JP2017182380A - Data management system and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data management system and method with which it is possible to flexibly manage data by a local and/or cloud server and make it continuously and optimally usable even at the time of a server or network trouble.SOLUTION: When a cloud management server, backup server and virtual server and a local main server and local terminal are connected by a network, the data of the local main server is backed up to the cloud backup server and a trouble occurs to the local main server, the data of the local main server is restored to the cloud virtual server using the backup data of the cloud backup server, the cloud virtual server is continued of system use as a temporary main server, and an access method for access from the local terminal to the cloud virtual server is switched, after restoration, to a file transfer method in place of a remote connection method at normal operation time.SELECTED DRAWING: Figure 1-1

Description

  The present invention relates to a data management system and method, and more particularly to a data management system and method in which application data is flexibly managed by a local and / or cloud server and can be continuously used even in the event of a server or network failure. About.

  Conventionally, for example, data of a local file server is backed up and restored to a replacement or repaired server when the data is damaged. For example, Patent Document 1 describes storing backup data via a network in a backup server of a virtual machine on the cloud. In Patent Document 2, when a local file server cannot be repaired or there is no replacement server with the same hardware configuration, virtual technology is used in another local real machine or a virtual machine on the cloud. It has been proposed to start a virtual server and restore system backup data to continue system operation.

  However, when the virtual server is built on a local real machine, the performance after recovery depends on the performance of the real machine, so high performance requires a high-spec real machine, which is expensive. (Patent Document 3). Also, when the virtual server is built on a cloud virtual machine, the performance depends on the network status, so the response time of processing at the local terminal becomes long, which may hinder business.

  When using virtual server data of virtual machines in the cloud, in order to improve the problem of performance depending on the network status, install a cache server to speed up processing, or connect a slave server locally. In some cases, it is installed to reinforce processing capacity (for example, Patent Documents 4 and 5).

  However, if a cache server that stores a copy of the virtual server data on the cloud is installed locally, the virtual server may fail (server down), or a network problem may occur and the connection to the virtual server may be interrupted. The cache function could not be maintained, and the operation on the local real terminal could not be continued until the failure of the virtual server or network was recovered. Also, when installing a slave server, the performance depends on the specifications of the slave server, and it is expensive to prepare a slave server with a high specification.

Japanese Patent No. 3768775 Patent No. 3022768 JP-A-4-141744 Japanese Patent No. 2701806 JP 62-172347 A

  The problem to be solved by the present invention is that in a system in which data of a local real server (primary server) is backed up to a virtual server in the cloud, operation can be continued even if a failure occurs in the primary server. Is to minimize performance dependence on network conditions.

  A further problem of the present invention is that in a system having a virtual server (main server) on the cloud and a local real server (small hardware cache server), a failure occurs in the main server or network on the cloud. Even if the cache server is replaced with a temporary primary server, the operation can be continued, and when the primary server or network on the cloud is restored, the updated data is reflected on the primary server so that it can be restored to the original operating state. Is to do.

  Further problems of the present invention are the first operation mode using the virtual server (backup server) on the cloud and the local main server, and the virtual server (main server) on the cloud and the local cache server. It is to provide a system that switches between the second operation mode according to conditions such as the number of connected users and an operation policy.

  A further problem of the present invention is that in a system including a virtual server of a cloud virtual machine and a real server of a local real machine, one of the virtual server and the real server is operated as a master server and the other as a slave server. The slave server is operated as a cache server or backup server according to the load and performance of the actual machine.

In order to solve the above problems, a data management system according to a first operation mode of the present invention includes a cloud management server that holds management information that defines the role of each server, a cloud backup server, and a cloud virtual server. A virtual machine for a local machine, a real server of a local real machine, and one or more local terminals are connected via a network. During normal operation, management information of the management server mainly includes a real server of the local real machine. A data management system defined to operate as a server,
First backup means for backing up the data of the local primary server to the backup server of the cloud;
When a failure occurs in a local primary server, a virtual server is generated from the cloud virtual server virtual machine in the management information of the management server, and the virtual server of the virtual machine is operated as a primary server. Write the information and use the backup data of the cloud backup server to restore the local primary server data to the virtual server of the cloud virtual server virtual machine, temporarily store the virtual machine virtual server As a primary server, a restoration method in case of failure to continue system use,
Access change means for switching the access method when accessing the virtual server of the virtual machine for the virtual server of the cloud from the one or more local terminals after the restoration to the file transfer method instead of the remote connection method during the normal operation When,
And when the failure of the local main server occurs, system use is continued using a cloud backup server.

The data management system of the second operation mode of the present invention includes a virtual server of a cloud virtual machine, a cloud management server that holds management information that defines the role of each server, a real server of a local real machine, One or more local terminals are connected via a network, and during normal operation, the management information of the management server uses the virtual server of the cloud virtual machine as a main server and the real server of the local real machine as a cache server A data management system that is prescribed to operate as
The local cache server is
Cache means for caching the data of the main server of the cloud and the management information of the management server in the cache server to be used by the one or more local terminals;
When a communication failure of the network or a failure of the main server of the virtual machine occurs, the management information of the management server and the cloud information are stored on the local cache server based on the data of the local cache server. Restore at least a part of the principal data of the virtual server and handle the cache data of the principal that cannot be directly manipulated as it is as a partial copy of the principal data that can be manipulated directly. A recovery means for rewriting the management information to operate the cache server as a temporary main server;
The system usage is continued when a communication failure or a failure of the main server in the cloud virtual machine occurs.

In a further embodiment of the present invention, a management server that holds management information defining the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are provided. Connected by network,
In the management information of the management server, the first operation mode and the second operation mode are registered as operation modes,
Based on a condition including the number of local terminals connected to the main server and / or a pre-defined operation policy, the contents of the management information are mutually exchanged between the first operation mode and the second operation mode. Operation mode switching means for switching is provided.

In a further embodiment of the present invention, a management server that holds management information defining the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are provided. Connected by network,
The management information of the management server is usually operated in a two-server configuration in which one of the virtual server of the cloud virtual machine and the real server of the local real machine is a master server and the other is a slave server. The operation mode and the second operation mode are registered,
An operation mode switching means for switching from the normal operation mode to the second operation mode based on environmental conditions including a load on the local terminal and a communication speed of the network is provided.

  In the first operation mode of the data management system according to the present invention, in a system including a local real server (main server), one or more local terminals, and a cloud backup server, the main server is regularly assigned to the backup server. If the primary server fails, the backup data is restored to the cloud virtual server, and the system operation can be continued by switching the virtual server to the primary server. When the main server is local, access is made through a LAN, and when the main server is in the cloud, switching to the file transfer method (Web connection) enables smooth business processing even when the communication speed is low.

  In the second operation mode of the data management system according to the present invention, in a system including a cloud virtual server (main server), a local real server (cache server), and one or more local terminals, The cache server accumulates the data cache every time the main server is accessed from a terminal, and the system is reconfigured so that the cache server can be temporarily used as the primary server if the primary server or the network fails. Therefore, when the primary server or network of the cloud is restored, the updated data during the failure is reflected in the primary server of the cloud, so that the business can be continued without any trouble even after the failure recovery.

  According to the present invention, the first operation mode and the second operation mode are switched to each other based on conditions such as the number of local terminals connected to the main server and / or a predetermined operation policy. As a result, an optimum data management method with redundancy according to the implementation environment becomes possible.

  According to the present invention, a normal operation mode in which one of a cloud virtual server and a local real server is operated as a master server and the other as a slave server, and the second operation mode described above are divided into a local terminal and a network. By switching according to the environmental conditions including the load, an optimum data management method with redundancy according to the implementation environment becomes possible.

