KR100763781B1 - Server management system and migration method - Google Patents

Abstract

A server management system and a migration method are provided to minimize the number of backup servers in server operation environment having plural heterogeneous servers, perform rapid server recovery for a work server where errors are generated and perform rapid server provisioning for a server where resources are increased more than a threshold value. An image storage(170) stores images for each work server configured in a work server pool(80). A management server(100), which includes an image management module(130), a resource usage monitoring module(110), an abnormality monitoring module(120), a backup sever full management module(140) and a converting module(150) and a service configuration recovery module(160), manages image storing operations of the image storage and server provisioning and server recovery operations. If a case needing server provisioning or server recovery is generated in one of the work servers, the management server copies images corresponding to the work server to a backup server(90) selected from the image storage to perform the server provisioning or server recovery.

Description

Server Management System and Migration Methods {SERVER MANAGEMENT SYSTEM AND MIGRATION METHOD}

In order to more fully understand the drawings used in the detailed description of the invention, a brief description of each drawing is provided.

1 is a block diagram showing a general connection structure of a server and a database.

2 is a block diagram illustrating an example of a server recovery method in which a homogeneous server is configured in a redundant manner.

3 is a block diagram illustrating an example of a server operating environment having a work server pool and a backup server pool including a plurality of homogeneous servers.

4 is a block diagram illustrating an example of a server operating environment having a work server pool and a backup server pool including a plurality of heterogeneous servers.

5 is a block diagram illustrating an example of a server operating environment having a work server pool composed of a plurality of heterogeneous servers and a minimized backup server pool.

6 is a block diagram showing the configuration of a server management system according to a preferred embodiment of the present invention.

7 is a block diagram showing a general server structure.

8 is a flowchart sequentially illustrating a control process of a management server according to an exemplary embodiment of the present invention.

* Description of the symbols for the main parts of the drawings *

8: network 10: server

20: database 30, 32: switch

40, 80: Job server pool 50, 90: Backup server pool

100: management server 110: resource usage monitoring module

120: abnormal situation monitoring module 130: image management module

140: backup server pool management module 150: heterogeneous server-to-server conversion module

160: service environment recovery module 170: image store

The present invention relates to a server management system and method for efficiently managing a server operating environment consisting of a plurality of servers, specifically, in a server operating environment having a plurality of heterogeneous servers minimizes the number of backup servers and rapid server recovery The present invention relates to a server management system and a method for enabling server recovery and server provisioning.

FIG. 1 is a block diagram illustrating a general connection structure between a server and a database, and FIG. 2 is a block diagram illustrating an example of a server recovery method in which a duplex method of a homogeneous server is configured.

As shown in FIG. 1, a method of reconfiguring the same software environment in a server operating environment including a server 10 connected to a network 8 and a database 20 generally includes an operating system mounted on the server 10. It is common to re-install all the software, from) to the database 20.

As shown in FIG. 2, in the case of a server recovery technology, a real-time server recovery is performed by configuring a homogeneous server 10, 12 and a database 20, 22 that are selectively connected through switches 30 and 32. Is making it possible. However, in order to be able to recover the server by using the redundancy method, the same type of backup server 12, backup database 22, switches 30 and 32 for redundancy and a management solution should be additionally established.

FIG. 3 is a block diagram illustrating an example of a server operating environment having a work server pool composed of a plurality of homogeneous servers and a backup server pool. FIG. 4 is a work server pool composed of a plurality of heterogeneous servers and a backup server pool. A block diagram showing an example of a server operating environment.

As shown in FIG. 3, in order to apply a server recovery technique in a server operating environment having a plurality of homogeneous servers 41 and a job server pool 40 in which a database or an application 42 is operated, the job server pool 40 may be used. As the number of servers 41 increases, a backup server pool 50 in which a plurality of homogeneous backup servers 51, backup databases 52, or applications are installed should be configured according to a management goal.

However, as shown in FIG. 4, the servers 61, 63, and 65 configured in the job server pool 60 are grouped into servers having different specifications due to various reasons such as differences in models or differences in introduction times. It is managed as a result, and the backup servers 71, 73, and 75 for recovery also increase in various server groups. Therefore, in order to enable server recovery or server provision between heterogeneous servers, a migration process between heterogeneous servers is required.

