KR102322121B1 - Automatic Disaster Recovery System And Recovery Method Thereof - Google Patents

Abstract

According to the present invention, by determining the response time of a plurality of servers constituting the operation center and determining whether a disaster has occurred in the operation center according to the response time, it is possible to more accurately determine whether a disaster has occurred, and the disaster occurrence of the operation center At the time of judgment, a disaster recovery automation system that enables faster startup of the backup center while preventing mutual conflicts between servers by automatically starting a plurality of servers constituting the backup center synchronized with the operation center in parallel for each classified server, and it's about how

Description

Automatic Disaster Recovery System And Recovery Method Thereof

One embodiment of the present invention relates to a disaster recovery automation system and method.

In general, the disaster recovery system protects against damage to the system by a path such as a virus that can invade the system, a catastrophic failure due to sudden accidents such as natural disasters such as fire and earthquake, or data loss that occurs over a period of time. exist to prepare.

The existing disaster recovery system performs a system role such as providing computerized service through a plurality of devices constituting the operation center in the general case where there is no disaster, but when a disaster such as a system failure occurs in the operation center, the In the backup center that is synchronized and backed up from the operation center, it performs the role of the operation center instead. Therefore, when a disaster occurs in the operation center, the operation center must be quickly switched to the backup center.

However, the conventional disaster recovery system depends on the manual operation of the operator for all procedures in the time from the occurrence of a disaster, until the recognition of the disaster and the declaration of the disaster, and the procedure of starting the system of the backup center after the declaration of the disaster. Each step such as conversion, network equipment, server environment setting, data verification and application operation is started one by one manually by the operator.

The operation of the disaster recovery system, which relied on the operator's manual operation, is initiated through the operator's approval process in the conversion stage to the backup center. , it can be a very cumbersome operation for the operator by the intervention of each step by the operator.

Therefore, in recent years, in recognizing the disaster of the operation center as a way to realize a more rapid disaster recovery process, each device constituting the operation center automatically controls the traffic flow, which is the amount of requests that the clients receive. A system has been developed that detects and performs a disaster recovery process by determining that a disaster has occurred in the operation center when the traffic flow is stopped or an abnormal flow pattern is detected.

However, in the disaster recovery system including the method for recognizing a traffic flow based on the traffic flow as described above, even when a disaster does not occur in the operation center, a sudden increase, a sharp drop, or an unstable traffic flow is temporarily detected. Since it can be determined that a disaster has occurred in the operation center even in the early days of the year, it cannot be said to be an ideal model for a disaster recovery system that requires both speed and accuracy in diagnosing the occurrence of a disaster.

Accordingly, there is a need to provide a disaster recovery automation system and method capable of performing a more accurate disaster detection capability and faster disaster recovery process.

The main object of the present invention is to detect and determine the response time of a plurality of servers constituting the operation center when judging whether a disaster has occurred in the operation center in activating the disaster recovery system, so that a more accurate determination of whether a disaster occurs It is to provide a disaster recovery automation system and method that can be provided.

In addition, in the event of a disaster in the operation center, the backup center that can handle the client's request instead of the operation center is started according to an automatic procedure, thereby providing a disaster recovery automation system and method that enables faster operation of the backup center. .

In addition, when a disaster occurs in the operation center and the backup center is started, a plurality of servers constituting the backup center are classified by function, and mutual collision between a plurality of servers that are activated by activating in parallel according to the classified functions To provide a disaster recovery automation system and method that can prevent

In the disaster recovery automation system according to an embodiment of the present invention, when a disaster occurs in an operation center that performs a client's request, a disaster recovery automation system for converting a processing subject of the request from the operation center to a backup center, A first automatic disaster recovery device for checking the operation state of the operation center and a second automatic disaster recovery apparatus for starting the backup center when an abnormality is detected in the operation state of the operation center by the first automatic disaster recovery apparatus Including, when the first automatic disaster recovery device checks the startup status of a plurality of operation servers constituting the operation center and detects an abnormality in the startup state of the plurality of operation servers, the second automatic disaster recovery device classifies a plurality of backup servers included in the backup center into a plurality of groups by function, sets an order for the groups, and sequentially activates the groups according to the order.

In the present invention, the first automated disaster recovery device includes an operation center agent that detects changes in the plurality of operation servers and a change information collection unit that collects the changes from the operation center agent, wherein the second The disaster recovery automation device receives the changes transmitted from the backup center agent and the first disaster recovery automation device that apply the changes to the backup server, so that the backup server can be applied according to the changes It may include a change information processing unit for transmitting the change to the backup center agent.

In the present invention, the first disaster recovery automation device detects a response response time that is a time when the plurality of operation servers respond to the request of the client, and the plurality of operation servers normally start the response response time The operation center is started by a pre-sensing unit and the pre-sensing unit that compares it with a reference reaction time range, which is a range of a response reaction time when the operation is performed, and detects whether an abnormality occurs in the operation center's starting state according to the comparison result. When it is detected that there is an abnormality in the state, it may include a pre-inspection unit for checking the synchronization progress state between the operation center and the backup center.

In the present invention, when the response reaction time is not within the reference reaction time range for a predetermined time, the pre-sensing unit may detect that an abnormality has occurred in the operating state of the operation center.

In the present invention, the pre-sensing unit, a response response time information receiving unit for receiving the information of the response response time from the plurality of operation servers, the longest of the maximum response response time of the response response time and the shortest of the response response time A reference reaction time range determining unit that detects a minimum response time and determines a range of the maximum response time to the minimum response time as the reference reaction time range, and analyzes whether the response reaction time is within the reference reaction time range When the response reaction time analysis unit and the response reaction time is not within the reference reaction time range, it may include a starting state detection unit for detecting that an abnormality has occurred in the starting state of the operation center.

In the present invention, the reference response time range is the range of the maximum response response time to the minimum response response time in which the plurality of operation servers capable of normally processing the request of the client respond to the request of the client. can

In the present invention, the second automated disaster recovery device may start the plurality of backup servers after synchronization between the operation center and the backup center is completed.

In the present invention, the pre-checking unit may further include a service range determining unit for determining the range of the service to be performed by the backup center among the services for which the operation center performs the request of the client.

In the present invention, when an abnormality is detected in the operation state of the plurality of operation servers, the first automated disaster recovery device transmits an operation center abnormality detection signal to the operator of the operation center or the backup center, and the second The disaster recovery automation device may start the plurality of backup servers according to the disaster occurrence determination of the operation center or the operator of the backup center.

In the present invention, when the plurality of backup servers are activated, the first automated disaster recovery device may stop the operation of the plurality of operation servers and block a path of synchronization between the operation center and the backup center.

In the present invention, the second automatic disaster recovery device includes a grouping unit for grouping the plurality of backup servers by function, a startup sequence determining unit for determining the startup sequence for each group of the grouped plurality of backup servers, and each group A start-up time calculator for calculating the start-up time, a start-up interval tuner for adjusting the start-up intervals for each group so that the start-ups for each group do not overlap with each other, and the plurality of backup servers are sequentially started for each group according to the start-up intervals for each group It may include a group-specific starting unit.

In the present invention, the starting time for each group may be an average value of values in which the starting time for each group is pre-measured a plurality of times, or may be a maximum value among values pre-measured a plurality of times.

In the present invention, the starting interval tuning unit may adjust the starting interval for each group so that the starting interval for each group is equal to or longer than the starting time for each group.

