing the traffic for the N servers identified through the traffic detection means provided in the network connection device according to the M service function categories, and if traffic for each service function is m (1 < m < = M) traffic for each service function; (4-5) identifying n (1 < = n < = N) system exclusion target servers associated with the identified reference traffic matched with the identified m service function traffic from the storage medium; (6) controlling the power of the n system elimination target servers to be excluded in the server system. The method of claim 3, (1 " = n &lt; = N) servers are excluded from the server system by the step (2 " Storing n '(n' <= n) system input target server information to be input to the server system in association with each category information in the storage medium; (3) when n servers out of the N servers are excluded from the server system, (7) when traffic for the (Nn) servers identified through the traffic detection means included in the network connection apparatus is M Identifying m '(1 <= m' <= M) traffic for each service function whose traffic according to service function is greater than or equal to a predetermined reference value by classifying / analyzing according to service function categories; (8-9) n "(n" &lt; = n ') system input destination servers associated with the reference traffic matched with the identified m' service function traffic among the n ' Checking; And (10) controlling power of the n "system input target server and inputting the power to the server system. A method of controlling power of a server system including N (N > = 1) servers according to traffic for each service function of each server, (1) one or more service functions provided in each server are classified into M (M > = 1) service function categories, and the M service function category information and N server information are classified by each service function Storing in a medium; (2) calculating at least one reference traffic information corresponding to each of the M service function categories, and for each reference traffic, resource usage information for each reference traffic for maintaining the service quality of the service function at a predetermined reference value or more Storing (or updating) in a storage medium in association with each category information; (3) when n (1 <= n <= N) servers out of the N servers are excluded from the server system, (7) (1 &lt; = m '&lt; = M) traffic for each service function having a predetermined reference value or more for each service function traffic by classifying / analyzing traffic for each of the M service function categories; (M "&lt; m &gt; &lt; m &gt; &lt; m & = m ') of resource usage information for each service function; (9) identifying n "(n" &lt; = n ') system input destination servers having the maximum resource capacity of at least m " (10) controlling the power of the n "number of system exclusion target servers, and inputting the control result into the server system. A method of controlling power of a server system including N (N > = 1) servers according to traffic for each service function of each server, (1) one or more service functions provided in each server are classified into M (M > = 1) service function categories, and the M service function category information and N server information are classified by each service function Storing in a medium; (1 &lt; = n < = N) servers out of the N servers are excluded from the server system, Storing n '(n' <= n) system input target server information to be input to the server system when the reference traffic is equal to or greater than a predetermined reference value, in association with each category information, in the storage medium; (3) when n servers out of the N servers are excluded from the server system, (7) when traffic for the (Nn) servers identified through the traffic detection means included in the network connection apparatus is M Identifying m '(1 <= m' <= M) traffic for each service function whose traffic according to service function is greater than or equal to a predetermined reference value by classifying / analyzing according to service function categories; (8-9) n "(n" &lt; = n ') system input destination servers associated with the reference traffic matched with the identified m' service function traffic among the n ' Checking; And (10) controlling power of the n "system input target server and inputting the power to the server system. The method of claim 1 or 3, or 5 or 6, wherein the step (1) (M > = 1) service function categories for each service function by classifying a plurality of service functions provided in the N servers according to a predetermined service function classification criterion, Wherein the server information and the M service function category information are interconnected based on respective service functions and stored in a storage medium. The method according to claim 1 or 5, wherein the step (2) Wherein the traffic monitoring unit checks traffic for a predetermined period of time for each of the N servers by a predetermined analysis unit time, and transmits user log information generated / accumulated by the N servers to the identified traffic Analyzing M traffic for each service function for a predetermined period; The method according to any one of claims 1 to 3, wherein the server is further configured to determine whether to exclude n servers out of the N servers based on the highest traffic and the lowest traffic among the M service- Setting a reference value for putting a number of servers into the server system; (Q > 2) or more traffic segments are set with the lowest traffic, the set reference value, and the highest traffic applied with the weight as the traffic segment reference points, Calculating a reference point as a reference traffic; Calculating a resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount for providing the service function in each reference traffic to a plurality of terminals within a predetermined time limit; Calculating resource usage information for each reference traffic by applying a predetermined weight to the calculated resource usage; And And storing (or updating) the reference traffic information and the resource usage information for each reference traffic in association with the category information corresponding to the service function in a storage medium. The method according to claim 1 or 3, Wherein the traffic monitoring unit checks traffic for a predetermined period of time for each of the N servers by a predetermined analysis unit time, and transmits user log information generated / accumulated by the N servers to the identified traffic Statistical processing of traffic for each of M service functions for a predetermined period on the basis of one or more of time, day, day, week, and month; The M-number of service function traffic patterns of the service function for the predetermined statistical period are analyzed to determine traffic decrease period information including at least one of a time zone, a day, a day, a week, And checking traffic of the traffic reduction period; Confirming n system elimination target servers for each traffic reduction period through steps (4) to (5) or (4-5); And Establishing system exclusion scheduling policy information including the traffic reduction period information, the traffic information of the traffic reduction period, and n system exclusion target server information, and storing the system exclusion scheduling policy information in the scheduling D / B , When the system exclusion scheduling policy information for excluding n system exclusion target servers for a predetermined traffic decrease period from the N servers is stored in the scheduling D / B, the system exclusion scheduling policy information through the scheduling D / Confirming whether a traffic reduction period has been reached; Wherein traffic for each of the N servers identified through the traffic detection means included in the network connection device is classified and analyzed according to the M service function categories when the traffic reduction period is reached, Checking whether the traffic information matches the traffic information of the scheduling policy information; If the traffic for each service function is matched with the traffic information of the system exclusion scheduling policy information, the control unit controls the power of n system exclusion target servers included in the system exclusion scheduling policy information to exclude them from the server system, If the number of the system excluding scheduling policy information does not match the traffic information of the system exclusion scheduling policy information, the power of n system elimination target servers identified through steps (4) through (5) or (4-5) And removing said power supply voltage from said power supply. The method according to claim 2 or 4 or 5 or 6, Wherein the traffic monitoring unit checks traffic for a predetermined period of time for each of the N servers by a predetermined analysis unit time, and transmits user log information generated / accumulated by the N servers to the identified traffic Statistical processing of traffic for each of M service functions for a predetermined period on the basis of one or more of time, day, day, week, and month; The M traffic service patterns for each service function for a predetermined period of time are analyzed to determine traffic increase interval information including at least one of time zone, day, day, week, and month, And checking traffic of the traffic increase period; Confirming n "number of system input target servers for each traffic increase period through steps (8) to (9) or (8-9) And setting up the system input scheduling policy information including the traffic increase period information, the traffic information of the traffic increase period, and n "system input target server information, and storing the system input scheduling policy information in the scheduling D / B In addition, When the system exclusion scheduling policy information for excluding n system exclusion target servers for a predetermined traffic increase period from the N servers is stored in the scheduling D / B, the system exclusion scheduling policy information through the scheduling D / Checking whether a traffic increase period is reached; (Nn) servers identified through traffic detection means included in the network connection device when the traffic increase period is reached, and traffic per service function, which is identified by analyzing / analyzing traffic for each of the M service function categories, Confirming whether the traffic information matches the traffic information of the exclusion scheduling policy information; When the traffic for each service function is matched with the traffic information of the system exclusion scheduling policy information, the power of the n "system exclusion target servers included in the system exclusion scheduling policy information is controlled and put into the server system, The system control unit controls power of n "number of system input target servers identified through steps (8) through (9) or (8-9) And inputting the generated power to the server system. 4. The method of claim 1 or 3, wherein the step (3) Checking traffic for a predetermined unit time for the N servers through the traffic detection unit, comparing / analyzing user log information generated / accumulated by the N servers with the identified traffic for each unit time And checking M traffic for each service function during the unit time. The method of claim 2, 4, 5, or 6, wherein the step (7) (Nn) servers by using the traffic detection means, and transmits user log information generated / accumulated by the (Nn) servers to the identified traffic Comparing and analyzing the M functions of the plurality of service functions, and checking traffic for each of M service functions during the unit time. 4. The method of claim 1 or 3, wherein the step (6) Checking whether the n servers normally respond to a ping; In response to the ping, when the n servers respond to the ping, the in-service request requested from the plurality of terminals to the n servers is deactivated through the network connection device and the in-service request deactivation is normally performed step; When the in-service request deactivation is confirmed, terminating a service function provided in the n servers and confirming that the service function is normally terminated; Shutting down the power of the n servers when the service function for the n servers is terminated; And And verifying that the n servers are not responsive to the ping. The method of claim 2, 4, 5, or 6, wherein the step (10) Checking whether the "n &quot; servers are not responsive to a ping; If the n "servers do not respond to the ping, periodically checking whether the power of the n" servers is connected and responsive to the ping; Performing the service functions provided in the n servers in a predetermined order when the n "servers respond to the ping, and checking whether the service functions are normally executed; Activating an in-service request from the plurality of terminals to the n servers through the network connection device and checking whether the in-service request activation is normally performed, when the service function is executed; Wherein the power supply control unit includes: A computer-readable recording medium storing a program for executing the method according to any one of claims 1 to 6. 