However, the migration process to maintain the same configuration to the heterogeneous server is a preliminary configuration status, reinstallation of the application and database of the operating system and experts in each field to maintain the same configuration of the current operating system, applications and databases of the server The same configuration work was involved. This causes a lot of cost and time.

As described above, the server recovery technology for server failure requires the installation of additional homogeneous servers, switches, and management solutions in order to provide redundancy in the same server operating environment. In addition, in a server operating environment in which a service is performed by a plurality of servers, the size of the backup server pool increases according to server specifications. In general, the configuration of the job server and the backup server is maintained in the same one-to-one configuration, and even in the case of the many-to-one configuration, it involves expensive software and hardware configurations, which causes difficulties in deployment and management.

An object of the present invention is to provide a server management system and method capable of minimizing the number of backup servers in a server operating environment having a plurality of heterogeneous servers.

It is another object of the present invention to provide a server management system and a migration method capable of providing a rapid server recovery for a server having an abnormality in a server operating environment having a plurality of heterogeneous servers and a server supply for a server having a resource increased beyond a threshold. To provide.

Another object of the present invention is to provide a server management system and method that can quickly perform migration between heterogeneous servers when server recovery or server supply in a server operating environment having a plurality of heterogeneous servers.

One aspect of the present invention for achieving the above technical problem is a server management for managing server recovery and server supply in a server operating environment including a work server pool consisting of a plurality of job servers and a backup server pool consisting of a plurality of backup servers System and migration method. The server management system of the present invention comprises: an image store in which images for respective job servers configured in the job server pool are stored; And a management server that manages image storage and server provisioning and server recovery of the image store, wherein the management server selects a suitable backup server from a backup server pool when a server supply or server recovery is required among the working servers, Server supply or server recovery is performed by copying the image corresponding to the job server from the image repository to the selected backup server.

Preferably, the management server: to create the image of the software content of each of the plurality of job servers to store in the image storage, and during the server recovery or server supply to the selected backup server the image of the job server stored in the image storage An image management module for copying; A resource usage monitoring module for triggering a server supply task for a corresponding job server when a job server of which a resource reaches a limit among a plurality of job servers is generated; An abnormal situation monitoring module that triggers a server recovery task for the corresponding job server when a job server among the plurality of job servers, the operation performance of which is lower than a reference value, is generated; A backup server pool management module configured to select one backup server most suitable from the backup server pool according to a system characteristic of the triggered job server when a server supply job or a server recovery job is triggered; If the triggered job server and the selected backup server are heterogeneous servers, a heterogeneous server-to-heterogeneous conversion module that performs migration to change one or more information belonging to the kernel from the image of the job server copied to the backup server to information appropriate for that backup server. ; And a service environment recovery module for recovering an application environment that the corresponding job server has for the selected backup server.

Preferably, the image management module periodically stores an image of a job server in an image repository according to an image management policy that determines an image creation time, an image creation target, and a life cycle of the image, and at least one of the image management policies. Setting values are user selectable.

Another aspect of the present invention relates to a server management method for managing server resiliency and server supply in a server operating environment including a job server pool consisting of a plurality of job servers and a backup server pool consisting of a plurality of backup servers. The server management method of the present invention includes: generating an image for a plurality of job servers and storing the image in an image repository; Monitoring resource usage and server operating performance for a plurality of job servers, and generating triggering for server supply or server recovery when server supply or server recovery is needed; When triggering occurs, selecting one backup server most suitable from the backup server pool according to the system characteristics of the triggered job server; Copying the server image corresponding to the triggered job server from the image store to the selected backup server; When the triggered job server and the selected backup server are heterogeneous servers, performing a conversion operation between heterogeneous servers as a migration job for the selected backup server; And a service environment recovery operation step of recovering an application environment that the job server has with respect to the selected backup server.

Preferably, the step of generating and storing images for the plurality of job servers periodically stores the image of the job server in the image repository according to an image management policy that determines image creation time, image creation target, and life cycle of the image. And one or more setting values belonging to the image management policy are user selectable.