In addition, in the present invention, when a disaster occurs in the operation center that performs the client's request, in the disaster recovery automation method for switching the processing subject of the request from the operation center to the backup center, a plurality of components constituting the operation center Detects the response response time, which is the time the operation server responds to the client's request, and checks whether the response response time is within the reference reaction time range, which is the range of the response response time when the plurality of operation servers are normally started a pre-sensing step of detecting whether an abnormality has occurred in the operation center's starting state, and when it is detected that an abnormality has occurred in the operation center's operation center, a preliminary check of checking the synchronization progress state between the operation center and the backup center and a parallel starting step of starting a plurality of backup servers constituting the backup center in parallel, wherein the parallel starting step classifies the plurality of backup servers into a plurality of groups by function, and the groups By setting an order to , the groups may be sequentially activated according to the order.

In the present invention, the pre-sensing step includes a response response time information receiving step of receiving information on the response response time from the plurality of operation servers, the longest maximum response response time and the response response time among the response response times A reference reaction time range determining step of detecting the shortest minimum response time among the minimum response time and determining the range of the maximum response time to the minimum response time as the reference reaction time range, the information derived from the response reaction time information A response reaction time analysis step of analyzing whether the response reaction time is within the reference reaction time range, and when the response reaction time is not within the reference reaction time range, a startup state of detecting that an abnormality has occurred in the startup state of the operation center It may include a detection step.

In the present invention, the reference response time range is the range of the maximum response response time to the minimum response response time in which the plurality of operation servers capable of normally processing the request of the client respond to the request of the client. can

In the present invention, in the pre-sensing step, when the response reaction time is not within the reference reaction time range for a predetermined time, it may be detected that an abnormality has occurred in the operating state of the operation center.

In the present invention, in the parallel starting step, after synchronization between the operation center and the backup center is completed, the plurality of backup servers may be started.

In the present invention, the pre-checking step may further include a service range determining step in which the operation center determines the range of the service to be performed by the backup center among the services for which the client's request is performed.

In the present invention, when the plurality of backup servers are activated, the operation of the plurality of operation servers is stopped, and the path of synchronization between the operation center and the backup center may be blocked.

In the present invention, the parallel startup step includes a grouping step of grouping the plurality of backup servers by function, a startup sequence determining step of determining the startup sequence for each group of the grouped plurality of backup servers, and the startup time for each group A startup time calculation step of calculating It may include a star startup phase.

In the present invention, the starting time for each group may be an average value of values pre-measured a plurality of times for the startup time for each group, or may be a maximum value among values pre-measured a plurality of times.

In the present invention, in the step of tuning the starting interval, the starting interval for each group may be adjusted so that the starting interval for each group is equal to or longer than the starting time for each group.

According to an embodiment of the present invention, in activating the disaster recovery system, by detecting the response time of a plurality of servers constituting the operation center, and determining whether a disaster has occurred in the operation center, more accurate determination of whether a disaster has occurred It is possible to provide a disaster recovery automation system and method that can be activated according to

In addition, it is possible to provide a disaster recovery automation system and method capable of faster business conversion from an operation center to a backup center by starting a backup center that can be started according to an automatic procedure according to whether a disaster has occurred.

In addition, when a disaster occurs in the operation center and the backup center is started, a plurality of servers constituting the backup center are started in parallel for each classified function, thereby preventing mutual conflict between the plurality of servers and starting the backup center. It is possible to provide a disaster recovery automation system and method capable of stable operation.

1 is a diagram illustrating a system configuration between an operation server, a backup server, and an automated disaster recovery device provided therein according to an embodiment of the present invention.
Figure 2 shows a disaster recovery automation system that works with the operation server and the backup server according to an embodiment of the present invention.
3 shows the normal response response time of the operation server and the response response time when a disaster occurs according to an embodiment of the present invention.
Figure 4 shows the start-up of the backup center according to an embodiment of the present invention.
5 is a flowchart of a disaster recovery automation method according to an embodiment of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and will be described in detail through the detailed description. However, this is not intended to limit the specific embodiments of the present invention, and may include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

In describing the present invention, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof may be omitted.

In addition, in this specification, when a component is referred to as “connected” or “connected” with another component, the component may be directly connected or directly connected to the other component, but in particular Unless there is an opposing description, it may be connected or connected through another element in the middle.

In the disaster recovery automation system according to an embodiment of the present invention, the operation center that performs the client's request is damaged by the intrusion of a virus or natural disasters such as fire or earthquake, so that it cannot normally receive or process the client's request. At the same time, it may be a system configured so that a backup center synchronized with the operation center and configured to include the same components as the operation center can take over the operation of the operation center.

Therefore, when a disaster occurs in the operation center, in order to minimize damage to clients who process business using the operation center, there is a need for rapid business conversion from the operation center to the backup center. That is, in order to minimize damage to the client that inevitably occurs according to a disaster in the operation center, there is a need to quickly provide the client with the startup of the backup center close to the normal operation of the operation center.

In addition, when a disaster occurs in the operation center, the technology for quickly switching to the backup center and determining whether a disaster occurs in the operation center may also be a very important factor in the disaster occurrence automation system. If a simple abnormal symptom that may appear temporarily even when a disaster does not occur in the operation center is determined to be a disaster occurrence factor in the operation center, it may cause great confusion to the operation center and clients using the operation center. have.

Therefore, in the following, it is possible to determine whether a disaster has occurred in the operation center through a more accurate diagnosis, and in the event of a disaster, an automated disaster recovery system capable of rapidly switching from the operation center to the backup center will be examined.

1 is a diagram illustrating a system configuration between an operation server, a backup server, and an automated disaster recovery device provided therein according to an embodiment of the present invention.

First, before describing the disaster recovery automation system according to an embodiment of the present invention, the operation center 100 and the backup center 200 to which the disaster recovery automation system can be applied may be briefly reviewed.

The operation center 100 for performing the client's request includes a plurality of operation servers 111, and through the plurality of operation servers 111, the client's request can be divided and performed for each unit task.

The plurality of operation servers 111 may be configured differently depending on the type of service that the operation center 100 provides to clients, for example, a web server, a linked server, an application server, an application server, a database server. and the like.

The backup center 200 may also include a plurality of backup servers 211 corresponding to the plurality of operation servers 111 constituting the operation center 100 . This may be to enable the backup center 200 to perform a service for a client on behalf of the operation center 100 when a disaster occurs in the operation center 100 .

Here, the plurality of backup servers 211 may be provided as a device or system equivalent to the same configuration as the plurality of operation servers 111, but may be provided more simplified than the plurality of operation servers 111, In a disaster recovery system, the backup center 200 is generally configured to prepare for a disaster in the operation center 100 with a simplified configuration including the essential configuration of the operation center 100 .

If the plurality of backup servers 211 are configured to have the same configuration as the plurality of operation servers 111 , the operation center together with the operation center 100 mainly used in normal times, that is, a situation in which a disaster does not occur. Since the backup center 200, which is temporarily used only in the event of a disaster of 100, also needs to perform the same maintenance and management work as the operation center 100, the cost required to maintain and manage the backup center 200 as well as In addition, the number of administrators is also increased, and as an effective disaster countermeasure means, the plurality of backup servers 211 constituting the backup center 200 are greater than the plurality of operation servers 111 constituting the operation center 100 . The configuration is simplified, but essential components among the plurality of operation servers 111 may be included.

The above-described operation center 100 and the backup center 200 are merely an embodiment to which the present invention can be applied to help understand the present invention, and the contents of the present invention to be described below can be applied and used. Any configuration is possible.

Hereinafter, a disaster recovery automation system according to an embodiment of the present invention will be described by looking at the disaster recovery automation system applied to the operation center 100 and the backup center 200 having the above configuration.

Figure 2 shows a disaster recovery automation system that works with the operation server 111 and the backup server 211 according to an embodiment of the present invention.