1-1 is a figure which shows the whole structure of the data management system concerning this invention. FIG. 1-2-1 is a diagram illustrating the relationship among the elements during normal operation in the first operation mode of the present invention. FIG. 1-2-2 is a diagram illustrating a relationship between elements when a failure occurs in the file server 220 in the first operation mode. FIG. 1-2-3 is a diagram illustrating the relationship among the elements during normal operation according to the second operation mode of the present invention. FIG. 1-2-4 is a diagram illustrating a relationship among elements when a failure occurs in the virtual server 130 or the network connection in the second operation mode. FIG. 2A is a schematic block diagram of the management server 110 shown in FIG. FIG. 2-2 is a schematic block diagram of the application server 120 shown in FIG. FIG. 2-3 is a schematic block diagram of the file server 220 shown in FIG. 2-4 is a schematic block diagram of the backup server 150 shown in FIG. 2-5 is a schematic block diagram of the cache server 230 shown in FIG. 2-6 is a schematic block diagram of terminal PC 210 shown in FIG. 2-7 is a schematic block diagram of the virtual server 130 shown in FIG. FIG. 3A and FIG. 3B are flowcharts for explaining the processing flow from the terminal 210 side during normal operation of the file server 220 and at the time of failure countermeasure in the first operation mode. FIG. 3B is a flowchart for explaining processing on the management server 110 side in the switching processing when a failure occurs in the file server 220 of the local 200 in the first operation mode. FIG. 3C is a flowchart for explaining the switching process at the time of failure recovery of the file server 230 of the local 200. FIG. 3-4-1 and FIG. 3-4-2 are flowcharts for explaining the processing flow from the terminal 210 side during normal operation of the cache server 230 and countermeasures against a network failure in the second operation mode. FIG. 3-5 is a diagram for explaining processing of the cache server 230 when a failure occurs in the virtual server 130 or the network in the second operation mode. FIG. 3-6-1 is a flowchart for explaining the switching process at the time of failure recovery in the second operation mode from the cache server 230 side. FIG. 3-6-2 is a flowchart for explaining the switching process at the time of failure recovery in the second operation mode from the management server 110 side of the cloud 100. FIG. 3-7 is a flowchart for explaining the operation of the first modified example of the present invention. FIG. 3-8 is a flowchart for explaining the operation of the second modified example of the present invention. FIG. 4A is a diagram illustrating an example of management information held by the management server 110 of the cloud 100. FIG. 4B is a diagram illustrating an example of the license management table 410 managed by the management server 110. FIG. 4C is a diagram illustrating an example of management information of the management server 110 during normal operation in the first operation mode. FIG. 4-4 is a diagram illustrating an example of management information of the management server 110 during normal operation in the second operation mode. In the 2nd modification of this application, it is a figure which shows the example of the management information of the management server 110 at the time of normal operation of the system configuration | structure shown to FIGS. 1-2. In the 2nd modification of this application, it is a figure which shows the example of the management information of the management server 110 when it becomes the system configuration | structure shown to FIGS. 1-2-3. FIG. 5A is a diagram illustrating an example of a management menu screen in the embodiment of the present invention. FIG. 5-2-1 is a diagram illustrating an example of a data management screen in the first operation mode. FIG. 5-2-2 is a diagram illustrating an example of a data management screen in the second operation mode. FIG. 5C is a diagram illustrating an example of a confirmation screen for restoring the principal data. FIG. 6 is an example showing a list of resources managed by the management server.

  DESCRIPTION OF EMBODIMENTS Embodiments for carrying out the present invention will be described in detail below with reference to the accompanying drawings. First, terms used in this specification will be explained, and then the present invention will be described using system configuration, system operation, various tables, screen examples, and the like.

(1) Explanation of terms “App”: An application program used by a contractor (accounting firm, consultant, etc.) who receives services in business processing.
“Data”: Business information input and output by an application. The data (file) stored in the target device may be expressed as XX data in order to distinguish any of principal data, copy data, cache data, and backup data.
“Principal data”: Data master.
"Updated data": Difference data when changes such as addition or deletion of data occur after reading of the principal data, but only the difference data added to the principal data or only the difference data Either may be sufficient.
“Copy data”: Data obtained by copying principal data.
“App Server”: A server that provides an appropriate version of an application required by a service contractor.
“Management Server”: A server that integrally manages a mechanism such as contractor information, billing information, application provision, principal data, copy data, management information management, and sharing. Although it is described assuming that it is mainly installed on the cloud side, it may be installed on the local side. In addition to a real server or virtual server of a real machine, a virtual server of a virtual machine may be used.
“Cloud (IDC)”: The management server, application server, backup server, cache server, data server, etc. necessary for the service provider to install physical servers, recording devices, communication devices, etc. A facility that stores and maintains data (Internet data center). One site may be configured, and a number of sites may be configured in a distributed, multiplexed, or virtual manner. Further, it is not always necessary to be a base, and it may be configured by a real server, or the center itself may be a virtual server.
“Local (office)”: A base such as an accounting office where a service contractor uses a service.
“Real machine”: A computer equipped with physical devices such as input, recording, and control.
“Virtual”: A pseudo computer built on a real server using software technology.
“Virtual machine”: A virtual computer that provides server functions. It is also possible to operate a virtual computer such as a virtual server on the virtual machine.
“Single device”: A computer that activates applications and performs data input / output operations. It is a computer that operates alone, and has a function of holding a principal, a copy, update data, and the like in a recording device inside the computer.
“Terminal”: A terminal that starts an application and performs data input / output operations.
“Terminal”: A computer that starts applications and inputs / outputs data, and uses a function to store update data in a storage device as necessary. Although it is described assuming that it is mainly installed on the local side, it may be installed on the cloud side, may be a real machine (terminal PC), or may be a virtual terminal.
“File”: A set of data that is business information input and output by an application.
“Cache”: When using data registered in the virtual server of the cloud virtual machine, temporarily store the data to be used in the storage device of the local cache server or the data, and to the same data Access speed is increased by accessing the data at the time of access. The data stored here can cache larger data such as the entire DB table and the entire file to be used, and realizes high-speed access to the data more efficiently. The difference between cache access and file transfer is that the terminal application is clearly aware of the location of the data to be updated. At the time of cache access, the terminal application is not aware of access to the cache and recognizes that it is accessing the storage device of the server with the principal data (actually, it accesses the cache data in the relaying cache server). ing). On the other hand, at the time of file transfer, the data used by the terminal application differs in that the principal data in the cloud server is once transferred to a local storage device, and the terminal application accesses the local device.
“Backup”: Refers to copying the contents of a server storage device in the event of a failure and storing it in the storage device of a cloud backup server. The target of backup data includes principal data, update data, management information, used programs, application information, and the like.
"Restore": When a failure occurs, restore the backup data from the backup server with the information necessary to have another server have the same role on behalf of the failed device (server, terminal, etc.) To build an available state.
"Recovery": "Restore" is a process to switch data handling within the same server while transferring data and restoring the environment after starting another server. It is called “recovery” to distinguish it from “restore” that restores data and environment. In the first operation mode, in the case of recovery processing in the cache server in the event of a network failure, recovery processing is performed on partial copy data of the principal, but the recovery processing is excluded on the entire principal. Not the purpose.
“Direct processing”: Directly opening and updating a data file. High-speed processing is not possible without an internal disk and local file server.
“File transfer”: Transfer (copy) of data to be handled between computers (terminals) in units of files to be handled.
“Direct processing method”: A method of directly opening and updating a file of principal data or copy data. This method is slow when the principal data is in the cloud and the terminal is remote such as local.
“File transfer method”: A method for inputting and outputting data without depending on the communication environment. When there is principal data in the cloud virtual server, the principal data is temporarily copied to a local storage device. High-speed processing is realized by directly manipulating the data.
“File server”: A server that serves as a management device for storing principal data or copy data. Although principal data is mainly stored and managed, copy data is stored and managed when it functions as a temporary file server when a failure occurs.
“Primary server”: A file server managed as a data principal location among file servers. For this reason, a temporary file server at the time of failure may be read as a temporary main server.
“Cache server”: a server having a role as a management device for managing cache data.
“Backup server”: A server having a role as a management device for managing backup data. Although it is described assuming that it is mainly installed on the cloud side, it may be installed on the local side. In addition to a real server or virtual server of a real machine, a virtual server of a virtual machine may be used.
“Management Menu”: Graphical user for service contractors to select services that can be used according to the contract type, and to set the data storage location such as principal and copy, and the type of terminal to be operated・ An interface.
“Management Information”: Information managed by the management server, “Management Category”, “Real Machine / Virtual”, “Name”, “Data”, “Role”, “Enable / Disable”, “Management” Information such as “location”, “status”, “connection source”, “presence / absence of charge” for managing the presence / absence of charge, etc.
“Account”: Managed by a user management table (not shown), and includes a user name, password, loginable terminal, access authority, etc., and is used in combination with the license management table (FIG. 4-2). .

