JP7141908B2 - Data management system and data management method - Google Patents

Description

The present invention relates generally to techniques for providing requested data.

In a business system within a company, data generated by a business application is stored in a storage device. A backup operation is performed for data protection against data corruption due to storage device hardware failure, software failure, or human error. Backups are acquired periodically, and multiple generations of backup data may be stored. Here, a storage device that stores data generated by a business application is called primary storage, and a storage device that stores backup data is called secondary storage. Primary storage and secondary storage may be provided in the same device, but backup data is for protection and is used infrequently, so it is stored on a low-performance disk or another device. In some cases. Backup data may also be stored in archival storage for long-term storage. The stored storage devices may have different specifications and protocols, such as block storage, file storage, object storage, and the like.

Business application data is requested to be used for uses other than business applications, such as in-house analysis, auditing, and application development testing. In that case, the data manager who manages the data extracts and provides the data that matches the usage request. The requested data may be in primary storage, secondary storage, or only in archival storage, and the data manager identifies the storage where the data resides and extracts the data. work load.

Here, when providing business application data for another purpose, it is desirable not to affect the business application. Therefore, there is a method of obtaining and providing a snapshot of the data in the primary storage. In this regard, in order to provide data, a method for automatically executing a workflow that acquires a snapshot of a data object, creates a live clone volume, and mounts it to an application is defined (Patent Document 1). reference).

U.S. Pat. No. 9,665,437

In the technique described in Patent Document 1, a workflow is defined according to usage, and data is provided from acquired snapshots. However, the technology described in Patent Literature 1 cannot be applied to backup data that has already been backed up. For this reason, if the unit you want to use is a consumption analysis for each customer in the past year, the necessary data is data for multiple applications (customer list management, sales management, etc.), and it is for a long period. There is a high possibility that it will not be stored in the storage, and there is a high possibility that the time specified in the data usage request will not match the backup acquisition time, so it will be difficult to provide the requested data. .

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above points, and intends to propose a data management system capable of providing data that matches the requested time.

In order to solve this problem, the present invention provides a data management system including a plurality of storage devices, comprising: Storage stored in the plurality of storage devices based on data management information in which information indicating the time when the storage data for storage was acquired and information indicating the storage device in which the storage data is stored are associated A data specifying unit for specifying, from among data, storage data that matches the requested time included in the request accepted by the request accepting unit; and an output unit for outputting the specified result by the data specifying unit. did.

In the above configuration, for example, the association between the requested time and the date and time when the stored data of the data was acquired is grasped, so the stored data that meets the time requirements of the requested data can be specified, and the stored data The stored data can be obtained from the information related to

According to the present invention, requested data can be provided.

It is a figure showing an example of composition concerning a data management system by a 1st embodiment. 3 illustrates an example hardware configuration of a management server according to the first embodiment; FIG. FIG. 3 is a diagram showing an example of a configuration related to functional modules of a data management program according to the first embodiment; FIG. 4 is a diagram showing an example of data management information according to the first embodiment; FIG. It is a figure which shows an example of AP management information by 1st Embodiment. 4 is a diagram showing an example of data attribute management information according to the first embodiment; FIG. FIG. 7 is a diagram showing an example of a flowchart relating to data provision processing according to the first embodiment; FIG. 10 is a diagram showing an example of a flowchart relating to data identification processing when date and time are specified according to the first embodiment; FIG. 7 is a diagram showing an example of a flowchart relating to data specifying processing when a period is specified according to the first embodiment; 4 is a conceptual diagram of selection of backup data when a period is specified according to the first embodiment; FIG. FIG. 3 is a diagram showing an example of storage device management information according to the first embodiment; FIG. FIG. 7 is a diagram showing an example of a flowchart relating to request level confirmation processing according to the first embodiment; FIG. 7 is a diagram showing an example of a flowchart relating to restoration time confirmation processing according to the first embodiment; FIG. 7 is a diagram showing an example of a flowchart relating to provision form determination processing according to the first embodiment; FIG. 3 is a diagram showing an example of an input/output interface according to the first embodiment; FIG. It is a figure which shows an example of business data management information by 2nd Embodiment. FIG. 12 is a diagram showing an example of storage tier management information according to the third embodiment; FIG. FIG. 13 is a diagram showing an example of usage history management information according to the third embodiment; FIG.