1 and 2 , the disaster recovery automation system according to an embodiment of the present invention is connected to the operation center 100 and a first disaster recovery automation device for checking the operating state of the operation center 100 . When an abnormality is detected in the operation state of the operation center 100 by the first automated disaster recovery device 1000, connected to the backup center 200 synchronized with the 1000 and the operation center 100 , a second automatic disaster recovery device 2000 for starting the backup center 200 may be included.

The backup center 200 needs to be prepared in preparation for the operation of the operation center 100, and in case of a disaster, the operation center 100 to implement the same and similar service as the normal service of the operation center 100 can be synchronized with

Specifically, the operation center 100 and the backup center 200 may be synchronized periodically or in real time. In order for the backup center 200 to approach the normal operation of the operation center 100, it is preferable that changes in the operation center 100, ie, changes such as installation or removal of new programs, are synchronized periodically or in real time. do.

The first disaster recovery automation device 1000 collects change information for collecting the changes from the operation center agent 1250 for detecting changes in the plurality of operation servers 111 and the operation center agent 1250 may include a unit 1200, wherein the second automated disaster recovery device 2000 includes a backup center agent 2250 that applies the changes to the backup server 211 and the first automated disaster recovery device ( 1000), and a change information processing unit 2200 for receiving the change transmitted from and transmitting the change to the backup center agent 211 so that the backup server 211 can be applied according to the change. can

Referring to FIG. 1 , the operation center agent 1250 is connected to the plurality of operation servers 111 , so that changes of the plurality of operation servers 111 constituting the operation center 100 are monitored in real time or periodically. may be collected, and the subject of collecting the changes may be the change information collecting unit 1200 . That is, the change information collection unit 1200 may collect the changes of the plurality of operation servers 111 through the operation center agent 1250, and collect the changes of the plurality of operation servers 111 through the operation center agent 1250. It can be transmitted to the change information processing unit 2200 installed in the second automatic disaster recovery device 2000 .

The change information processing unit 2200 that has received the changes of the plurality of operation servers 111 transmits the changes to the plurality of backups via the backup center agent 2250 connected to the plurality of backup servers 211 . By applying the above changes to the server 211, it is possible to synchronize the operation server and the backup server.

As described above, the change information collection unit 1200 and the change information processing unit 2200 may provide a synchronization path between the operation center 100 and the backup center 200, and thus, the backup center 200 ) is synchronized by reflecting the changes of the operation center 100, in preparation for a disaster of the operation center 100, it is possible to prepare to provide the same and similar service as the service of the operation center 100 to the client. .

The first disaster recovery automation device 1000 connected to the backup center 200 detects whether a disaster has occurred in the operation center 100 , and thus utilizes the backup center 200 synchronized with the operation center 100 . provide an excuse

As shown in FIGS. 1 and 2 , the first automated disaster recovery device 1000 includes the change information collection unit 1200 , the operation center agent 1250 and the first automated disaster recovery server 1100 . can be

As described above, the change information collection unit 1200 and the operation center agent 1250 are provided for the main purpose of synchronizing the backup server 211 and the operation server 111, and automate the first disaster recovery. The server 1100 may be provided to detect whether a disaster has occurred in the operation center 100 .

The first automated disaster recovery server 1100 may detect whether an abnormality has occurred in the operation center 100 by detecting the activation state of the plurality of operation servers 111 constituting the operation center 100, When the first automated disaster recovery server 1100 detects an abnormality in the activation state of the plurality of operation servers 111 , the second automated disaster recovery device 2000 starts the plurality of backup servers 211 . to convert the client's request processing subject to the backup center 200 .

Here, when the first automated disaster recovery server 1100 detects an abnormality in the operation state of the plurality of operation servers 111 , the first automated disaster recovery server 1100 is the operation center 100 or the It is possible to transmit an abnormality detection signal to the operation center 100 to the operator of the backup center 200, which indicates the abnormality of the operation state of the plurality of operation servers 111 detected by the first automated disaster recovery server 1100. By checking the operation center 100 or the operator of the backup center 200 , it is for more accurately determining whether a disaster has occurred in the operation center 100 .

If the operator of the operation center 100 or the backup center 200 examines the operation center 100 abnormal detection signal, the plurality of operation servers 111 detected by the first automated disaster recovery server 1100 ), when it is determined that a disaster has occurred in the operation center 100, the operator of the operation center 100 or the backup center 200 determines that a disaster has occurred in the operation center 100 It can be down, and accordingly, the plurality of backup servers 211 can be activated.

That is, the second automated disaster recovery server 2100 is the operation center 100 or the backup center 200 determined based on the detection signal of an abnormality in the operation center 100 transmitted from the first automated disaster recovery server 1100 . ), the plurality of backup servers 211 may be activated according to the disaster occurrence decision of the operator. The contents regarding the activation of the plurality of backup servers 211 will be described later.

Looking more closely at the first automated disaster recovery server 1100, which serves to detect the operating state of the operation server 111 among the first automated disaster recovery device 1000, the first automated disaster recovery server ( 1100) detects a response reaction time, which is a time when the plurality of operation servers 111 respond to a client's request, and determines the response reaction time as the response reaction time when the plurality of operation servers 111 are normally started. It may include a pre-sensing unit 1110 for detecting whether an abnormality occurs in the operation state of the operation center 100 according to the comparison result compared with the reference reaction time range, which is the range, and the pre-sensing unit 1110 may include a pre-inspection unit 1120 for checking a synchronization progress state between the operation center 100 and the backup center 200 synchronized with the operation center 100 .

Since the plurality of operation servers 111 constituting the operation center 100 respond to perform functions for each unit task according to the request of the client, the time that the plurality of operation servers 111 respond to the request of the client This may be the response reaction time.

The response reaction time may be used as a measure for determining whether a disaster has occurred in the operation center 100 . That is, when a disaster occurs in the operation center 100 and the plurality of operation servers 111 constituting the operation center 100 do not function properly, the response response time may be slow. In addition, when the plurality of operation servers 111 are infected with a virus or malfunction due to an error, the response response time may increase abnormally in speed, so the response response time is a disaster of the plurality of operation centers 100 . It may be a very suitable means for diagnosing whether or not it has occurred.

The conventional disaster recovery automation system disclosed a technique for determining whether a disaster occurred in the operation center 100 based on the number of client traffic, which is a record of the client's request to the plurality of operation servers 111, In this case, since the number of traffic is very different for each time period or various environmental factors such as seasons, even if a disaster does not occur in the operation center 100, when the number of traffic temporarily increases or decreases, the corresponding traffic Due to the number, a disaster of the operation center 100 may be declared, and the accuracy of the determination may be very low.

On the other hand, the disaster recovery automation system according to the present invention determines whether a disaster has occurred in the operation center 100 through the response time of the plurality of operation servers 111 detected periodically or in real time, rather than the number of traffic. It is possible to provide very high accuracy in determining whether a disaster has occurred in the operation center 100 .

Here, since the response response time is used as a measure to determine whether an abnormality has occurred in the operating state of the plurality of operation servers 111, a comparison group of the response response time that can determine whether the response response time is located within a normal value Thus, the reference reaction time range can be suggested.

3 shows the normal response response time of the operation server 111 and the response response time when a disaster occurs according to an embodiment of the present invention.

The reference reaction time range is the time during which the plurality of operation servers 111 in normal times where no disaster occurs, that is, the plurality of operation servers 111 that can normally process the client's request respond to the client's request. The maximum response time and the minimum response time may be calculated and may be a range set as a maximum value and a minimum value, respectively.

The reference reaction time may be a value previously measured and stored a plurality of times through the plurality of normal operation servers 111 , and the range may be provided as a value measured for a certain time in order to provide a more accurate range.

That is, as shown in FIG. 3 , the reference reaction time range provided as a range (y-axis) of the maximum response time to the minimum response time is provided as a value (x-axis) provided as a range for a certain time. By being, the range of the maximum to the minimum value of the response response time according to the request to the client for the plurality of operation servers 111 is more normal and similar to the range value of the response response time of the plurality of operation servers 111 can be provided. have.