(2) Description of System Configuration and Simple Operation Next, the system configuration of the present invention will be described. 1-1 is a figure which shows the whole structure of the data management system concerning this invention. As shown in FIG. 1, a local (office) system 200 such as an accounting office computer system is connected to a cloud (IDC) 100. In the local 200, one or more terminal PCs (1 to N) 210, a file server 220, and a cache server 230 are LAN-connected and connected to the Internet via a firewall FW. In the cloud 100, a management server 110, an application server 120, a virtual machine 140 on which a virtual server 130 is mounted, and a backup server 150 are LAN-connected and connected to the Internet via a firewall FW.

  Here, it is assumed that the cloud-side virtual machine 140 includes not only a case where it is already activated as a virtual machine, but also a case where it is not activated, waiting for activation, or not yet prepared for activation. For this reason, in addition to existing virtual machine virtual servers, the cloud virtual machine includes a stage that starts from the start of the virtual machine and before it is configured as a virtual server. You may make it connect to. As described above, the components of the cloud 100 may be scattered in a plurality of locations and interconnected via the Internet. Further, as will be apparent from the following description, the present invention does not always require all of these components, and selectively uses some or all of these components depending on the embodiment and / or the implementation environment. Is.

  FIG. 1-2-1 is a diagram illustrating a relationship among elements during normal operation in the first operation mode of the present invention. In the first operation mode, the data principal used by each local terminal 210 is stored in the local file server 220, and the terminal 210 can directly operate a desired file via the LAN. Data of the file server 220 is backed up to the backup server 150 of the cloud 100 at any time or periodically. Information such as the role of each server, the validity / invalidity at that time, and the connection destination is defined in the management information of the management server 110. Each server or terminal 210 refers to the management information of the management server 110 as necessary, and Acquire information such as the access destination and backup data destination. Note that the cloud backup server, management server, and local file server may be virtual servers as well as real servers. In addition, the management server and the backup server are described assuming that they are mainly installed on the cloud side, but may be installed on the local side.

  FIG. 1-2-2 is a diagram illustrating a relationship among elements when a failure occurs in the file server 220 in the first operation mode. When a failure occurs in the file server 220 during operation, as will be described in detail later, the management information of the management server 110 is changed, the virtual server 130 is prepared in the virtual machine of the cloud 100, and this virtual server is transferred from the backup server 150. The backup data is restored to 130. The terminal 210 subsequently accesses the file of the virtual server 130 according to the management information of the management server 110, and the business is continued. By this switching processing, the operator of the terminal 210 can continue the business without being aware of the failure of the file server 220 and the switching of the access destination to the virtual server 130 on the cloud. Furthermore, as the connection destination is switched from the local 200 file server 220 to the virtual server 130 of the cloud 100, the access method is changed from the normal connection method to the file transfer method that is less affected by the network environment, resulting in a response delay. Do not. Note that the cloud backup server, management server, and local file server may be virtual servers as well as real servers. In addition, the management server and the backup server are described assuming that they are mainly installed on the cloud side, but may be installed on the local side.

  FIG. 1-2-3 is a diagram illustrating the relationship between the elements during normal operation in the second operation mode of the present invention. In the second operation mode, the principal data used in the terminal 210 of the local 200 is stored in the virtual server 130 configured in the virtual machine 140 of the cloud 100 and further backed up to the backup server 150 of the cloud 100. A cache server 230 is provided in the local 200, and a copy of data accessed from the terminal 210 is stored (cached) therein. Again, the file transfer method is used as the access method, and the principal data is read into the cache server 230 at the time of access, and the cache data is directly operated from the terminal 210. Even in this case, it appears to the operator of the terminal 210 that the principal data of the virtual server 130 is being operated. The management information of the management server 110 is also referred to first and fetched as cache data.

  The cloud backup server, the management server, and the local cache server may be virtual servers as well as real servers. In addition, the management server and the backup server are described assuming that they are mainly installed on the cloud side, but may be installed on the local side. The cache server is described assuming that it is mainly installed on the local side, but may be installed on the cloud side. According to the present system, the cache server as a local real server only needs to have a cache function during normal operation, and on the other hand, in the event of a failure, it only functions as a temporary file server. Therefore, there is a merit that a relatively small-scale hardware configuration is sufficient because the processing speed may be somewhat reduced.

  FIG. 1-2-4 is a diagram illustrating a relationship among elements when a failure occurs in the virtual server 130 or the network connection in the second operation mode. If a failure occurs in the virtual server 130 or the network connection, the principal data of the virtual server 130 cannot be accessed. The cache server 230 creates a copy of the management information and file data (cache data) cached at that time, and changes the access destination information of the management information file to the cache server. This is notified to the terminal 210, the access path is changed, and the cache server can serve as a temporary file server during the occurrence of a failure, and the business can be continued.

  As described above, cache data of a cache server that cannot be directly manipulated is handled as partial copy data of the principal that can be directly manipulated, so that the cache server functions as a temporary file server. The recovery process for making it happen has been described. Here, the “restoration means” of the first operation mode is called “restoration” because it performs data transfer and environment recovery after starting another server, etc., whereas in the second operation mode, This is a process that changes the handling of cache data in the same cache server to partial copy data of the principal, and is called “recovery” because it is different from the “restore means” in the first operation mode that restores the data and environment to another server. Will be distinguished. In the first operation mode, in the case of recovery processing in the cache server in the event of a network failure, recovery processing is performed for partial copy data of the principal, but when the virtual server failure occurs, the backup server data is It is not intended to exclude that recovery processing is performed on the entire principal, such as when performing recovery processing by using it.

  When the failure is recovered, the update data is reflected on the virtual server 130 via the backup server 150 and then the original role as the cache server is restored. The cloud backup server, the management server, and the local cache server may be virtual servers as well as real servers. The master server, the management server, and the backup server are described assuming that they are mainly installed on the cloud side, but may be installed on the local side. Although it is described on the assumption that the slave server is mainly installed on the local side, it may be installed on the cloud side. Further, according to the present system, there is an advantage that the machine can be efficiently used by using the obsolete slave server as a cache server.

  2-1 to 2-7 are schematic block diagrams of the management server 110, application server 120, file server 220, backup server 150, cache server 230, terminal PC 210, and virtual server 130 shown in FIG. Management information, license information, user information, and the like are stored in the storage unit 112 of the management server 110 in FIG. The application server 120 in FIG. 2-2 stores application data and application versions. The storage unit 232 of the cache server 230 in FIG. 2-5 stores management information, license information, user information, and the like that the management server 110 has in addition to the cache data. Further, as shown in FIG. 1-2-4 A copy data storage unit is provided to replace the file server in the event of a failure. The local terminal 210 in FIG. 2-6 stores processing data when functioning as a single machine and a management server access path indicating a connection destination to the management server, and a keyboard for the operator to operate the file. Etc., and an output unit 215 to the display.