One embodiment of the present invention will be described in detail below with reference to the drawings.

(1) First Embodiment In FIG. 1, 100 indicates a data management system according to the first embodiment as a whole.

FIG. 1 is a diagram showing an example of the configuration of a data management system 100. As shown in FIG. The data management system 100 includes a management server 110 , a host computer 120 , an SDS (Software Defined Storage) system 130 and a client terminal 140 .

The management server 110 and the host computer 120 are connected by a NW (Network) 150 for management. Management server 110 and SDS system 130 are connected by NW 150 for management. The management server 110 and the client terminal 140 are connected by an access NW 160 . The host computer 120 and the SDS system 130 are connected by a NW 170 (IP (Internet Protocol) network or SAN (Storage Area Network)). In addition, the configuration of the network is not limited to the configuration described above, and an appropriate configuration can be adopted.

The management server 110 executes the data management program 111 and manages data stored (stored) in the SDS system 130 .

One or a plurality of host computers 120 are provided to provide business APs 121 including one or more APs (Application Programs), and store data in the SDS system 130 .

The SDS system 130 has a control node 131 . The control node 131 executes control software, virtualizes the storage device 132 that constitutes the SDS system 130 , and provides an interface to the host computer 120 . As the storage device 132 constituting the SDS system 130, primary storage for storing the data of the business AP 121, storage data of the data (backup data obtained by backing up the data, archive data obtained by archiving the data, snapshot There is a secondary storage that stores snapshot data acquired by Hereinafter, backup data, archive data, and snapshot data will not be particularly distinguished, and the case of backup data will be described as an example, but the present invention can also be applied to archive data and snapshot data. Note that the storage device 132 may be a storage obtained by virtualizing disks of a plurality of general-purpose servers, or may be a general storage subsystem.

The client terminal 140 is a terminal used (operated) by a user who uses data, and accesses the data management program 111 (eg, transmits a use request for the user to request data).

FIG. 2 is a diagram showing an example of the hardware configuration of the management server 110 that executes the data management program 111. As shown in FIG. The management server 110 includes one or more CPU 201 (central processing unit), memory 202, secondary storage device 203 such as a hard disk, input/output IF (interface) for controlling input from a keyboard and mouse and output to a display. 204, which includes a network IF 205 for connecting to a network (NW 150 for management, NW 160 for access, etc.).

A data management program 111 is loaded onto the memory 202 of the management server 110 and executed by the CPU 201 . Note that the functions of the management server 110 (for example, the functional modules of the data management program 111 shown in FIG. 3) may be realized by software as described above, or may be realized by hardware such as a dedicated circuit. However, it may be realized by combining software and hardware. Also, part of the functions of the management server 110 may be implemented by another computer that can communicate with the management server 110 .

The secondary storage device 203 also stores a data management DB (database) 206 used by the data management program 111 . Note that the management server 110 may be implemented as a virtual machine instead of a physical machine.

As an additional remark, the hardware configuration of the general-purpose server in the storage device 132 configured as the SDS system 130 is the same as the configuration shown in FIG. 2, and the description thereof will be omitted.

FIG. 3 is a diagram showing an example of a configuration related to functional modules of the data management program 111. As shown in FIG. The data management program 111 includes a self-service portal IF section 301, a data management DB section 302, a request reception section 303, a data identification section 304, a data information collection section 305, a backup information collection section 306, a data evaluation section 307, and an output section 308. Consists of

The self-service portal IF unit 301 communicates with the client terminal 140 and provides an interface for users who use data. The data management DB unit 302 manages the data management DB 206 (attributes related to data, backup information, restoration information, etc.). The request receiving unit 303 receives, for example, a request (for example, a data usage request from a user) including information indicating a requested time (for example, a target time to be described later).

The data identification unit 304 includes, for example, information indicating the time (date, time, etc.) when the storage data for storing the data was acquired, and the storage device (storage location, storage identifier, volume identifier, etc.) in which the storage data is stored. ) is associated with information indicating the Identify the stored data that matches the requested time included in the request.

The data information collection unit 305 collects information about data from the business AP 121 or according to an input from a data manager. The backup information collection unit 306 collects data backup information from the SDS system 130 or from the backup execution tool of the host computer 120 .