Hereinafter, the detection of the activation state of the operation center 100 through the response reaction time and the reference reaction time range can be viewed in detail.

The pre-sensing unit 1110 may serve to detect in advance whether a disaster has occurred in the operation center 100 through the above-described response reaction time and the reference reaction time range.

Specifically, the pre-sensing unit 1110 is a response response time information receiving unit 1111 for receiving the information of the response response time from the plurality of operation servers 111, the range of the maximum response time to the minimum response time. A reference reaction time range determining unit 1112 for calculating and determining the reference reaction time range, and a response reaction time analyzing unit 1113 for analyzing whether the response reaction time derived from the information on the response reaction time is within the reference reaction time range ), and when the response reaction time is not within the range of the reference reaction time, it may include a start state detection unit 1114 for detecting that an abnormality has occurred in the start state of the operation center 100 .

The response response time information receiving unit 1111 may receive information on the response response time from the plurality of operation servers 111 .

The information of the response response time may be information recorded in each of the plurality of operation servers 111 while the plurality of operation servers 111 respond to the client's request, and the plurality of operation servers 111 and When the connected operation center agent 1250 requests information on the response response time from the plurality of operation servers 111 , it may be a value received by the response response time information receiving unit 1111 .

However, the response response time information is not limited to being received by the response response time information receiving unit 1111 through the above path, and the response response time information receiving unit 1111 may receive the response response time information. Any possible path is possible, and as the response response time information receiving unit 1111 is connected to the plurality of operation servers 111 in a wired or network form, the plurality of operation servers 111 respond to the client's request. It may be configured as a path in which the response response time information is automatically received by the response response time information receiving unit 1111 at the same time as responding.

Subsequently, the reference reaction time range determining unit 1112 detects the longest maximum response time among the response time and the shortest minimum response time among the response time, and the maximum response time to the minimum response time. The time range may be determined as the reference reaction time range.

As described above, the reference reaction time range determining unit 1112 may be determined by calculating the range of the maximum response time to the minimum response time through a previously measured value, but the operation center 100 or the It is also possible to be determined according to the numerical value set by the operator of the backup center 200 .

The response response time received through the response response time receiver may be analyzed by the response response time analyzer 1113 whether it is within the reference response time range. Since the reference reaction time range is provided in a form in which the range of the maximum response time to the minimum response time is continued for a predetermined time as described above, the response reaction time analyzer 1113 determines that the response reaction time is a predetermined time. It can be analyzed whether it is within the reference reaction time range.

As a result of the analysis of the response reaction time analysis unit 1113 , when the response reaction time is not within the reference reaction time range for a predetermined time, the starting state detection unit 1114 is the operation center 100 in the starting state. It can be detected that an abnormality has occurred.

The response reaction time measured for each time period may temporarily deviate from the reference reaction time range. In this abnormal situation, it may be determined that an abnormality has occurred in the activation state of the plurality of operation servers 111 , but in general, the temporary abnormal situation is affected by the network or the device used by the plurality of operation servers 111 , or , and even may exhibit temporary abnormalities under the influence of environmental requirements such as weather.

Therefore, even if the response time is out of the range of the reference reaction time, it may not be an accurate determination of the occurrence of a disaster to immediately make a decision in an abnormal situation, and it is preferable to observe the time zone value of the response time for a predetermined time. do.

Based on the above reasons, the pre-sensing unit 1110 according to the present invention has an abnormality in the starting state of the operation center 100 when the response reaction time is not within the reference reaction time range for a predetermined time. can be detected as occurring.

The response response time for each time period detected from the plurality of operation servers 111 in normal times when a disaster does not occur can be confirmed through (a) of FIG. 3 , and referring to FIG. Although it was detected that some of the measured response response times of the response response times were out of the reference response time range, it can be confirmed that they are measured again within a normal range, that is, within the reference response time range.

On the other hand, as shown in (b) of FIG. 3, the response response time for each measurement time period detected from the plurality of operation servers 111 in which the abnormality of the activation state occurred is out of the reference reaction time range for a predetermined time. can be checked

In this case, since the response reaction time is out of the reference reaction time range due to an abnormal behavior for more than a predetermined time, the startup state detection unit 1114 detects that an abnormality has occurred in the startup state of the plurality of operation servers 111 . It can be done, and it is necessary to promptly follow the response accordingly.

As described above, when it is detected that an abnormality has occurred in the operation state of the operation center 100, that is, a plurality of operation servers 111 constituting the operation center 100, the second automatic disaster recovery server ( 2100) may start the plurality of backup servers 211 constituting the backup center 200, and before that, the operation center 100 and the operation center 100 through the pre-inspection unit 1120 A synchronization progress state between the backup center 200 synchronized with the .

As described above, the operation center 100 and the backup center 200 may be synchronized periodically or in real time, and the plurality of operation servers 111 and the backup center 200 constituting the operation center 100 may be synchronized. ) of the plurality of backup servers 211 constituting the synchronization in progress, an abnormality in the activation state of the plurality of operation servers 111 may occur.

Here, if the plurality of backup servers 211 are activated during synchronization with the plurality of operation servers 111 , the subject of the client's request is switched from the operation center 100 to the backup center 200 . The operation server 111 should be stopped from starting, and accordingly, synchronization between the plurality of operation servers 111 and the plurality of backup servers 211 is stopped.

However, in the above case, when the synchronization between the plurality of operation servers 111 and the plurality of backup servers 211 is stopped in an incomplete state, the plurality of backup servers 211 are synchronized to an incomplete synchronization. Due to this, it is not possible to implement a close operation state of the plurality of operation servers 111, and an error may occur due to semi-forced synchronization termination.

Therefore, when the synchronization between the operation center 100 and the backup center 200 is checked by the pre-inspection unit 1120 as being in progress, the second automated disaster recovery server 2100 is the operation center 100 . And after synchronization between the backup centers 200 is completed, the plurality of backup servers 211 may be started.

Of course, when it is checked by the pre-inspection unit 1120 that synchronization between the operation center 100 and the backup center 200 is not in progress, the second automated disaster recovery server 2100 performs the plurality of backups. The server 211 may be started immediately.

The pre-inspection unit 1120 further includes a service range determination unit 1122 for determining the scope of the task to be performed in the backup center 200 among the tasks in which the operation center 100 performs the client's request. can do.

As described above, the backup center 200 is provided in preparation for a disaster that may suddenly occur in the operation center 100 , so the plurality of the backup servers 211 constituting the backup center 200 are operated in the operation center 100 . Compared to the plurality of operation servers 111 constituting the center 100, it may be provided to be simplified.

Therefore, when a disaster occurs in the operation center 100 and the subject of processing the client's request is switched to the backup center 200, all tasks in which the operation center 100 performs the client's request are performed by the backup center 200 ) may not be converted to Accordingly, the service range determining unit 1122 determines the scope of the task to be switched from the task of the operation center 100 before the plurality of backup servers 211 constituting the backup center 200 are started. can

The service range determining unit 1122 may be changed according to the configuration scale of the plurality of backup servers 211 for the plurality of operation servers 111 , but the plurality of backup servers 211 may be configured for the operation of the plurality of operation servers 111 . Since all essential components of the server 111 are configured to be included, there is no problem in being converted and used as the processing subject of the client's request.

However, the backup center 200 includes the plurality of backup servers 211 in the same configuration as the plurality of operation servers 111 constituting the operation center 100, so that the operation center 100 has the same configuration. Since it may be configured, in this case, the service range determination unit 1122 may determine the work range of the backup center 200 to be equal to or different from the range of the operation center 100 .