(3) Description of Management Information and Table FIG. 4A is a diagram illustrating an example of management information held by the management server 110 of the cloud 100. As shown in the figure, the management information lists various servers and terminals that can be incorporated into or used in the system and their contents. The information included in the management information is as follows.
“Number”: The item number of the management number, which is automatically assigned by the system.
“Management category”: This is an object to be managed by the management server. In this specification, the types of a single device (device that handles operations and data), a server (device that handles data), and a terminal (device that handles operations) are classified. Including. Of these, the stand-alone device and the server also serve as a management device for storing principal data or copy data. This is a device category, which includes a single machine, server, application server, cache server, backup server, slave server, and terminal.
“Real machine / virtual”: Used to distinguish whether the device is in a virtual environment or on a real machine.
“Name”: a name assigned for identifying a server, a PC, or the like.
“Data”: Refers to the type of data managed by the management apparatus, and distinguishes among principal data, copy data, cache data, and backup data.
“Role”: A function that can be executed by a server such as a file SV (server), an application SV, a backup SV, a cache SV, a management SV, a single machine, a terminal, or a device such as a terminal.
“Valid / Invalid”: Settings when using the role. Valid when using, invalid when canceling use. Also, valid if available, invalid. Invalid display. Enable / Disable can be set / displayed in toggle format.
“Latest data”: The original data, backup data, copy data, etc. of each server or terminal is the latest (updated), or the failure occurred, the server was stopped, or the data was not updated. Indicates whether this data is different from the latest one (not updated).
“Management location”: indicates a location where a device such as a server or a terminal physically exists, the cloud indicates an IDC, and the local indicates an office other than an IDC including an office such as an accounting office or an outside office such as a consultant Indicates the location.
“Status”: Indicates the status of a device such as communication failure due to operation, stoppage, failure, network failure, etc., and the status of the role of the device when a plurality of roles exist in one device.
“Connection source”: Indicates the name of the mobile device (smart phone, tablet, etc.) logged in to the virtual PC when using the virtual PC.

  FIG. 4B is a diagram illustrating an example of the license management table 410 managed by the management server 110. For example, when the provider of the service of the present invention provides a plurality of tax accounting applications to a plurality of accounting firms, companies, individuals, etc., the types of applications and licenses contracted for each company or office It is a table for managing the number and the number of users online. The license management table 410 registers information such as the user name, the name of the contracted application, the total number of licenses granted, and the number of licenses currently used. The license is registered in this table when the usage contract is concluded. For example, when the application is accessed from the terminal 210 of the accounting office and the management server 110 is accessed, the number of used licenses is grasped.

(4) Description of First Operation Mode The first operation mode shown in FIGS. 1-2-1 and 1-2-2 will be described in detail with reference to flowcharts, contents of a management information table, and screen examples. FIG. 3-1 (3-1-1, 3-1-2) is a flowchart for explaining the processing flow from the terminal 210 side when the file server 220 is in normal operation and at the time of failure countermeasure in the first operation mode. . For example, in an accounting office system, when an office worker (that is, an operator) intends to operate office business data (for example, customer account book data), the operator selects a terminal (S2-1). In this case, the terminal 210 may be an office desktop PC or a mobile terminal connected to a network. When the terminal 210 logs in to the management server 110 of the cloud 100, an access lock process from another connection terminal to the management server 110 is performed (S2-2). Also, the management server 110 is notified by setting the terminal installation location (local), valid / invalid (invalid), role (terminal), etc., among the current terminal information settings in the management information of the terminal used. The information on the terminal 210 is confirmed in the license management table 410 (FIG. 4-2), and the information regarding the terminal is updated in the management information (S2-3). This management information is the management information table of FIG. Based on the updated management information and information in the license management table, the management server notifies the user terminal of available accounts. The management information at the time of login has the contents shown in FIG. 4-3, and the screen displayed on the terminal 210 is as shown in FIG. As shown in the management information and the screen example, there is principal data in the real server 220 of the local 200 during normal operation, which is backed up to the backup server 150 of the cloud 100, and the virtual server on the cloud is currently invalidated. As described above, the management information shown in FIG. 4A is updated from time to time as shown in FIG. 4A to FIG. It is dynamically rewritten and managed.

  The terminal 210 acquires information on the management location of the principal data, the management location of the used terminal 210, and the management location of the backup data from the management server 110 (S2-5). At this time, the latest version of the application to be used is registered in the application server 120 of the cloud 100, and the management server 110 can update the latest version by notifying the terminal 210 of the update information of the latest application. . In this case (normal time), the local 200 file server 220 is the access path. The operator activates the application on the terminal 210 and selects data to be accessed (S2-6).

  In order to determine the access method to the server, the registration location of the principal data (that is, the file server of the local (office) 200 or the virtual server 130 of the cloud (IDC) 100) is compared with the installation location of the terminal 210 (S2- 7). Since the terminal 210 and the file server 220 are in the same local 200 during normal operation, the access method is LAN access, so the file data is updated by directly operating the principal data of the file server 220 (S2-17). On the other hand, when the failure of the file server 210 as shown in FIG. 1-2-2, the principal data is replaced with the virtual server 130 of the cloud 100, and the location of the principal data is different from the terminal 210. Since access is made by file transfer by connection, the data of the virtual server 130 is not directly manipulated, but the data is transferred from the virtual server 130 to the storage device of the terminal 210 and registered as copy data, and after this is manipulated The update data is returned to the virtual server 130 again by the file transfer method, and the principal data of the virtual server 130 is updated (S2-8 to S2-11).

  For example, when using a remote desktop on a mobile terminal such as when accessing from a mobile terminal at home to a local server in the office or a terminal connected to the real server, the mobile terminal Does not necessarily have to be in the office, but may be in the home, in a moving car, or in a train. Even in such a case, the location of the terminal in the management information is local. In this case, step S2-7 is handled as it is because it is handled as a local in the above case. The principal data is backed up to the backup server 150 (S2-12). Also, application information used in the terminal 210 and management information of the management server 110 are backed up to the backup server 150 (S2-14 to S2-16). The backup timing may be simultaneous with the update or may be a predetermined date and time. The backup may be a bull backup or a differential backup.

  FIG. 3B is a flowchart for explaining processing on the management server 110 side in the switching processing when a failure occurs in the file server 220 of the local 200 in the first operation mode. The management server 110 monitors the operation of the file server 220 at any time, and when a failure such as a hardware failure such as a power source occurs, this may be detected and notified to the terminal 210 or the file server 220 may be notified. The terminal 210 that has become inaccessible may detect a failure. When logging in to the cloud management server 110 from the terminal 210 to continue the operation, access lock processing from another terminal to the management server 130 is performed (S3-1), and management information regarding the terminal 210 (terminal installation location ( (Local), valid / invalid (valid), role (change from terminal to single machine), etc.) are updated (S3-2). An example of a screen displayed on the terminal 210 at this time is shown in FIG.

  Next, the management server 110 obtains a list of servers that can be used as file servers among the resources managed by the management server, and notifies the operator of the terminal 210 to select them (S3-3). An example of this list is shown in FIG. As shown in FIG. 6, the management server 110 manages a plurality of servers in the cloud 100 and / or the local 200, and the presence / absence of data, whether the data is the latest, whether the server is valid / invalid, Whether it is a real machine or a virtual machine is managed. Unconfigured servers can function as file servers, backup servers, and cache servers depending on the settings. With regard to the slave server states 1 to 3, state 1 means cold standby, state 2 means warm standby, and state 3 means hot standby. The number of these servers and their status are appropriately determined according to the implementation environment and the service operation policy of the present invention, and are not limited to the example shown in FIG.

  In the environment shown in FIG. 6, in the case of condition 1 or 2 (non-operating server (function not set), stop, invalid, no data), the selected server is started (S3-7, S3-16) The server function is set (S3-8), and the data is restored using the backup data of the backup server 150 (S3-9). If condition 3 or 4 (non-operating file server, stop, valid, data not set), the backup data is restored as it is (S3-9). If condition 5 or 6 (slave server, cold standby, stop, valid, copy data, not updated), start the selected server (S3-13), restore the backup data from the backup server 150, and copy the copy data Update to the latest data (S3-14). In the case of condition 7 or 8 (slave server, warm standby, stop, valid, copy data, not updated), since it has already started, the backup data is restored from the backup server 150 and the copy data is updated to the latest data. (S3-14). In the case of condition 9 or 10 (slave server, hot standby, stop, valid, copy data, updated), since the data has already been started and the data is synchronized, the process directly proceeds to the switching process. When restoring backup data, the copy data may be updated by restoring full backup data or by restoring only the difference from the latest data. Further, a confirmation screen as shown in FIG. 5-3 may be displayed. In the switching process, the selected server (virtual machine or real machine) has been prepared as a new file server (principal data management device) for the management information (see FIG. 4-3) of the management server 110 of the cloud 100. (File server usage environment construction), management information (data (changed to principal data), enabled / disabled (changed to enabled)) of the selected server is updated (S3-10, S3-11). In the example shown in FIG. 1-2-2, the virtual server 130 of the virtual machine is configured as an alternative server at the time of failure. When the management information is updated (S3-11), an access unlock process for the management server 110 is performed (S3-12). Thereby, the management information is referred to from the terminal 210, and a new file server (the virtual server 130 of the cloud 100) is accessed. This access method is the file transfer method as described above.