For example, based on the data management information, the data evaluation unit 307 calculates the time difference between the storage data specified by the data specifying unit 304 and the requested time included in the request. Evaluate whether the included level of requirements is met. Further, for example, the data evaluation unit 307 calculates the time required to restore the stored data identified by the data identification unit 304, and determines whether the calculated time required for restoration is within the possible waiting time included in the request. evaluate. The output unit 308 outputs the result of identification by the data identification unit 304 (performs format conversion, transfer, etc. for providing the identified storage data). For example, the output unit 308 outputs information related to storage data specified by the data specifying unit 304 . The output unit 308 may output information indicating the storage destination of the stored data, may output the stored data, or may output an identifier for identifying the stored data and the data. An identifier for identifying a storage location to store the data may be output.

FIG. 4 is a diagram showing an example of data management information (data management table 400) included in the data management DB 206. As shown in FIG. The data management table 400 manages backup information for each piece of data for which backup has been acquired, on the time axis. The data management table 400 includes a data identifier field 401 , date field 402 , identifier field 403 , backup acquisition unit field 404 , storage location field 405 , size field 406 and acquisition time field 407 .

The data identifier field 401 stores information indicating a data identifier that identifies data. The date field 402 stores information indicating the date of backup data acquisition (backup date). The identifier field 403 stores an identifier that identifies backup data (backup file). The backup acquisition unit field 404 stores information indicating the backup acquisition unit. The storage location field 405 stores storage information indicating the backup destination (storage destination). The size field 406 stores information indicating the size of backup data. The acquisition time field 407 stores information indicating the acquisition time of the backup data.

Here, the data identifier is a data unit for identifying AP data, such as a DB table name or column name in the case of DB data, an object identifier in the case of an object, and a file identifier in the case of a file. . The storage location is a storage identifier, a volume identifier in the storage, a directory, or the like, and is used to identify the storage destination of backup data. In the example of FIG. 4, the storage location is indicated by the storage identifier (storage 1) and the volume identifier (LU1). As a unit of acquisition, backup methods such as full backup, differential backup, and incremental backup are managed.

A full backup is a backup method that retains all data. A differential backup is a backup method that retains a difference from a full backup. Incremental backup is a backup method that retains increments (changes) from the previous backup. For example, the difference between a differential backup and an incremental backup will be explained when data is updated and a backup is taken as follows.
Time A: Data X added to Data W Time B: Full backup Time C: Data Y added Time D: Differential backup (or incremental backup)
Time E: Addition of data Z Time F: Differential backup (or incremental backup)

In the full backup at time B, all data (data W and data X) are saved as backup data.

In the differential backup at time D, differential data (data Y) from the backup data (data W and data X) of the previous full backup is saved as backup data. In the incremental backup at time D, the changed data (data Y) from the previous backup (time B) is saved as backup data.

In the differential backup at time F, differential data (data Y and data Z) from the backup data (data W and data X) of the previous full backup is saved as backup data. In the incremental backup at time F, the changed data (data Z) from the previous backup (time D) is saved as backup data.

In this embodiment, full backup, differential backup, and incremental backup are used as examples of backup methods. can also

The data management program 111 may acquire and register the backup information from the backup execution tool, or may acquire and register information on data stored in the secondary storage from the SDS system 130 . The data identifier may be obtained by the data management program 111 from the business AP 121, or may be restored by restoring the backup data when obtaining the backup information, and reading and registering the contents of the data. The registered data identifiers are updated, if any, by restoring and reading the backup data each time backup information is acquired or periodically.

Although not shown in FIG. 4, for example, as shown in FIG. 10, in the data management table 400, if there is a time stamp in the data, the initial time stamp, which is the first time stamp in the data, and the data Additionally manage the final timestamp, which is the last timestamp in the . The time stamp information is registered together with the data management program 111, for example, when the backup data is temporarily restored and the contents of the data are read.

FIG. 5 is a diagram showing an example of AP management information (AP management table 500). AP management table 500 is included in data management DB 206 . The AP management table 500 is a table for managing information on APs from which backup data has been obtained. The AP management table 500 includes an AP identifier field 501 , an AP type field 502 and a backup means field 503 .