1. A system for controlling power of a server system composed of N (N > = 1) servers according to traffic for each service function of each server, A storage medium for storing N server information and M (M > = 1) service function category information for service functions provided in the N servers, for each service function; Information registration means for storing (or updating) one or more reference traffic information corresponding to the M service function categories and resource usage information for each reference traffic, in association with each category information, in a storage medium; (1 < = m < = m) that is less than or equal to a preset reference value by classifying / analyzing traffic for the N servers identified by the traffic detection means provided in the network connection device, M) pieces of traffic for each service function; (Or estimating) resource usage information for each of m service functions corresponding to the m service function traffic through the storage medium, and allocating N "(N" <= N) Resource analyzing means for identifying a server-specific resource usage amount and a resource availability amount for each server; (1 &lt; = n < = N) &quot; nodes (1 &lt; = n &lt; = N) for transferring the resource usage corresponding to the service function among the N & Server confirmation means for confirming a system exclusion target server; And And power supply control means for controlling power of the n system elimination target servers to be excluded from the server system. 17. The method of claim 16, Wherein the traffic analysis means comprises: (1 &lt; = m < = M) number of traffic that is equal to or greater than a preset reference value by classifying / analyzing traffic for (Nn) And checking traffic for each service function. Wherein the resource analyzing means comprises: (M "<= m ') matched with a service function provided in n' (n '<= n) servers having at least one m' service function among the n servers through the storage medium, (Or calculating) resource usage information for each service function, Wherein the server confirmation means comprises: (N "&lt; = n ") number of system input target servers having the maximum resource capacity equal to or larger than the resource usage amount for each m" service functions among the n ' Wherein the power source control means includes: Wherein the control unit controls the power of the n "number of system exclusion target servers to be input to the server system. 1. A system for controlling power of a server system composed of N (N > = 1) servers according to traffic for each service function of each server, One or more service functions included in each server are classified into M (M > = 1) service function categories, and the M service function category information and N server information are stored for each service function media; Calculating at least one reference traffic information corresponding to each of the M service function categories, and associating n pieces of system exclusion target server information to be excluded in the server system with each category information when the reference traffic is equal to or less than a preset reference value Information storing means for storing the information in the storage medium; (1 < = m < = m) that is less than or equal to a preset reference value by classifying / analyzing traffic for the N servers identified by the traffic detection means provided in the network connection device, M) pieces of traffic for each service function; Server confirmation means for identifying n (1 < = n < = N) system exclusion target servers associated with the identified reference traffic matched with the identified m service function functions from the storage medium; And power supply control means for controlling power of the n system elimination target servers to be excluded from the server system. 19. The method of claim 18, The information registration means, (1 < = n < = N) of N servers among the M service function categories, And n '(n' <= n) system input target server information to be input to the server system when the number of the system input target server information is equal to or greater than the reference value, is stored in the storage medium in association with each category information. Wherein the traffic analysis means comprises: When n servers among the N servers are excluded from the server system, traffic for (Nn) servers identified through traffic detection means included in the network connection device is classified / analyzed according to the M service function categories M '(1 < = m &lt; = M) traffic for each service function whose traffic per service function is equal to or greater than a preset reference value, Wherein the server confirmation means comprises: (N "&lt; = n &quot;) number of system entry target servers associated with the reference traffic matched with the identified m 'service function traffic among the n' number of servers through the storage medium In addition, Wherein the power source control means includes: And controls the power of the n "system input target servers to be input to the server system. 1. A system for controlling power of a server system composed of N (N > = 1) servers according to traffic for each service function of each server, A storage medium for storing N server information and M (M > = 1) service function category information for service functions provided in the N servers, for each service function; Information registration means for storing (or updating) one or more reference traffic information corresponding to the M service function categories and resource usage information for each reference traffic, in association with each category information, in a storage medium; If N (1 < = n < = N) servers out of the N servers are excluded from the server system, traffic for the (Nn) (1 &lt; = m < = M) traffic functions for each service function, the traffic for each service function being classified according to service function categories, (M "<= m ') matched with a service function provided in n' (n '<= n) servers having at least one m' service function among the n servers through the storage medium, Resource analyzing means for identifying (or calculating) resource usage information for each service function; Server confirmation means for confirming n "(n" <= n ') system input destination servers having the maximum resource capacity equal to or larger than the resource usage amount for each m "service functions among the n' And power control means for controlling power of the n "number of system exclusion target servers and inputting the power to the server system. 