Subsequently, when the plurality of backup servers 211 are started, the first automated disaster recovery apparatus 1000 stops the operation of the plurality of operation servers 111 , and the operation center 100 and the backup center It is possible to block the synchronization path between (200).

The actuation of the plurality of backup servers 211 is a process for the plurality of backup servers 211 to be converted as subjects of client request processing, and thus the plurality of backup servers 211 are operated while the plurality of backup servers 211 are running. If the server 111 is started, the path through which the client's request is received is confused, and the disaster situation occurring in the operation center 100 may be further aggravated.

Above all, since the start-up of the plurality of backup servers 211 proceeds when a disaster occurs in the operation center 100, the backup center 200 is started in a state that completely excludes the effect of an error caused by the disaster. it is preferable

Therefore, when the plurality of backup servers 211 are activated, the operation of the plurality of operation servers 111 must be stopped, and through this, the synchronization path between the operation center 100 and the backup center 200 is also fundamental. can be blocked with

Subsequently, the second automated disaster recovery server 2100 may start the plurality of backup servers 211 .

As described above, the backup center 200 is the operation center 100 and the operator of the backup center 200 based on the detection signal of the operation center 100 abnormality received from the first disaster recovery automation device 1000. Since it is activated according to the decision on the occurrence of a disaster, it can be activated through a more accurate judgment of the occurrence of a disaster.

In addition, since the second automated disaster recovery server 2100 automatically starts the plurality of backup servers 211 according to the disaster occurrence decision of the operator of the operation center 100 or the backup center 200, the conventional It may be possible to start the backup server 211 faster than the method in which it was started through the approval of the operator for each unit step constituting the start-up step.

Specifically, the second automatic disaster recovery server 2100 is a grouping unit 2111 for grouping the plurality of backup servers 211 by function, and determining the startup sequence for each group of the grouped plurality of backup servers 211 A start-up sequence determining unit 2112, a start-up time calculator 2113 for calculating the start-up times for each group, a start-up interval arbiter 2114 for adjusting the start intervals for each group so that the group-by-group starts do not overlap with each other, and the plurality of It may include a group activation unit 2115 for starting the backup server 211 for each group.

Figure 4 shows the start-up of the backup center 200 according to an embodiment of the present invention.

In order that the plurality of backup servers 211 constituting the backup center 200 do not collide with each other during operation, it is preferable to perform a similar task or to perform a similar task or divide by the same server, that is, group by function and start by group. . Accordingly, the grouping unit 2111 may group the plurality of backup servers 211 by function to prevent mutual conflicts in the system by sequentially starting the plurality of backup servers 211 by function. .

Here, the plurality of backup servers 211 may be grouped with the same function, but even servers having different functions may be classified into the same group and grouped as long as they do not conflict with each other even when they are started at the same time.

Referring to an exemplary embodiment with reference to FIG. 4 , it can be seen that the plurality of backup servers 211 are classified by function (first group to fourth group). Here, the first to fourth groups classified by function may be configured with the same servers or servers that are different from each other but do not conflict with each other during simultaneous startup.

The grouping unit 2111 classifies the plurality of backup servers 211 constituting the backup center 200 by function, so that when the backup center 200 is started, the plurality of backup servers 211 are individually Since the groups classified by function are not activated, but are activated collectively, the speed of maneuvering can be greatly improved while preventing mutual conflicts between groups.

Subsequently, the start-up order determining unit 2112 may determine the group-by-group start-up order of the grouped plurality of backup servers 211 .

When the plurality of backup servers 211 are activated for each group, if the startup order for each group is not taken into account, each group classified from the plurality of backup servers 211 is activated through another group activated before and after startup. Interference may cause errors. In other words, the plurality of backup servers 211 are sequentially activated for each group, so that the entire system can be built. There may be errors in the build.

As an example, when one group consisting of a server in charge of data verification and application driving is started up before another group running the system, data verification and application running on the system that are not started can act so that the backup center An error may occur in (200).

Therefore, it is preferable that the startup sequence is determined by considering that the plurality of groups classified from the plurality of backup servers 211 do not cause errors according to context when starting sequentially, and for this, the startup sequence determining unit 2112 ) may determine the startup sequence for each group based on data through the startup results of the plurality of backup servers 211 that have been normally activated in advance.

When the plurality of backup servers 211 are started, the second automated disaster recovery server 2100 determines the start-up order for each group classified so that they can start normally without causing a mutual collision so that they can be started.

However, when the plurality of groups classified by function from the plurality of backup servers 211 start up, if the start-up time overlaps, this may also cause an error before the start-up of the backup center 200 .

Accordingly, the start-up time calculator 2113 can calculate the time required for each of the plurality of groups to be started, and based on this, the start-up interval tuner 2114 to be described later determines the start-up intervals of each of the groups. can be adjusted

The start-up time calculator 2113 may calculate the start-up time for each group based on the previously-measured start-up time for each group. Specifically, the start time for each group may be a value pre-measured a plurality of times, and the start time calculator 2113 may determine the pre-measured value as the start time for each group.

Here, the startup time for each group may be different depending on the condition of the device at startup or the network connection state, etc., so for more reliable error prevention, a value greater than or equal to the average value of the values measured a plurality of times in advance is the startup time for each group. may be determined, and more preferably, a maximum value among values previously measured a plurality of times may be determined as the startup time for each group.

Subsequently, the starting interval tuning unit 2114 may adjust the group-by-group starting interval so that the group-by-group maneuvers do not overlap each other.

Specifically, in the startup interval arranging unit 2114, each of the plurality of groups classified from the plurality of backup servers 211 is activated according to the startup sequence determined through the startup sequence determining unit 2112, the startup sequence Accordingly, the starting interval may be adjusted based on the starting time for each group calculated through the starting time calculation unit 2113 so that starting does not occur at a time overlapping with other groups activated before and after.

Since the time at which each of the plurality of groups is started can be calculated through the start time calculation unit 2113, it is preferable to set an interval between the plurality of groups to be longer than the time at which each start is completed.

Accordingly, the startup interval tuning unit 2114 may adjust the startup interval for each group so that the startup interval for each group is greater than or equal to the startup time for each group. 211) until the start of one group constituting the group is completed, the duplication of maneuvers with other groups can be prevented, so that mutual conflicts for each group can be prevented.

Subsequently, the group start unit 2115 may sequentially start the plurality of backup servers 211 for each group according to the group start interval.

The first disaster recovery automation device 1000 detects the response response time of the plurality of operation servers 111 constituting the operation server 111 in real time or periodically to detect abnormalities in the operation state of the plurality of operation servers 111 . can be detected, so it is possible to detect whether an abnormality has occurred in the operating state of the operation center 100 in a more accurate way.

In addition, the second automated disaster recovery device 2000 is based on the operation center 100 abnormality detection signal provided from the first automated disaster recovery device 1000 of the operation center 100 or the backup center 200 . The backup center 200 can be automatically started according to the operator's decision on the occurrence of a disaster, and when the plurality of backup servers 211 constituting the backup center 200 are started, the plurality of backup servers 211 can be classified into a plurality of groups by function, and the groups can be sequentially started according to the order by setting an order to the groups. It makes it possible to provide a disaster recovery automation system capable of promptly starting the backup center 200 .

Hereinafter, a disaster recovery automation method according to an embodiment of the present invention can be looked at.

The disaster recovery automation method is not particularly limited as long as it is a disaster recovery-related device or system to which the method to be described below can be applied, and may be applied to various types of disaster recovery means. By looking at the disaster recovery automation method as an example, the disaster recovery automation method according to the present invention can be understood.

Accordingly, in describing the disaster recovery automation method, the aforementioned disaster recovery automation system and all components constituting the same may be shared below.