  FIG. 3C is a flowchart for explaining the switching process at the time of failure recovery of the file server 230 of the local 200. After the file server 230 is restored, when the user terminal 210 logs in to the management server 110 of the cloud 100, an access lock process to the management server 130 is performed (S3-21). The management server 110 updates information on the terminal 210 (terminal installation location (local), valid / invalid (valid), role (change from a single machine to a terminal machine), etc.) on the management information (S3-22). Next, the management server 110 presents a list of servers that can be used again (see FIG. 6) to the terminal 210, and accepts selection of a recovery destination server (S3-23). In order to return the system after recovery of the file server 230, a local file server 230 is selected in this case (not shown). An example of the confirmation screen after selection is shown in Fig. 5-3. It should be noted that the recovery server may be automatically determined by the management server 110 according to a predetermined policy (server performance, installation location, real machine / virtual, etc.). Next, the management server 110 backs up the data of the virtual server of the cloud 100 used as the file server during the failure to the backup server 150 (S3-24). Next, the management server 110 restores the backup data from the backup server of the cloud 100 to the file server 220 of the real machine of the local 200 of the recovery destination, and constructs the usage environment of the file server 220 (S3-25). Then, the management server 110 updates the management information so as to switch the management location of the principal data to the local file server 230 restored from the cloud virtual server 130 (S3-26). Specifically, after performing an access lock process to the virtual server 130, the virtual server data setting is changed from “principal” to “none” in the management information, and the data setting of the file server 220 is “copy”. To “principal”. Delete the data stored in the virtual server after updating the management information. Next, the virtual server 130 of the cloud is stopped (S3-27), and the virtual server 130 is invalidated in the management information and updated to validate the local file server 230 (S3-28, S3-29). ). Then, an access unlock process to the management server 110 is performed to return the system to the original state (S3-30). In this embodiment, the file server 220 is a real server, but it may be a virtual server.

(5) Description of Second Operation Mode Next, the second operation mode shown in FIG. 1-2-3 and FIG. 1-2-4 will be described in detail using a flowchart, the contents of the management information table, and the like. . FIG. 3-4 (3-4-1, 3-4-2) is a flowchart for explaining the processing flow from the terminal 210 side during normal operation of the cache server 230 in the second operation mode. For example, in an accounting office system, when an office worker (that is, an operator) tries to operate an office file (for example, customer account book data), the operator selects a terminal (actual machine PC1 in FIG. 5-2-2). (S4-1). The terminal 210 accesses the management server indicated by the management server access path (S4-2). When the terminal 210 logs in to the management server 110 of the cloud 100, access lock processing from another connection terminal to the management server 110 is performed (S4-3). The management server 110 is notified by setting “role” to “terminal” and “valid / invalid” to “invalid” among the settings of the current terminal information in the management information of the terminal used. The information of the terminal 210 is confirmed with the license management table 410 (FIG. 4-2), and the information regarding the terminal is updated in the management information (registering the terminal 210) (S4-4). Based on the updated management information and information in the license management table, the management server notifies the user terminal of available accounts. The management information at the time of login has the contents shown in FIG. 4-4, and the screen displayed on the terminal 210 is as shown in FIG. 5-2-2. As shown in these drawings, the principal data is stored in the virtual server 130 of the cloud 100 during normal operation, and this is backed up to the backup server 150, and the cache server 230 is installed in the local 200.

  The terminal 210 acquires information on the management location of the principal data, the management location of the used terminal 210, and the management location of the backup data from the management server 110. At this time, the latest version of the application to be used is registered in the application server 120 of the cloud 100, and the management server 110 can update the latest version by notifying the terminal 210 of the update information of the latest application. . In this case (normal time), the local 200 file server 220 is the access path. The operator activates an application on the terminal 210 and selects data to be accessed. The latest version of the application to be used is registered in the application server 120 of the cloud 100, and the management server 110 notifies the terminal 210 of the update information of the latest application and the access path for principal data (S4-5). In this case (during normal operation), the access path is the virtual server 130 of the cloud 100. The terminal 210 acquires the management location of the principal data, the management location of the used terminal, and the management location of the backup data (S4-6). At this time, the use program, data, application update information, principal data access path, principal data management location, principal data management location, management location of the use terminal, which are notified or acquired by the terminal 210 and managed by the management server 110 of the cloud 100, Management information such as a backup data management location is stored in the cache server 230 (S4-7). The operation of the cache server 230 is performed in the background, and the operator of the terminal 210 is not aware of it.

  When these environments are ready, an application is started on the terminal 210 and data to be accessed is selected (S4-8). At the time of data access, in order to determine an access method to the server, the terminal 210 compares the location of the principal data with the installation location of the terminal 210 (S4-9). During normal operation, the principal data is in the virtual server 130 of the cloud 100, and thus the process proceeds to step S4-10. As will be described in detail later, since the cache server 230 of the local 200 substitutes for the virtual server (file server) when a failure occurs due to the failure of the virtual server 130, the principal data and the location of the terminal 210 are the same local 200, and the cache server Access to the principal data 230 is performed (S4-20). Since the virtual server 130 cannot be accessed when a network failure occurs, at least a part of the principal data of the cloud virtual server is restored based on the data of the local cache server to obtain partial copy data of the principal. By handling the partial cache data of the principal that cannot be directly manipulated as it is as partial copy data of the principal data that can be directly manipulated, the cache server 230 of the local 200 can manage the management server and the virtual server (file). Therefore, the principal data and the location of the terminal 210 are the same local 200, and the copy data of the cache server 230 is accessed (S4-20). When a network failure occurs, the management server access path is changed from the cloud management server to the management server (alternative function) in the cache server.

  During normal operation, the cache server 230 monitors a request from the terminal 210 and checks whether or not the cache of data to be accessed by the terminal 210 is in the cache server 230. This is provided (S4-12). If there is no cache data in the cache server 230, the designated data is transferred from the virtual server 130 of the cloud 100 to the local cache server 230, and the terminal 210 directly updates and updates the data in the cache server (S4). -11, S4-12). After the processing is completed, the cache data of the local cache server 230 is transferred to the management location of the principal data of the cloud virtual server, and this is reflected (S4-13). The principal data is backed up to the backup server 150 (S4-14). When another data is processed (S4-15: No), the process returns to the data selection process (S4-8) to be accessed, and when the data process is finished, the process is ended (S4-15: Yes).

  FIG. 3-5 is a diagram for explaining processing of the cache server 230 when a failure occurs in the virtual server 130 or the network in the second operation mode. In the second operation mode, as illustrated in FIG. 1-2-4, when the virtual server 130 of the cloud 100 fails or a failure occurs in the network connecting the cloud 100 and the local 200, the virtual server 130 Unable to access principal data. In this case, the operator logs in to the management server indicated by the management server access path from the terminal 210 (S5-1). In the case of a failure of the virtual server 130, the management tool of the cloud management server 110 is activated. The cache server 230 performs an access lock process for the cache server, and stops the operation of the local real machine as a cache server (S5-3). The cache server 230 updates the management information to invalidate the backup server 150 (S5-9). Then, the role of the local cache server 230 is switched from the cache server (cache SV) to the file server (file SV), and the cache server is changed from invalid to valid in the management information (S5-10). The data access process in the event of a failure follows the flow of FIG. 3-4, but the access destination is the copy area of the cache server 230.