The AP identifier field 501 stores information indicating an AP identifier that identifies an AP. The AP type field 502 stores information indicating the type of AP (for example, software name). The backup means field 503 stores information indicating backup execution means (information that is used when data is restored when data is provided and that can identify how the backup was performed). FIG. 5 shows an example in which the AP is a DB system, the DB name is registered as the AP identifier, and the name of the DB management system is registered as the AP type. Information on the AP identifier, AP type, and backup means is registered by, for example, an administrator.

FIG. 6 is a diagram showing an example of data attribute management information (data attribute management table 600). A data attribute management table 600 is included in the data management DB 206 . The data attribute management table 600 is a table for managing data attributes. The data attribute management table 600 includes a data identifier field 601 , type field 602 and timestamp column field 603 .

A data identifier field 601 stores information indicating a data identifier that identifies data. The type field 602 stores information indicating the type of data. A timestamp column field 603 stores information indicating the timestamp in the data. As for the type of data, it is managed whether the specifications for adding, updating, or deleting data are write-once or update.

FIG. 7 is a diagram showing an example of a flowchart relating to data provision processing in which the data management program 111 provides data in response to a data utilization request.

In step S<b>701 , the data management program 111 receives a data usage request from the client terminal 140 . The usage request includes the data name (data identifier) of the data requested by the user, the time range for data acquisition (target time), the required level of data accuracy, and the time allowed to wait until data is provided (possible waiting time). time) and other information. Additionally, the usage request is not limited to including all of the above information. For example, if there is only one type of data to be requested, the data identifier may not be included.

An example (screen) of the user interface will be described later with reference to FIG. The required level requires accuracy with respect to the difference between the requested target time and the time of data that can be provided, and is selected from two levels (low, high), for example. As the possible waiting time, the time (for example, within one hour) that the user desires to receive the data is specified.

In step S702, the data management program 111 searches the data management table 400 for the data name (data identifier) requested by the user.

In step S703, the data management program 111 determines whether or not data of the requested data name (data identifier) exists (whether or not there is a matching data name). , and if it is determined that it does not exist, the process moves to step S711.

In step S704, the data management program 111 determines whether the target time of the requested data is date and time specification or period specification (whether date and time is specified) in order to extract data candidates that meet the conditions of the requested data. If it is determined that the date and time are specified, the process proceeds to step S705, and if it is determined that the period is specified, the process proceeds to step S706.

In step S705, the data management program 111 executes data identification processing at the time of specifying the date and time, and shifts the processing to step S707. The details of the data specifying process when specifying the date and time will be described later with reference to FIG.

In step S706, the data management program 111 executes data identification processing when specifying a period, and moves the processing to step S707. The details of the data specifying process when specifying a period will be described later with reference to FIG. 9 .

In step S707, the data management program 111 performs processing (request level confirmation processing) for confirming whether or not data satisfying the requested level can be provided when the requested level is specified in the usage request. Details of the request level confirmation process will be described later with reference to FIG. 12 .

In step S708, the data management program 111 calculates the time required for restoration of the identified data candidate, and checks whether the requested possible waiting time is satisfied. (restoration time confirmation process). Details of the restoration time confirmation process will be described later with reference to FIG. 13 .

In step S709, the data management program 111 performs provision form determination processing. In addition, in the provision form determination process, the provision form of the data to be provided is determined, and the details will be described later with reference to FIG. 14 .

In step S710, the data management program 111 restores the data and creates data in the provision form. For data restoration, for example, the data management program 111 identifies backup means (for example, a DB instance) based on the AP management table 500 and restores data using the identified backup means.

In step S711, the data management program 111 outputs the result of the utilization request, and terminates the data provision process. For example, the data management program 111 outputs, as a result, a method of accessing data in the form of provision. Note that an output example will be described later with reference to FIG. 15 .

Here, if the data that matches the time of the requested data is after the latest backup data, it means that there is no data that matches the requested data in the backup data. In that case, a snapshot of the data volume in the primary storage may be obtained and stored. As for the acquired data, the acquisition time, storage destination, etc. may be managed in the same manner as backup data.

FIG. 8 is a diagram showing an example of a flowchart relating to the data specifying process (step S705) when the date and time are specified.

In step S801, the data management program 111 searches the data management table 400 for backup data whose acquisition date and time (date and time) match the specified date and time.

In step S802, the data management program 111 determines whether or not there is a match as a result of the search. Transfer.

In step S803, the data management program 111 identifies the corresponding backup data, and shifts the process to step S805.