5 is a flowchart of a disaster recovery automation method according to an embodiment of the present invention. Hereinafter, the disaster recovery automation method can be viewed with reference to FIGS. 1 to 5 .

In the disaster recovery automation method according to an embodiment of the present invention, when a disaster occurs in the operation center 100 that performs the client's request, the processing subject of the request is switched from the operation center 100 to the backup center 200 As a disaster recovery automation method, the plurality of operation servers 111 constituting the operation center 100 detects a response response time, which is a time to respond to a client's request, and sets the response response time to the plurality of operation servers ( 111) is a pre-detection step (S100) of detecting whether an abnormality occurs in the starting state of the operation center 100 by checking whether it is within the reference reaction time range, which is the range of the response reaction time when the operation center is normally started (S100), the operation center When it is detected that an abnormality has occurred in the startup state of (100), a pre-check step (S200) of checking the synchronization progress state between the operation center 100 and the backup center 200, and the backup center 200 It may include a parallel starting step (S300) of starting the plurality of backup servers 211 in parallel to configure.

Since the plurality of operation servers 111 constituting the operation center 100 respond to perform functions for each unit task according to the request of the client, the time that the plurality of operation servers 111 respond to the request of the client This may be the response reaction time.

The response reaction time may be used as a measure for determining whether a disaster has occurred in the operation center 100 . That is, when a disaster occurs in the operation center 100 and the plurality of operation servers 111 constituting the operation center 100 do not function properly, the response response time may be slow, and When the plurality of operation servers 111 are infected with a virus or malfunction due to an error, the response response time may be abnormally fast, so the response response time is very difficult to diagnose whether the plurality of disasters have occurred. It may be a suitable means.

Here, since the response response time is used as a measure to determine whether an abnormality has occurred in the operation state of the plurality of operation servers 111, a comparison of the response response time that can determine whether the response response time is located within a normal balance As a group, the reference reaction time range may be presented.

The reference reaction time range is the time during which the plurality of operation servers 111 in normal times in which a disaster does not occur, that is, the plurality of operation servers 111 that can normally process the client's request respond to the client's request. The maximum response time and the minimum response time may be calculated and set to a maximum value and a minimum value, respectively.

The reference reaction time may be a value previously measured and stored a plurality of times through the plurality of normal operation servers 111 . In order to provide a more accurate range, the range may be provided as a value measured for a certain period of time.

Hereinafter, the detection of the activation state of the operation center 100 through the response reaction time and the reference reaction time range can be viewed in detail.

The pre-sensing step (S100) may serve to detect in advance whether a disaster has occurred in the operation center 100 through the above-described response reaction time and the reference reaction time range.

Specifically, the pre-sensing step (S100) is a response response time information receiving step (S110) of receiving the information of the response response time from the plurality of operation servers 111, the maximum response time to the minimum response time A reference reaction time range determining step (S120) of determining the reference reaction time range by calculating a range, a response reaction time analysis step of analyzing whether the response reaction time derived from the information on the response reaction time is within the reference reaction time range (S130) and when the response reaction time is not within the reference reaction time range, it may include a starting state detection step (S140) of detecting that an abnormality has occurred in the starting state of the operation center 100.

In the response response time information receiving step ( S110 ), the response response time information receiving unit 1111 may receive information on the response response time from the plurality of operation servers 111 .

The information of the response response time may be information recorded in each of the plurality of operation servers 111 while the plurality of operation servers 111 respond to the client's request, and the plurality of operation servers 111 and When the connected operation center agent 1250 requests information on the response response time from the plurality of operation servers 111 , it may be a value received by the response response time information receiving unit 1111 .

However, the response response time information is not limited to being received by the response response time information receiving unit 1111 through the above path, and the response response time information receiving unit 1111 may receive the response response time information. Any possible path is possible, and as the response response time information receiving unit 1111 is connected to the plurality of operation servers 111 in a wired or network form, the plurality of operation servers 111 respond to the client's request. It may be configured as a path in which the response response time information is automatically received by the response response time information receiving unit 1111 at the same time as responding.

Subsequently, in the step of determining the reference reaction time range ( S120 ), the reference reaction time range determiner 1112 may determine the reference reaction time range as a comparison group of the response reaction times.

In the reference reaction time range determining step (S120), as described above, the range of the maximum response time to the minimum response time may be calculated and determined based on a previously measured value, but the operation center 100 or the It is also possible to be determined according to the numerical value set by the operator of the backup center 200 .

The response reaction time received by the response reaction time receiving unit through the response reaction time information receiving step (S110) is the reference reaction time through the response reaction time analysis unit 1113 in the response reaction time analysis step (S130). It can be analyzed whether it is within the range. Since the reference reaction time range is provided in a form in which the range of the maximum response time to the minimum response time is continued for a certain time as described above, the response reaction time in the analysis step S130 is the response time It can be analyzed whether it is within the reference reaction time range for a predetermined time.

As a result of the analysis in the response reaction time analysis step (S130), when the response reaction time is not within the reference reaction time range for a predetermined time, the starting state detecting unit 1114 in the starting state detecting step (S140) may detect that an abnormality has occurred in the operating state of the operation center 100 .

The response reaction time measured for each time period may temporarily deviate from the reference reaction time range. In this abnormal situation, it may be determined that an abnormality has occurred in the activation state of the plurality of operation servers 111 , but in general, a temporary abnormal situation is the influence of the network or the device used by the plurality of operation servers 111 . It is often affected by environmental factors, such as weather, or even temporarily anomalies.

Therefore, even if the response response time is out of the range of the reference response time, it is not an accurate diagnosis to immediately make a decision in an abnormal situation, and it is preferable to observe the time value of the response response time for a predetermined time.

When the response reaction time is not within the reference reaction time range for a predetermined time in the pre-sensing step (S100) according to the present invention based on the above reasons, the operation center 100 has an abnormality in the starting state. can be detected as occurring.

As described above, when it is detected that an abnormality has occurred in the operation state of the operation center 100 , that is, a plurality of operation servers 111 constituting the operation center 100 through the pre-detection, the pre-inspection In step S200, a synchronization progress state between the operation center 100 and the backup center 200 synchronized with the operation center 100 may be checked through the pre-inspection unit 1120 .

The operation center 100 and the backup center 200 may be synchronized periodically or in real time, and thus the plurality of operation servers 111 and the backup center 200 constituting the operation center 100 may be synchronized. In the plurality of backup servers 211 constituting the synchronization, an abnormality in the activation state of the plurality of operation servers 111 may occur.

Here, if the plurality of backup servers 211 are activated during synchronization with the plurality of operation servers 111 , the subject of the client's request is switched from the operation server 111 to the backup server 211 , the The operation of the plurality of operation servers 111 should be stopped, and accordingly, synchronization between the plurality of operation servers 111 and the plurality of backup servers 211 is stopped.

However, when the synchronization between the plurality of operation servers 111 and the plurality of backup servers 211 is stopped in an incomplete state, the plurality of backup servers 211 are synchronized with the uncompleted synchronization as described above. Due to this, it is not possible to implement a close operation state of the plurality of operation servers 111, and an error may occur due to semi-forced synchronization termination.

Therefore, when it is checked that the synchronization between the operation center 100 and the backup center 200 is in progress by the pre-checking step (S200), the parallel starting step (S300) is the operation center 100 and the After synchronization between the backup centers 200 is completed, it is preferable to start up the plurality of backup servers 211 .

Of course, when it is checked that the synchronization between the operation center 100 and the backup center 200 is not in progress by the pre-checking step (S200), the plurality of backup servers in the parallel starting step (S300) (211) can also be started immediately.

The pre-checking step (S200) further includes a service range determining step (S220) of determining the scope of the task to be performed in the backup center 200 among the tasks in which the operation center 100 performs the client's request. can do.