(Example)

Hereinafter, the server management system and the migration method of the present invention will be described in detail with reference to the accompanying drawings. The embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited by the embodiments described below. This embodiment is provided to more completely explain the present invention to those skilled in the art. And detailed description of known functions and configurations that are determined to unnecessarily obscure the subject matter of the present invention is omitted.

5 is a block diagram illustrating an example of a server operating environment having a work server pool composed of a plurality of heterogeneous servers and a minimized backup server pool.

Referring to Figure 5, according to the server management system of the present invention as described below, even if the job server pool 80 is composed of a plurality of heterogeneous servers (81, 83, 85) is configured a backup server pool 90 for this Backup server pool 90 with fewer heterogeneous servers 91 and 93 than the number of servers configured in the job server pool 80 to facilitate server recovery, server provisioning, and migration between heterogeneous servers. It can be done effectively. For example, when the job server pool 80 is composed of a plurality of heterogeneous server families 80a, 80b, 80c each consisting of four pairs of servers 81, 83, 85 and databases 82, 84, 86, respectively. For this purpose, the backup server pool 90 may be composed of only two heterogeneous servers 91 and 93 and a database 92 and 94.

6 is a block diagram showing the configuration of a server management system according to a preferred embodiment of the present invention.

Referring to FIG. 6, a server management system according to a preferred embodiment of the present invention is largely composed of a management server 100 and an image storage 120. Although a detailed description will be described later, the management server 100 stores each image of each of the job servers 81, 83, and 85 configured in the job server pool 80 in the image storage 170. And when the resource usage of a specific server is increased or an abnormal situation is detected among the job servers 81, 83, and 85, an image of the server is copied to the selected backup server to perform server supply or server recovery. do.

In the job server pool 80, various types of servers are operated to provide specific services through the network 8, and server groups 80a, 80b, and 80c, which are composed of heterogeneous servers 81, 83, and 85, are provided. Forming. The backup server pool 90 is a set of backup servers 91 and 93 to replace and operate the corresponding server when a problem occurs in the job server pool 80. The image repository 170 is an image for storing images of software information composed of internal operating systems, applications, data, and the like for respective job servers 81, 83, and 85 configured in the job server pool 80. It is a repository (or database).

The management server 100 includes the resource usage monitoring module 110 and the abnormal situation monitoring module 120 for the job server pool 80, the image management module 130 for the image storage 170, and the backup server pool 90. Backup server pool management module 140, heterogeneous server-to-server conversion module 150 and service environment recovery module 160 for.

The image management module 130 generates an image of the software contents of the job servers 81, 83, and 85 configured in the job server pool 80 as an image at a predetermined time and data range according to an image management policy. Store in The image management module 130 classifies cases having different configurations based on the combination information of the respective operating systems and applications installed in the job servers 81, 83, and 85, and then, for each of the other configurations, The installation state is copied and compressed as it is to be stored in the designated area of the image storage 170.

The image management policy determines the image creation time, the image creation target, and the life cycle of the image, and is determined according to the purpose of the job server. In addition, the user can specify the setting value to cope with various situations. For example, if the job server frequently updates real-time information and maintains the latest information, specify a short image storage cycle. In addition, when management of the system change history of the corresponding job server is required, the previous image is designated to be preserved for some time without deleting. The method of generating an image to be stored in the image store 170 may use a conventional commercial imaging tool or an open source method as it is.

The image management module 170 copies the image of the corresponding job server stored in the image repository 170 to the selected backup server at the time of server recovery or server supply. The process of copying an image to a backup server leverages existing imaging techniques and utilizes technologies such as Wake on LAN and Fast Copy for management efficiency and rapid copying.

The resource usage monitoring module 110 and the abnormal situation monitoring module 120 are modules for monitoring the job server pool 80. There are two management indicators of the job servers 81, 83, and 85, one for resource utilization and the other for resource failure. The resource utilization rate includes the processor, memory, network, disk utilization rate, etc. of the job servers 81, 83, and 85. The resource usage monitoring module 150 triggers a server supply task when the resource of the task server reaches a threshold. A resource failure is when a server becomes inoperable due to an application or hardware problem. The abnormal situation monitoring module 120 considers the abnormal situation when the operation performance of the job server is less than the reference value and triggers a server recovery task.