  In the case of a network failure, the user logs in to the cache server 230 of the local 200 with user authority, and activates the management tool of the cache server 230 (S5-2). The cache server 230 performs an access lock process for the cache server, and stops the operation of the local real machine as a cache server (S5-3). Next, when there is a failure in the network, the management information registered in the cache data of the cache server 230 (see FIG. 4-4) is copied to the copy data area in the cache server 230 (S5-4). ). In addition, the update data registered in the cache data is copied to a copy area in the cache server to construct a use environment (S5-5). Then, in the management server (alternative function) in the cache server constructed from the copied management information, the management information management location is rewritten from the management server 110 of the cloud 100 to the management server (alternative function) in the local cache server ( S5-6), the management information is updated to change the management location of the principal data from the cloud to the local cache server 230 (S5-7). In response to this, the terminal 210 switches the login management server access path to the management server (alternative function) in the cache server 230 of the local 100 (S5-8). The cache server 230 updates the management information to invalidate the backup server 150 (S5-9). Then, the role of the local cache server 230 is switched from the cache server to the file server, and the management server (alternative function) information (number 7 in FIG. 4-6) in the cache server is added to the management information. In step S5-10, the cache server is changed from invalid to valid. The data access process in the event of a failure follows the flow of FIG. 3-4, but the access destination is the copy area of the cache server 230.

  FIG. 3-6-1 is a flowchart for explaining the switching process at the time of failure recovery in the second operation mode from the cache server 230 side. When the failure of the virtual server 110 of the cloud 100, which is the original main server, or the network failure is recovered, the management server 110 of the cloud 100 detects this and notifies the terminal 210 of the local 200 (S6-1-1). When the network failure is recovered, the terminal 210 receiving this logs in to the management server (alternative function) in the local cache server 230 (S6-1-2). The cache server 230 of the local 200 performs access lock processing (S6-1-3), and the management server (alternative function) in the local cache server 230 has information (management) of the location of the management server to which the terminal 210 is logged in The server access path is changed from the management server (alternative function) in the cache server to the management server 110 of the cloud 100 (S6-1-4). At this time, the access path of the management server is also updated to the cloud 100 in the management information. Thereafter, the management server (alternative function) in the local 100 cache server accepts the logout of the use terminal 210 (S6-1-5).

  FIG. 3-6-2 is a flowchart for explaining the switching process at the time of failure recovery in the second operation mode from the management server 110 side of the cloud 100. When the management server 110 of the cloud 100 accepts the login of the terminal 210, the access lock processing to the management server 110 is performed (S6-2-1). The management server 110 acquires the information of the terminal 210 and updates the management information of the management server 110 (S6-2-2). The available account is notified to the terminal 210. Next, the management server 110 backs up the data (copy data, update data, management information) of the local 200 cache server to the backup server 150 of the cloud 100 (S6-2-3). In addition, among the resources managed by the cloud management server 110, a list of available servers is notified to the terminal 210 (see FIG. 6), and the operator selects which server to restore the principal data (see FIG. 6). S6-2-4, FIG. 6). This may be automatically determined by the management server 110 according to a predetermined policy (server performance, installation location, real machine / virtual, etc.). When the recovery destination server (cloud virtual server) is determined, the server is started up (S6-2-5), the backup data is restored from the backup server 150 of the cloud 100, and the use environment of the virtual server 130 is constructed. (S6-2-6). Thereafter, after invalidating the cache server 230 of the real machine of the local 200, the role of the cache server 230 is switched again from the file server at the time of failure operation to the cache server, and the contents are reflected in the management information (S6-). 2-7). Then, the management server 110 deletes the copy data created in the cache server 230 of the local 200, and updates the management information to validate the cache server 230 (S6-2-8). Thereafter, access unlock processing to the management server 110 of the cloud 100 is performed (S6-2-9), and logout from the terminal 210 is accepted (S6-2-10). Thereafter, the operation at the normal operation is performed by accessing the management server 110 and the virtual server 130 of the cloud 100 from the terminal 110 of the local 100.

  The first and second operation modes of the present invention have been described in detail above. However, the present invention is not limited to the above-described embodiment, and various modifications and changes are made using the available resources shown in FIG. An example can be realized. Hereinafter, some modified examples will be described.

(6) Description of First Modified Example As a first modified example of the embodiment of the present invention, it is shown in FIG. 1-2-3 from the normal operation time of the first operation mode shown in FIG. 1-2-1. An operation mode for shifting to the second operation mode will be described. As shown in FIG. 1-2, the file server 220 of the local 200 is the main server, and the terminal 210 of the local 100 accesses the principal data of the file server 220 in normal times. The management server 110 of the cloud 110 defines this operation form as management information, and the principal data of the file server 220 is backed up to the backup server 150 of the cloud 100. From this state, conditions such as the number of terminals 210 accessing the file server 220, which is the main server, installation locations (whether they are in the office or accessed via the Internet from a remote location), or pre-defined operations Based on the policy, when a predetermined condition is satisfied, the process shifts to the second operation mode illustrated in FIG. 1-2-3. For example, when the number of terminals 210 to be remotely connected from a remote place becomes 10 or more, the load on the office (local) file server 220 becomes excessive, or the access from the outside to the office LAN is increased. From the viewpoint of being unpreferable, it is conceivable to move the main server onto the cloud 100 to be in the second operation mode. When switching to the second operation mode, the virtual server 130 is set up in the virtual server virtual machine 140 (this may be a real machine) on the cloud 100 and the backup data is restored from the backup server 150. Thereafter, the terminal 210 The virtual server 130 is accessed as a main server. At this time, the file server 220 of the local 100 is switched to the cache server, and the terminal 210 in the office (local 100) can perform business using the cache data.

  FIG. 3-7 is a flowchart for explaining the switching operation for shifting to the second operation mode shown in FIG. 1-2-3 from the normal operation of FIG. 1-2-1 in the first modification. It is. The conditions for triggering the operation switching are monitored by the management server 110 of the cloud 100, and when the conditions are satisfied, when a special input instruction is given by the operator of the terminal 210 or the system administrator, or from the terminal 210, the management server When the user logs in, the management server 110 of the cloud 100 performs an access lock process (S7-1-1), and the management information is updated based on the information of the terminal 210 (S7-1-2). The management server 110 notifies the terminal 210 of an available account. Next, a list of available servers among resources managed by the cloud management server 110 is presented to the operator (see FIG. 6), and selection of a switching destination server is accepted (S7-1-3). In this case, a confirmation screen shown in FIG. 5-3 may be presented. Next, the cloud management server 110 performs an access lock process for preventing each terminal from connecting to the file server 220 of the local 200 real machine that has been used as the main server until now (S7-1-4). Thereafter, if the backup data is not the latest, the data of the file server 220 is backed up to the backup server 150 of the cloud 100, and the principal data stored in the local file server 220 is deleted (S7-1-5). ). Thereafter, the selected server (virtual server 130 in the example of FIG. 1-2-3) is activated (S7-1-6), the backup data is restored from the backup server 150, and the usage environment is constructed (S7-). 1-7). Then, after disabling both the management information file server 220 and the virtual server 130 of the cloud 100, the management location of the principal data is changed from the file server 220 of the local 200 to the virtual server 130 of the cloud 100, The management information of the management server 110 is updated so that the role of the local 200 file server used as the main server is switched to the cache server, and both the file server 220 and the virtual server 130 of the cloud 100 are enabled. (S7-1-8, S7-1-9). When the above is completed, access unlock processing of the management server 110 is performed (S7-1-10), and the terminal 210 logs out of the management server 110 (S7-1-11). Thereafter, the operation form shown in FIG. 1-2-3 is performed, and the terminal 210 operates data using the virtual server 130 of the cloud 100 as a main server. Although not described in detail, for example, a condition that the number of terminals accessing the main server is reduced to a predetermined number or less, or a predetermined operation policy (a predetermined time zone of the day (for example, daytime) / Night), or on weekdays / all day, or when failure occurs, priority is given to recovery or operation continuation, etc.). It is possible to return to the operation mode 1.