As mentioned above, since the backup center 200 is provided in preparation for a disaster that may occur suddenly in the operation center 100 , a plurality of the backup servers 211 constituting the backup center 200 are operated in the operation center 100 . Compared to the plurality of operation servers 111 constituting the center 100, it may be provided to be simplified.

Therefore, when a disaster occurs in the operation center 100 and the subject of processing the client's request is switched to the backup center 200, all tasks in which the operation center 100 performs the client's request are performed by the backup center ( 200) cannot be converted. Accordingly, in the service range determining step (S220), the service range determining unit 1122 performs the operation of the operation center 100 before the plurality of backup servers 211 constituting the backup center 200 are started. may determine the scope of the work to be transferred from.

The scope of the task determined in the service range determining step (S220) may be changed according to the configuration scale of the plurality of backup servers 211 for the plurality of operation servers 111, but the plurality of backup servers ( 211) is configured to include all essential components among the plurality of operation servers 111, so there is no problem in being converted and used as the processing subject of the client's request.

However, the backup center 200 includes the plurality of backup servers 211 in the same configuration as the plurality of operation servers 111 constituting the operation center 100, so that the operation center 100 is equivalent to Since it may be configured as a configuration, in this case, the work range of the backup center 200 in the service range determination step (S220) may be determined to be equal to or different from the range of the operation center 100.

Then, when the plurality of backup servers 211 are started in the parallel startup step (S300), the operation of the plurality of operation servers 111 is stopped, and the operation center 100 and the backup center 200 ) the path of synchronization may be blocked.

The actuation of the plurality of backup servers 211 is a process for the plurality of backup servers 211 to be converted as subjects of client request processing, and thus the plurality of backup servers 211 are operated while the plurality of backup servers 211 are running. If the server 111 is started, the path through which the client's request is received is confused, and the disaster situation occurring in the operation center 100 may be further aggravated.

Above all, since the start-up of the plurality of backup servers 211 proceeds when a disaster occurs in the operation center 100, it is important that the backup center 200 is started in a state that completely excludes the effect of an error caused by the disaster. desirable.

Therefore, when the plurality of backup servers 211 are started, the start of the plurality of operation servers 111 may be stopped, and through this, the synchronization path between the operation center 100 and the backup center 200 is also shown. can be blocked in the first place.

Subsequently, in the parallel starting step (S300), the plurality of backup servers 211 may be activated.

In order that the plurality of backup servers 211 constituting the backup center 200 do not collide with each other during operation, it is preferable to perform a similar task or to perform a similar task or divide by the same server, that is, group by function and start by group. . Therefore, in the grouping step (S310), the plurality of backup servers 211 can be grouped by function to prevent mutual conflicts in the system by sequentially starting each function by function. .

Here, the plurality of backup servers 211 may be grouped with the same function, but even servers having different functions may be classified into the same group and grouped as long as they do not conflict with each other even when they are started at the same time.

In the grouping step (S310), the grouping unit 2111 classifies the plurality of backup servers 211 constituting the backup center 200 by function, so that when the backup center 200 is started, the plurality of backup Since the server 211 is not started individually, but is started collectively among groups classified by function, it is possible to greatly improve the speed of the start while preventing mutual conflict between groups.

Subsequently, in the starting sequence determining step (S320), the startup sequence determining unit 3112 may determine the group startup sequence of the grouped plurality of backup servers 211 .

When the plurality of backup servers 211 are activated for each group, if the startup order for each group is not taken into account, each group classified from the plurality of backup servers 211 is activated through another group activated before and after startup. Interference may cause errors. In other words, the plurality of backup servers 211 are sequentially started for each group, so that the entire system can be built. may cause an error.

In the parallel start-up step (S300), when the plurality of backup servers 211 are started, the start-up order is determined for each group classified so that they can be started normally without causing a mutual collision so that they can be started.

However, when the plurality of groups classified by function from the plurality of backup servers 211 start up, if the start-up time overlaps, this may also lead to an error before the start-up of the backup center 200 .

Accordingly, in the starting time calculation step (S330), the time required for each of the plurality of groups to be activated can be calculated, and based on this, the starting interval of each group is determined in the starting interval tuning step (S340), which will be described later. can be coordinated

In the starting time calculation step ( S330 ), the starting time calculator 2113 may calculate the starting time for each group based on the previously measured starting time for each group. Specifically, the start-up time for each group may be a value pre-measured a plurality of times, and the start-up time calculator 2113 may determine the pre-measured value as the start time for each group.

Here, the startup time for each group may be different depending on the condition of the device at startup or the network connection state, etc., so in order to more reliably prevent errors, a value greater than or equal to the average value of the values measured in advance a plurality of times is the startup time for each group. may be determined, and more preferably, a maximum value among values previously measured a plurality of times may be determined as the startup time for each group.

Subsequently, in the maneuvering interval tuning step ( S340 ), the group-by-group start-up interval may be adjusted so that the group-by-group maneuvers do not overlap each other.

Specifically, in the starting interval tuning step (S340), the starting interval tuning unit 2114 includes each of the plurality of groups classified from the plurality of backup servers 211 in the starting order determined through the starting sequence determining unit 2112. In starting according to the starting order, the starting interval is adjusted based on the starting time for each group calculated through the starting time calculation step (S330) so that the startup does not occur at a time overlapping with other groups that are activated before and after according to the starting order. can

Since the time at which each of the plurality of groups is activated can be calculated through the step of calculating the starting time ( S330 ), it is preferable to set an interval between the plurality of groups longer than the time at which each startup is completed.

Therefore, the starting interval for each group may be adjusted so that the starting interval for each group is equal to or greater than the starting time for each group through the starting interval tuning step (S340), and through this, the plurality of backup servers 211 are configured Since the overlapping of the maneuvers with other groups can be prevented until the start of one group is completed, the mutual collision for each group can be prevented.

Subsequently, in the group-by-group startup step ( S350 ), the plurality of backup servers 211 may be sequentially started for each group according to the group-by-group startup interval through the group-by-group startup unit 2115 .

In the disaster recovery automation method described above, the response response time of the plurality of operation servers 111 constituting the operation server 111 is detected in real time or periodically in the pre-detection step (S100), and the plurality of operation servers By detecting an abnormality in the operation state of (111), a more accurate basis for determining the occurrence of a disaster can be provided, and the synchronization progress state of the operation center 100 and the backup center 200 is checked in the preliminary inspection step (S200). By doing so, preparations for activating the plurality of backup servers 211 constituting the backup center 200 may proceed.

In addition, in the parallel starting step ( S300 ), the operator of the operation center 100 or the backup center 200 based on the detection signal of the operation center 100 abnormality provided from the first disaster recovery automation device 1000 . In accordance with the disaster occurrence determination of A mutual collision between the 211 is prevented, so that the start-up stability is improved, and a disaster recovery automation method capable of more rapid start-up of the backup center 200 may be provided.

The above description is merely illustrative of the technical spirit of the present invention, and various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains.

Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments.

The protection scope of the present invention should be construed by the following claims, and all technical ideas within the equivalent range should be construed as being included in the scope of the present invention.