In step S804, the data management program 111 identifies the backup data whose backup acquisition date and time is closest to the specified date and time, and shifts the process to step S805.

In step S805, the data management program 111 determines whether or not the selected backup data (the data) is differential data (backup data obtained by differential backup). The process moves to step S806, and when it is determined that the data is not difference data, the process moves to step S807.

In step S806, the data management program 111 identifies the original full backup data, and terminates the data identification processing when the date and time are specified.

In step S807, the data management program 111 determines whether or not the selected backup data (data) is incremental data (backup data obtained by incremental backup). When the process is shifted to S808 and it is determined that the data is not the incremental data, the data is the backup data of the full backup, so the data specifying process at the time of designating the date and time ends.

In step S808, the data management program 111 identifies the full backup data and the incremental data from the full backup data to the data, and terminates the data identification processing at the date and time designation.

In this way, the data identification unit 304, for example, based on the data management information, when it determines that there is stored data corresponding to the data identifier included in the request (for example, "Yes" in step S703), the request (eg, step S802). When determined, the storage data with the closest date and time to the requested time included in the request is identified (eg, step 804).

FIG. 9 is a diagram showing an example of a flowchart relating to the data specifying process (step S706) when specifying a period. The data specifying process when specifying the period will be explained with reference to FIG. 10 when appropriate.

FIG. 10 is a conceptual diagram of selection of backup data when a period is specified. FIG. 10A is a diagram showing an example of retrieving backup data for a requested target period with respect to the backup acquisition time on the time axis. FIG. 10(B) shows an example of initial timestamp 408 and final timestamp 409 that can be managed as part of data management table 400 in conjunction with the particular example of backup in (A).

In step S901, the data management program 111 determines whether or not the requested data (target data) is of the write-once type. If so, the process moves to step S908.

In step S902, the data management program 111 determines whether or not there is a time stamp column in the target data. .

In step S903, the data management program 111 identifies the latest backup data (here, backup data A) before the start date and time when the first time stamp of the data management table 400 is requested. In the example of FIG. 10, the target period is from 4/25 to 5/10, so the backup data of the record 1001 whose first time stamp is before the start date and time 4/24 and is the latest 4/20 is specified.

In step S904, the data management program 111 determines whether or not the final time stamp of the specified backup data is after the requested end date and time. If it is determined that it is not after that, the process moves to step S905. In the example of FIG. 10, it is determined that the final timestamp (2017/4/29) of record 1001 is not after the requested end date and time (2017/5/10).

In step S905, the data management program 111 repeats confirmation of the backup data that was backed up after the backup data until the final time stamp is after the requested end date and time.

In step S906, the data management program 111 identifies backup data (here, backup data B) whose last time stamp is after the end date and time. In the example of FIG. 10, record 1003 is information of backup data having the latest end time stamp (May 13, 2017) after the requested end date and time (May 10, 2017).

In step S907, the data management program 111 specifies all the backup data from the first selected backup data A to the backup data B including the end date and time, and shifts the process to step S911. In the example of FIG. 10, the backup data of record 1002 between records 1001 and 1003 is specified. Note that the data management program 111 specifies only the backup data A when the final time stamp of the first specified backup data A is after the end date and time.

In step S908 (if the data is update type, or if the data is write-once type and there is no timestamp column), the data management program 111 determines that the backup acquisition date and time is before the requested start date and time, and The backup data (backup data C) with the closest date and time to the start date and time is specified.

In step S909, the data management program 111 identifies the backup data (backup data D) whose backup acquisition date and time is after the requested end date and time and is closest to the end date and time.

In step S910, the data management program 111 identifies all backup data from the identified backup data C to backup data D as targets.

In step S911, the data management program 111 determines whether or not the specified backup data (specific data) has differential data. The process moves to S913.

In step S912, the data management program 111 identifies the original fullback data, and terminates the data identification processing when specifying a period.

In step S913, the data management program 111 determines whether or not the identified backup data (specific data) has incremental data. If yes, the process proceeds to step S914. Ends data identification processing at the time of designation.

In step S914, the data management program 111 identifies the full backup data and the incremental data from the full backup data to the data, and terminates the data identification processing when specifying a period.