When the server supply job or the server recovery job is triggered, the backup server pool management module 140 selects one backup server most suitable from the backup server pool 90 according to the system characteristics of the triggered job server. For example, one or more criteria are selected from various system components such as processor performance, memory size, and internal device characteristics to select the most suitable backup server.

The heterogeneous server-to-server conversion module 150 performs a migration operation of changing some information from the image of the job server copied to the backup server to information suitable for the corresponding backup server. When you copy a job server image to a backup server, the backup server will not work properly unless both system configurations are exactly the same. For example, when the network card of the job server and the backup server to which the image is copied is different from each other, the network function is modified by modifying device information of the network card among the copied job server images. This task applies to all hardware devices in the backup server. As such, the heterogeneous server-to-server conversion module 190 is involved in the normal operation of all hardware devices of the backup server.

More specifically, as shown in FIG. 7, the internal structure of a general server is largely composed of hardware 200, kernel 300, and application program 400. The hardware 200 includes, for example, a CPU, a RAM, a disk and a CD, a console, a serial port, a special board, a network interface, and the like, and various types of devices are used by vendors. The kernel 300 is largely composed of a kernel unit 310, a function implementation unit 320, a software support unit 330, and a hardware control unit 340. Among them, the kernel 300 is directly connected to the hardware 200 by the software support unit 330. And a hardware control unit 340. The kernel unit 310 is software for process management, memory management, file system, device control, network, etc., which is independent of the hardware 200, and receives various system calls from user programs and applications. You will receive the desired hardware service. The function implementation unit 320 is a part for implementing functions such as simultaneous multitasking, virtual memory, files and directories, TTY and device control, and connection. The software support unit 330 and the hardware control unit 340 include components such as architecture dependent code, memory manager, file system manager, block device, character device manager, network subsystem, interface driver, and the like.

When the image of a specific job server is copied to the selected backup server by the image management module 130, only the hardware 200 portion is changed to the hardware of the backup server, and the remaining applications 400 and the kernel 300 are configured as the specific job server. It remains intact. However, in this state, even if the application program 400 and the kernel 300 of the backup server request the operation of the hardware from the software support unit 330 and the hardware control unit 340 for the operation of the server, the hardware 200 changes the request method. Will not respond to. Therefore, for migration between heterogeneous servers, a software support unit and a hardware control unit suitable for the hardware components of the server are secured for each vendor. The heterogeneous server-to-server conversion module 150 finds a portion to be changed in the kernel 300 and replaces existing information to complete the migration.

The service environment recovery module 160 restores the application environment originally owned by the job server. For example, this may include an external environment such as setting a system network address other than job server image information, restoring data that could not be stored in an image of a job server, and changing an application configuration value by an external environment.

The following is a management server 100 for quickly responding to server recovery and server supply by using a backup server pool minimized in a server operating environment composed of a plurality of heterogeneous servers by the server management system having the above configuration. The control process of) will be explained.

8 is a flowchart sequentially illustrating a control process of a management server according to an exemplary embodiment of the present invention.

Referring to FIG. 8, in step S100, the image management module 130 of the management server 100 generates an image of each of the job servers 81, 83, and 85 of the job server pool 80 according to the image management policy. To store in the image store 170.

In step S200, resource usage management module 110 and the abnormal situation management module 120 monitors the work status for each job server 81, 83, 85 in the job server pool. In detail, the resource usage monitoring module 150 monitors resource usage of each of the job servers 81, 83, and 85 of the job server pool 80 in steps S210 and S220, and determines whether a server supply is necessary. If it is determined that the server supply is necessary, the resource usage monitoring module 150 proceeds to step S230 to trigger a server supply operation. In addition, the abnormal situation monitoring module 120 monitors server operation performance of each of the job servers 81, 83, and 85 of the job server pool 80 in steps S212 and S222, and determines whether server recovery is necessary. If it is determined that the server recovery is necessary, the abnormal situation monitoring module 120 proceeds to step S232 to trigger a server recovery operation.

When triggering for server supply or server recovery occurs, the backup server pool management module 180 responds to selecting one backup server most suitable for the backup server pool 90 according to the system characteristics of the job server triggered in step S300. Select. In operation S310, the image management module 130 copies the server image corresponding to the triggered job server to the backup server selected from the image storage 170.