(7) Description of Second Modification As a next example, from a state in which a virtual server of a cloud virtual machine and a real server of a local real machine are operating as a master server and a slave server, respectively, for example, local If the response from the virtual server (master server) of the cloud virtual machine slows down due to the concentration of accesses from other terminals, the load increases, or the network communication speed decreases, the slave server is cached. An example of switching to operation used as a server (second operation mode shown in FIG. 1-2-3) will be described. In normal operation, in the overall configuration diagram shown in FIG. 1-1, the virtual server 130 of the virtual machine in the cloud 100 has a data principal (ie, master server), and copy data is provided in the local 200 file server (ie, slave). Server) and the cloud 100 have a backup server 150. In addition to being in the local 100, the terminal 230 may be connected to the network from a remote value. The location and role of these servers are managed by the management information of the management server 110. An example of management information in this case is shown in FIG.

  From this state, for example, when the access from the terminal is concentrated and the load becomes large, or when the traffic of the network connecting the local and the cloud increases and the response delay occurs, the management server 110 is the slave server until then. A process of substituting the cache server for the file server 220 of the local 200 used is performed. FIG. 3-8 is a flowchart for explaining the switching process of the embodiment.

  As shown in FIG. 3-8, for example, when the switching condition of the operation mode as described above is satisfied, when there is a special input instruction from the operator of the terminal 210 or the system administrator, or from the terminal 210 to the management server When the user logs in, the management server 110 of the cloud 100 performs an access lock process (S8-1), and the management information is updated based on the information of the terminal 210 (S8-2). The management server 110 notifies the terminal 210 of an available account. Next, a list of available servers among resources managed by the cloud management server 110 is presented to the operator (see FIG. 6), and selection of a switching destination server is accepted (S8-3). In this case, a confirmation screen shown in FIG. 5-3 may be presented. The management server 110 performs access lock processing of the file server 220 of the local machine 200 (S8-4), disables the real machine, and switches the role of the real machine from the file server to the cache server (see FIG. 4-5, the “role” “file SV” of number 2 in FIG. 4-5 is rewritten to “cache SV” in number 2 of “role” in FIG. 4-6), and the contents of the management information are updated by enabling the real machine (S8-5). Further, the management server 110 deletes copy data of the local real machine (S8-6), and performs an access lock process to the local cache server (S8-7). Thereafter, access unlock processing of the management server 110 is performed (S8-8), and the use terminal 210 logs out of the management server 110 (S8-9). An example of the management information in this state is shown in FIG. As shown in FIG. 1-2-3, this management information has a data principal in the virtual server 130, backup data is taken in the backup server 150, and the local real server becomes the cache server 230. It becomes an operational form. From this point, when a failure further occurs in the master server (virtual server 130) or the network, the operation mode at the time of failure shown in FIG.

  Also, the transition (change) of the management information in the management information table (FIGS. 4-1, 4-3, etc.) is recorded as a log together with the date and time, etc. Later, the system administrator on the cloud side or the local side A system administrator can check the list. Also, based on the management information transition (change) log, aggregate the status of the failed device, the net failure, the failure occurrence of the virtual server of the virtual machine on the cloud side, generate a tabulated list, etc. In addition, it is possible to grasp the occurrence status or operation rate of a failure or failure. In addition, the trend of the number of terminal connections over time, the occurrence of network failures, the availability of virtual servers on virtual machines on the cloud side, the availability and failure rate of real servers on local real machines Such information can be fed back to the operation policy and reflected to be used as a reference for switching between the first operation mode and the second operation mode.

  Here, the contents of the failure occurrence include the following examples in addition to the failure of the local real server, the network failure, and the failure of the cloud virtual server described above.

  First, when operating in the first operation mode, there may be a case where a network failure or a management server failure further occurs at the time of a failure of a local real server (a state in which a single machine is operated).

  In this case, when a user normally starts and logs in the real machine PC1 (single machine) of the accounting office and accesses the management server of the cloud (IDC center), there is a failure in the management server. Since the latest application information and the latest management information cannot be obtained, the application is started based on the current old application stored in the actual machine PC1 (single machine) and the current old management information, and the principal data is accessed. Input process.

  After that, when the trouble of the management server is resolved and the latest application information can be obtained, if there is input data opened by the old application, the old application is registered and the old application is terminated. . After that, it connects to the management server, downloads and installs the latest application to the actual machine PC1 (single machine), and then performs normal processing.

  Alternatively, if a communication failure occurs before the app is started, the communication failure occurs even if the user tries to access the management server by logging in after starting the actual PC 1 (single device) of the accounting office. Since the latest application information and the latest management information cannot be obtained, the data management application starts based on the current application stored in the actual PC 1 (single machine) and the current management information, and the data management menu is displayed. .

  In the data management menu, the location where the management location is “center” on the previous time is grayed out and “No communication failure has occurred. No. 1. Do you want to switch the copy data of the actual machine PC1 to the principal data? "A confirmation message appears. If you agree with this, the copy data of the actual machine PC1 (single machine) is switched to the principal data. The real machine PC1 (single machine) directly accesses the principal data of the switched real machine PC1 (single machine) to perform input and registration processing.

  When the communication failure is resolved and communication with the IDC center is established, when the user starts and logs in to the accounting office's actual PC 1 (single machine) and accesses the management server of the IDC center, the data management application Start up and display the data management menu. The location where the management location is “Center” on the data management menu returns to the original display, “Because communication failure has been recovered, number 1. Do you want to switch the principal data of PC1 to number 2. Virtual SV1? If you agree, the copy data of the real machine PC1 (single machine) is copied to the virtual SV1 (server) that is the virtual server of the IDC server, and the principal data of the real machine PC1 (single machine) is copied. Switch to data.

  Alternatively, when operating in the normal operation state in the second operation mode, the network is normal, but there may be a failure in the real server (actual server) of the local real machine (cloud virtual server and terminal) And operation).

  As described above, depending on whether the system components are valid or invalid, the operation can be switched from the first operation mode or the second operation mode to operation with a single machine or operation with a terminal and a virtual server However, even if the network (Internet side) connection goes offline due to some failure, or when a failure occurs in the system components, the system can be switched and used in the same way. Is possible.

  For example, when the network (on the Internet side) goes offline, the user terminal itself can determine the state of data used on the user terminal without inquiring of the management server.

  More specifically, for example, the network cannot be connected between the user terminal or file server (local real server) and the management server, and the user terminal or file server (local real server) shifts to offline operation. If a failure occurs in any of the system components including the file server (cloud virtual server) and it is determined that the data is offline, whether the data to be processed is the original data Instead of inquiring to the management server whether it is a copy of this data (pre-update data), the control unit of the user terminal or the control unit of the file server (local real server) makes an independent determination. And continue to use the system.

  Alternatively, when it is determined that a failure has occurred in the file server (cloud virtual server), if the data to be processed in the user terminal is a copy of the principal data (data before update), a copy of the principal data (Pre-update data) can be temporarily treated as principal data (temporarily shifted to the same processing as that of a single machine), and the system can be used continuously.

  In this case, at the user terminal, the original data or the copy of the original data (pre-update data) is used to update the data, and the updated data that has been created is restored after the network connection is restored. Return to the active component (virtual server etc.) of the system configuration set as the storage location of this data.

  In this way, depending on the components of the system that can no longer be used offline or the components of the system in which a failure has occurred, the management location of the principal data and the management location of the copy of the principal data are switched to operate dynamically. it can.

  In addition, the user can freely set the management location of the principal data, or if a failure occurs, the management location of the principal data can be temporarily changed dynamically according to the components of the system where the failure has occurred. Can be switched to. In other words, it is possible to provide an integrated new service that can be flexibly operated according to various forms according to the management location of the set principal data and the copy data of the principal and the validity of the system components. it can.

  As described above, in the present invention, any of the servers shown in FIG. 1-1 may not exist, and other virtual servers or real servers may exist in the cloud or locally in addition to those illustrated. The information and roles of these servers are managed by the management information of the management server 110, and the operations of the file server (master or slave), cache server, backup server, etc. are switched appropriately according to the failure occurrence status and operation policy. Can do. As a result, redundancy can be ensured and optimum system operation can be performed according to the situation.

  The present invention can be used in the information processing industry that uses application data from local and / or cloud servers.