100: operation center 111: operation server
200: backup center 211: backup server
1000: first automated disaster recovery device 1100: first automated disaster recovery server
1200: change information collection unit 1250: operation server agent
2000: second automatic disaster recovery device 2100: second automatic disaster recovery server
2200: change information processing unit 2250: backup server agent

Claims (23)

In the case of a disaster in the operation center that performs the client's request, in the disaster recovery automation system for converting the processing subject of the request from the operation center to the backup center,
a first automatic disaster recovery device for checking the operating state of the operation center; and
and a second automated disaster recovery device that starts the backup center when an abnormality is detected in the operating state of the operation center by the first automated disaster recovery device;
When the first automated disaster recovery device checks the startup status of a plurality of operation servers constituting the operation center and detects an abnormality in the startup status of the plurality of operation servers, the second automatic disaster recovery device is configured to operate the backup center Classifies a plurality of backup servers included in a plurality of groups by function, sets an order for the groups, and sequentially starts the groups according to the order,
The first disaster recovery automation device,
The plurality of operation servers detects a response reaction time that is a time for responding to the request of the client, and sets the response reaction time to a reference reaction time range that is a range of response reaction times when the plurality of operation servers are normally started; a pre-sensing unit that compares and detects whether or not an abnormality occurs in the operating state of the operation center according to the comparison result; and
and a pre-checking unit for checking the synchronization progress state between the operation center and the backup center when it is detected that there is an abnormality in the operating state of the operation center by the prior detection unit; and
The pre-sensing unit,
a response response time information receiving unit for receiving information on the response response time from the plurality of operation servers;
Criteria for detecting the longest maximum response time of the response time and the shortest minimum response time among the response time, and determining the range of the maximum response time to the minimum response time as the reference reaction time range Reaction time range determining unit;
a response reaction time analyzer analyzing whether the response reaction time is within the reference reaction time range; and
Disaster recovery automation system comprising a; when the response reaction time deviates from the reference reaction time range for a predetermined period of time, a startup state detection unit that detects that an abnormality has occurred in the operation center's startup state. According to claim 1,
The first disaster recovery automation device,
an operation center agent for detecting changes in the plurality of operation servers; and
a change information collection unit for collecting the changes from the operation center agent; includes,
The second automatic disaster recovery device,
a backup center agent for applying the changes to the backup server; and
a change information processing unit that receives the change transmitted from the first automated disaster recovery device and transmits the change to the backup center agent so that the backup server can be applied according to the change; Disaster recovery automation system comprising a. delete According to claim 1,
The automatic disaster recovery system, characterized in that the pre-sensing unit detects that an abnormality has occurred in the operating state of the operation center when the response reaction time is not within the reference reaction time range for a predetermined time. delete According to claim 1,
The reference reaction time range is a disaster, characterized in that the range of the maximum response time to the minimum response time in which the plurality of operation servers that can normally process the request of the client respond to the request of the client recovery automation system. According to claim 1,
The disaster recovery automation system, characterized in that the second automatic disaster recovery device starts the plurality of backup servers after synchronization between the operation center and the backup center is completed. According to claim 1,
The pre-inspection unit further comprises a service range determining unit for determining the range of the service to be performed by the backup center among the services for which the operation center performs the request of the client. According to claim 1,
When an abnormality is detected in the operation state of the plurality of operation servers, the first automatic disaster recovery device transmits an operation center abnormality detection signal to the operator of the operation center or the backup center,
The second automated disaster recovery device is a disaster recovery automation system, characterized in that the activation of the plurality of backup servers according to the disaster occurrence determination of the operator of the operation center or the backup center. According to claim 1,
Disaster recovery automation, characterized in that when the plurality of backup servers are started, the first automated disaster recovery device stops the operation of the plurality of operation servers and blocks a synchronization path between the operation center and the backup center system. According to claim 1,
The second automatic disaster recovery device,
a grouping unit for grouping the plurality of backup servers by function;
a start-up order determining unit for determining a start-up order for each group of the grouped plurality of backup servers;
a start-up time calculation unit for calculating start-up times for each group of the plurality of backup servers;
a start-up interval arranging unit for adjusting the group-by-group start-up intervals of the plurality of backup servers so that the group-by-group start-ups of the plurality of backup servers do not overlap with each other; and
Disaster recovery automation system comprising a; group-by-group activation unit for sequentially starting the plurality of backup servers for each group according to the group-by-group activation interval of the plurality of backup servers. 12. The method of claim 11,
The start-up time for each group of the plurality of backup servers is an average value of values pre-measured a plurality of times for the start-up times for each group of the plurality of backup servers, or a maximum value among values pre-measured a plurality of times. . 12. The method of claim 11,
Disaster recovery automation system, characterized in that the start-up interval arranging unit adjusts the group-by-group start-up intervals of the plurality of backup servers so that the group-by-group start-up intervals of the plurality of backup servers are equal to or greater than the group-by-group start-up times of the plurality of backup servers. . In the case of a disaster in the operation center that performs the client's request, in the disaster recovery automation method for switching the processing subject of the request from the operation center to the backup center,
A reference reaction that detects a response reaction time, which is a time when a plurality of operation servers constituting the operation center responds to a client's request, and sets the response reaction time to a range of response reaction times when the plurality of operation servers are normally started A pre-sensing step of detecting whether an abnormality occurs in the operating state of the operation center by checking whether it is within a time range;
a pre-checking step of checking a synchronization progress state between the operation center and the backup center when it is detected that an abnormality has occurred in the operating state of the operation center; and
A parallel startup step of starting a plurality of backup servers constituting the backup center in parallel;
The parallel starting step classifies the plurality of backup servers into a plurality of groups by function, sets an order for the groups, and sequentially starts the groups according to the order,
The pre-detection step is
a response response time information receiving step of receiving information on the response response time from the plurality of operation servers;
Criteria for detecting the longest maximum response time of the response time and the shortest minimum response time among the response time, and determining the range of the maximum response time to the minimum response time as the reference reaction time range determining the reaction time range;
a response reaction time analysis step of analyzing whether the response reaction time derived from the information on the response reaction time is within the reference reaction time range; and
and an operation state detection step of detecting that an abnormality has occurred in the operation state of the operation center when the response reaction time deviates from the reference reaction time range for a predetermined period of time. delete 15. The method of claim 14,
The reference reaction time range is a disaster, characterized in that the range of the maximum response time to the minimum response time in which the plurality of operation servers that can normally process the request of the client respond to the request of the client How to automate recovery. 15. The method of claim 14,
In the pre-sensing step, when the response response time is not within the reference response time range for a predetermined time, the disaster recovery automation method characterized in that it detects that an abnormality has occurred in the operating state of the operation center. 15. The method of claim 14,
In the parallel starting step, after synchronization between the operation center and the backup center is completed, the disaster recovery automation method, characterized in that the plurality of backup servers are started. 15. The method of claim 14,
The pre-checking step further comprises a service range determination step of determining the range of the service to be performed by the backup center from among the services for which the operation center performs a request from a client. 15. The method of claim 14,
When the plurality of backup servers are activated, the operation of the plurality of operation servers is stopped, and the synchronization path between the operation center and the backup center is blocked. 15. The method of claim 14,
The parallel starting step is
a grouping step of grouping the plurality of backup servers by function;
a start-up order determining step of determining the start-up order for each group of the grouped plurality of backup servers;
a start-up time calculation step of calculating the start-up time for each group of the plurality of backup servers;
a start-up interval tuning step of adjusting the group-by-group start-up intervals of the plurality of backup servers so that the group-by-group start-ups of the plurality of backup servers do not overlap each other; and
Disaster recovery automation method comprising: a group-by-group starting step of sequentially starting the plurality of backup servers for each group according to the group-by-group startup interval of the plurality of backup servers. 22. The method of claim 21,
The startup time for each group of the plurality of backup servers is an average value of values pre-measured a plurality of times for the startup times for each group of the plurality of backup servers, or a maximum value among values pre-measured a plurality of times. . 22. The method of claim 21,
Disaster recovery automation, characterized in that in the step of tuning the startup interval, the startup interval for each group of the plurality of backup servers is adjusted so that the startup interval for each group of the plurality of backup servers is equal to or greater than the startup time for each group of the plurality of backup servers. Way.

Images