Subsequently, the heterogeneous server conversion module 150 determines whether the triggered job server and the selected backup server are mutually heterogeneous servers in operation S400. In the case of a heterogeneous server, the process proceeds to step S410 to perform the conversion between heterogeneous servers as a migration operation. In the final step of server supply or server recovery, in step S420, the service environment recovery module 160 restores the external environment such as the system network address setting other than the job server image information and the data that could not be stored in the image of the job server. Restores the application environment that the original job server had, such as changes in application configuration values by the environment. This completes the server supply or server recovery by the backup server.

Embodiments of the server management system and method of the present invention described above are merely exemplary, and those skilled in the art to which the present invention pertains can make various modifications and equivalent other embodiments. You can see well. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

According to the server management system and method of the present invention as described above, the number of backup servers can be minimized in a server operating environment having a plurality of heterogeneous servers, and the server recovery and resources for a job server having an abnormality This enables rapid server provision for servers that have grown beyond the threshold. In server recovery or server provisioning, migration between heterogeneous servers can be accomplished quickly. The present invention can achieve both tasks of server supply and server recovery through the operation of a minimum backup server pool based on a migration technology that allows an operating system and an application running on an existing server to operate on a server having a different specification. . Server administrators can perform server supply and server recovery with cost and management issues incurred in the operation of multiple backup servers, the use of available server resources, and rational procedures beyond the simple countermeasures of backup software.

Claims (7)

In a server management system for managing server recovery and server provisioning in a server operating environment that includes a job server pool consisting of multiple job servers and a backup server pool consisting of multiple backup servers: An image repository in which images for respective job servers configured in the job server pool are stored; And Includes a management server that manages image storage and server provisioning and server recovery of image repositories, The management server selects a suitable backup server from the backup server pool when a server supply or server recovery is required among the job servers, and copies the image corresponding to the job server from the image repository to the selected backup server to supply the server or server. A server management system, characterized in that to perform a recovery. The method of claim 1, The management server is: An image management module for generating a software content of each of the plurality of job servers as an image and storing it in an image repository, and copying an image of the corresponding job server stored in the image repository to a selected backup server at the time of server recovery or server supply; A resource usage monitoring module for triggering a server supply task for a corresponding job server when a job server of which a resource reaches a limit among a plurality of job servers is generated; An abnormal situation monitoring module that triggers a server recovery task for the corresponding job server when a job server among the plurality of job servers, the operation performance of which is lower than a reference value, is generated; A backup server pool management module configured to select one backup server most suitable from the backup server pool according to a system characteristic of the triggered job server when a server supply job or a server recovery job is triggered; If the triggered job server and the selected backup server are heterogeneous servers, a heterogeneous server-to-heterogeneous conversion module that performs migration to change one or more information belonging to the kernel from the image of the job server copied to the backup server to information appropriate for that backup server. ; And And a service environment recovery module for restoring the application environment of the job server with respect to the selected backup server. The method of claim 2, And the image management module periodically stores an image of a job server in an image repository according to an image management policy that determines an image creation time, an image creation target, and an image life cycle. The method of claim 3, And one or more setting values belonging to the image management policy are user selectable. In a server management method for managing server recovery and server provisioning in a server operating environment that includes a job server pool consisting of multiple job servers and a backup server pool consisting of multiple backup servers: Generating and storing images for a plurality of job servers in an image repository; Monitoring resource usage and server operating performance for a plurality of job servers, and generating triggering for server supply or server recovery when server supply or server recovery is needed; When triggering occurs, selecting one backup server most suitable from the backup server pool according to the system characteristics of the triggered job server; Copying the server image corresponding to the triggered job server from the image store to the selected backup server; When the triggered job server and the selected backup server are heterogeneous servers, performing a conversion operation between heterogeneous servers as a migration job for the selected backup server; And And a service environment recovery operation step of recovering the application environment that the job server has with respect to the selected backup server. The method of claim 5, Generating and storing images for the plurality of job servers periodically stores images of the job server in an image repository according to an image management policy that determines an image creation time, an image creation target, and a life cycle of the image. Featuring server management methods. The method of claim 6, The one or more setting values belonging to the image management policy are user selectable.

Images