100 cloud (IDC)
110 Management server 120 Application server 130 Virtual server 140 Virtual machine 150 Backup server 200 Local (office)
210 terminal 220 file server 230 cache server

Claims (14)

A cloud management server that holds management information that defines the role of each server, a cloud backup server, a cloud virtual server virtual machine, a local real machine real server, and one or more local terminals Connected to the network and during normal operation, the management information of the management server is a data management system defined to operate the real server of the local real machine as a main server,
First backup means for backing up the data of the local primary server to the backup server of the cloud;
When a failure occurs in a local primary server, a virtual server is generated from the cloud virtual server virtual machine in the management information of the management server, and the virtual server of the virtual machine is operated as a primary server. Write the information and use the backup data of the cloud backup server to restore the local primary server data to the virtual server of the cloud virtual server virtual machine, temporarily store the virtual machine virtual server As a primary server, a restoration method in case of failure to continue system use,
Access change means for switching the access method when accessing the virtual server of the virtual machine for the virtual server of the cloud from the one or more local terminals after the restoration to the file transfer method instead of the remote connection method during the normal operation When,
A data management system comprising: using a cloud backup server in the event of a failure of the local primary server. The data management system according to claim 1, further comprising:
As a means for recovering the system to a normal operation state when the failure of the local primary server is recovered,
Based on the management information of the management server, as a server that can be used as a data recovery destination, a real server in which the failed local real server is repaired, another real server in the local or cloud, and a virtual in the local or cloud A recovery destination server selection means for selecting a data recovery destination server from the servers,
Means for backing up the updated contents of the data of the virtual server in the cloud operated as the temporary main server to the backup server;
The backup data of the backup server is restored to the server selected as the recovery destination, and the virtual server of the cloud that has operated the primary server as the temporary primary server in the management information of the management server after the restoration. A data management system comprising: a restoration unit for restoration at the time of writing information for switching from the server to the server selected as the restoration destination, and restoring the system to a normal operation state. In the data management system according to claim 1 or 2,
Using a management server constructed virtually as the cloud management server,
Using a virtually built backup server as the cloud backup server,
Using a server constructed virtually instead of the real server of the local real machine,
A data management system characterized by operation. A virtual server of a cloud virtual machine, a cloud management server that holds management information that defines the role of each server, a real server of a local real machine, and one or more local terminals are connected via a network,
During normal operation, the management information of the management server is a data management system defined to operate the virtual server of the cloud virtual machine as a main server and the real server of the local real machine as a cache server. And
The local cache server is
Cache means for caching the data of the main server of the cloud and the management information of the management server in the cache server to be used by the one or more local terminals;
When a communication failure of the network or a failure of the main server of the virtual machine occurs, the management information of the management server and the cloud information are stored on the local cache server based on the data of the local cache server. By restoring at least a part of the principal data of the virtual server to make partial copy data of the principal, principal data that can directly manipulate the partial cache data of the principal that cannot be directly manipulated as it is Recovering means for rewriting the management information so as to temporarily operate the local cache server as a temporary main server,
A data management system characterized by continuing system use in the event of communication failure or failure of a primary server in a cloud virtual machine. The data management system according to claim 4, further comprising:
When the communication failure of the network or the failure of the primary server of the virtual machine is recovered, the update contents of the data of the local cache server that has been operated as the temporary primary server are transferred to the virtual machine of the virtual machine of the cloud. A data management system comprising failure recovery processing means to be reflected on a server. A management server that holds management information that defines the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are connected via a network,
In the management information of the management server, the configuration of the data management system according to any one of claims 1 to 3 is a first operation mode, and the configuration of the data management system according to claim 4 or 5 is a second configuration. It is registered as an operation form,
Based on a condition including the number of local terminals connected to the main server and / or a pre-defined operation policy, the contents of the management information are mutually exchanged between the first operation mode and the second operation mode. A data management system comprising operation mode switching means for switching. A management server that holds management information that defines the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are connected via a network,
The management information of the management server is usually operated in a two-server configuration in which one of the virtual server of the cloud virtual machine and the real server of the local real machine is a master server and the other is a slave server. Operation mode,
The configuration of the data management system according to claim 4 or 5 is registered as a second operation mode,
A data management system comprising operation mode switching means for switching from the normal operation mode to the second operation mode based on environmental conditions including a load on the local terminal and a communication speed of the network. A cloud management server that holds management information that defines the role of each server, a cloud backup server, a cloud virtual server virtual machine, a local real machine real server, and one or more local terminals In the operation method of a data management system connected via a network,
During normal operation, the management server sets the management information to operate the real server of the local real machine as a main server,
The cloud backup server stores backups of the local primary server;
If the local primary server fails,
The management server generates a virtual server from the virtual machine for the virtual server of the cloud, writes information for operating the virtual server of the virtual machine as a main server in the management information, and The backup data is used to restore the local primary server data to the virtual server of the cloud virtual server virtual machine, the virtual machine virtual server is used as a temporary primary server, and system use is continued.
After the restoration, the access method when accessing the virtual server of the virtual machine for the virtual server of the cloud from the one or more local terminals is switched to the file transfer method instead of the remote connection method during the normal operation,
A method of operating a data management system, characterized in that in the event of a failure of the local primary server, system use is continued using a cloud backup server. The method of claim 8, further comprising:
When the failure of the local primary server is recovered,
Based on the management information, the management server recovers the data recovery destination from a real server that has repaired the failed local real server, another local or cloud real server, and a local or cloud virtual server. Select your server,
The backup server backs up the updated contents of the cloud virtual server data that was operating as the temporary primary server,
The backup data of the backup server is restored to the server selected as the recovery destination, and the virtual server of the cloud in which the management server has operated the primary server as the temporary primary server in the management information after the restoration. An operation method of a data management system, wherein the management information is updated so as to switch from a server to a server selected as the recovery destination, and the system is restored to a normal operation state. The method according to claim 9 or 10, wherein
Using a management server constructed virtually as the cloud management server,
Using a virtually built backup server as the cloud backup server,
Using a server constructed virtually instead of the real server of the local real machine,
A method of operating a data management system characterized by operating. Data in which a virtual server of a cloud virtual machine, a cloud management server that holds management information that defines the role of each server, a real server of a local real machine, and one or more local terminals are connected via a network In the management system operation method,
During normal operation, the management server sets the management information to operate the virtual server of the cloud virtual machine as a main server and the real server of the local real machine as a cache server,
The local cache server caches the data of the cloud main server and the management information of the management server in the cache server to be used by the one or more local terminals,
When the network communication failure or the failure occurs in the main server of the virtual machine, the cache server manages the management information of the management server on the local cache server based on the data of the local cache server. Then, by restoring at least a part of the principal data of the cloud virtual server to make a partial copy of the principal, the partial cache data of the principal that cannot be directly manipulated as it is is directly An operation method of a data management system, wherein the management information is rewritten so that the local cache server is temporarily operated as a temporary main server, and is handled as partial copy data of operable principal data. The method of claim 12, further comprising:
When the network communication failure or the failure of the primary server of the virtual machine is recovered, the local cache server updates the update contents of the data of the local cache server that has been operated as the temporary primary server. A method of operating a data management system, which is reflected in a virtual server of a virtual machine in the cloud. A data management system in which a management server that holds management information that defines the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are connected via a network In the operation method of
In the management information of the management server, the operation method according to any one of claims 8 to 10 is registered as the first operation mode, and the operation method according to claim 11 or 12 is registered as the second operation mode. And
Based on a condition including the number of local terminals connected to the main server and / or a predetermined operation policy, the management server converts the contents of the management information into the first operation mode and the second operation. A method of operating a data management system, characterized in that the mode is switched between modes. A data management system in which a management server that holds management information that defines the role of each server, a virtual server of a cloud virtual machine, a real server of a local real machine, and one or more local terminals are connected via a network In the operation method of
The management information of the management server is usually operated in a two-server configuration in which one of the virtual server of the cloud virtual machine and the real server of the local real machine is a master server and the other is a slave server. Operation mode,
The operation method of the data management system according to claim 11 or 12 is registered as a second operation mode,
An operation method of a data management system, wherein the management server switches from the normal operation mode to the second operation mode based on an environmental condition including a load on the local terminal and a communication speed of the network. .

Images