1. A system for controlling power of a server system composed of N (N > = 1) servers according to traffic for each service function of each server, (M > = 1) service function categories provided in each server, and stores the M service function category information and the N server information classified and linked for each service function, ; (1 < = n < = N) of N servers among the M service function categories, Information registering means for storing n '(n' <= n) system input target server information to be input to the server system when the input value is equal to or greater than a reference value, in association with each category information, in the storage medium; When n servers among the N servers are excluded from the server system, traffic for (Nn) servers identified through traffic detection means included in the network connection device is classified / analyzed according to the M service function categories Traffic analysis means for confirming m '(1 <= m' <= M) traffic for each service function whose traffic per service function is equal to or greater than a preset reference value; (N "&lt; = n &quot;) number of system input destination servers associated with the reference traffic matched with the identified m 'service function traffic among the n' number of servers through the storage medium, ; And And power control means for controlling power of the n "system input target servers and inputting the power to the server system. The information processing apparatus according to claim 16, 18, 20, or 21, (M > = 1) service function categories for each service function by classifying a plurality of service functions provided in the N servers according to a predetermined service function classification criterion, Wherein the server information and the M service function category information are interconnected on the basis of respective service functions and stored in a storage medium. 21. The method according to claim 16 or 20, Wherein the traffic analysis means comprises: Wherein the traffic detection unit checks traffic for a predetermined period of time for the N servers by a predetermined analysis unit time, and detects user log information generated / accumulated by the N servers, And analyzing M traffic for each service function for a predetermined period, The information registration means, The method according to any one of claims 1 to 3, wherein the server is further configured to determine whether to exclude n servers out of the N servers based on the highest traffic and the lowest traffic among the M service- Sets a reference value for putting a number of servers into the server system, A reference point for q (q> 2) or more traffic sections set with the highest traffic to which the predetermined weight is applied, the lowest traffic, and the reference value as the traffic section reference point, A predetermined weight is set for a resource usage amount including at least one of a CPU usage amount, a memory usage amount, a process / thread usage amount, and a session allocation amount for providing the service function to a plurality of terminals within a predetermined time limit in each reference traffic, Traffic information for each traffic is calculated, And stores (or updates) the reference traffic information and the resource usage information for each reference traffic in association with the category information corresponding to the service function in a storage medium. 19. The method according to claim 16 or 18, Wherein the traffic analysis unit is configured to statistically process M traffic for each service function for a predetermined period of time based on at least one of time, day, day, week, and month, The system may be configured to check the traffic reduction period information including at least one of time, day, day, week, and month, traffic of the traffic reduction period and n system elimination target servers for each traffic reduction period, And a scheduling policy processing means for, Wherein the scheduling policy processing means comprises: The method of claim 1, further comprising: storing traffic reduction information in the scheduling D / B, traffic information in the traffic reduction section, and n server exclusion target server information in association with each other, Wherein the control unit controls power of n system excluding servers stored in the scheduling D / B to exclude the system from the server system if the matching information matches the traffic information stored in the scheduling D / If the traffic for each service function identified through the traffic analyzing means does not match the traffic information stored in the scheduling D / B, the power of the n system elimination target servers checked in real time through the resource analyzing means and the server checking means is controlled Wherein the server system is configured to control the power supply of the server. The method according to claim 17 or 19, 20 or 21, The M traffic processing method according to any one of claims 1 to 3, wherein the traffic analyzing means analyzes the M traffic for each service function for a predetermined period based on at least one of time, day, day, week, And a scheduling policy processing means for checking and connecting to the scheduling D / B of the traffic injection period information including at least one of the time zone, day, day, week, and month and the traffic n " And is characterized in that, Wherein the scheduling policy processing means comprises: The system according to claim 1, wherein, when the traffic increase interval information in the scheduling D / B is stored in association with the traffic information of the traffic increase interval and n "system insertion target server information, When the traffic is matched with the traffic information stored in the scheduling D / B, the power of the n "number of system exclusion target servers stored in the scheduling D / B is controlled and put into the server system, If the traffic for each service function identified through the traffic analyzing means does not match the traffic information stored in the scheduling D / B, the power of the n "number of system exclusion target servers checked in real time by the resource analyzing means and the server checking means is controlled To the server system.

Images