In this way, for example, based on the data management information, the data identification unit 304 determines that there is stored data corresponding to the data identifier included in the request (for example, "Yes" in step S703). It is determined whether or not the stored data has timestamp information (for example, step S902). Identify the first stored data that is previous and latest (for example, step S903), and identify the second stored data whose final time stamp is later than the end date and time included in the request (for example, step S903). Step S906), specify all the stored data from the first stored data to the second stored data (eg, step S907).

Further, for example, the data identification unit 304 determines whether or not there is backup data matching the requested time included in the request based on the data management information (for example, steps S703, steps S801 to S804, steps S901 to step S910), if it is determined that the backup data exists, and if the backup data acquisition unit is a differential backup, the full backup data of the most recent full backup of the backup data is specified (for example, steps S806 and S912), and the backup data When the unit of data acquisition is incremental backup, the full backup data of the most recent full backup of the backup data and the incremental backup data from the full backup data to the backup data are specified (for example, steps S808 and S914). .

FIG. 11 is a diagram showing an example of storage device management information (storage device management table 1100). A storage device management table 1100 is included in the data management DB 206 . The storage device management table 1100 includes a storage identifier field 1101, type field 1102, protocol field 1103, disk type field 1104, and performance field 1105. FIG.

The storage identifier field 1101 stores information indicating a storage identifier that identifies the storage device 132 . The type field 1102 stores information indicating the access type. The protocol field 1103 stores information indicating the corresponding protocol. The disk type field 1104 stores information indicating the type of the constituting disk. The performance field 1105 stores information indicating disk performance (for example, transfer speed). These may be set by an administrator, or may be acquired from a storage management tool and registered.

Furthermore, a volume identifier field 1106 may be provided, and information indicating the volume identifier may be registered together with the type of the corresponding disk when the volume is created.

FIG. 12 is a diagram showing an example of a flowchart relating to the requested level confirmation process (step S707) for confirming whether or not data that satisfies the requested level can be provided.

In step S1201, the data management program 111 calculates the time difference from the target time for the specified backup data. For example, when the target time of the requested data is specified by date and time, the data management program 111 calculates the difference between the acquisition date and time of the specified backup data and the specified date and time. In addition, for example, when the target time is specified for a period, the data management program 111 calculates the difference between the specified start date and time and the first date and time of the specified backup data (if there is no timestamp column, the backup acquisition date and time), and , the difference between the designated end date and time and the last date and time of the specified backup data is calculated, and the difference is taken as the total.

In step S1202, the data management program 111 acquires the requested level from the usage request (the input usage condition).

In step S1203, the data management program 111 determines whether or not the time difference satisfies the required level. . For example, the request level is set in two stages, low and high, and a threshold for determination is provided for each level. If the request level is "low", the first threshold is higher than if the time difference is "high", and if it is "high", the second threshold is lower than if the time difference is "low". value. The settings may be defined in advance as information of the data management program 111, defined by the administrator, or defined by the user. In this case, the data management program 111 determines whether or not the required level is satisfied by determining whether the time difference is within a threshold value corresponding to the required level.

In step S1204, the data management program 111 determines that the required level determination result is "satisfactory", and terminates the required level confirmation process.

In step S1205, the data management program 111 sets the determination result of the required level to "unsatisfactory", and terminates the required level confirmation process.

If the required level is not satisfied here, the data management program 111 may perform restoration from the data update record (for example, journal file) as additional processing. As an additional process, there is a method of using a backup of past update records of data. The data management program 111 determines whether or not a backup of the past update record of the data is saved and is available, and if it is not available, the request level determination result is "unsatisfactory". If the data update record is available, the data management program 111 designates the time period that is the difference between the date and time of the specified backup data and the target date and time (the end date and time if the period is specified), and performs the update record. Search if there is As a result of the search, if there is an update record in the time period of the difference, the update record is added as target data so as to be restored. In this case, the data management program 111 recalculates the difference between the latest date and time of the update record to be applied and the target date and time, and determines the request level.

FIG. 13 is a diagram showing an example of a flowchart relating to the restoration time confirmation process (step S708) for confirming whether the requested possible waiting time is satisfied.

In step S1301, the data management program 111 calculates the restoration time for the identified backup data. In the calculation process, for example, the data management program 111 reads the storage location and size from the backup information managed by the data management table 400 for the full backup data among the backup data specified as targets in the data specifying process. The data management program 111 determines the disk used by the volume from the configuration information of the storage device 132 (for example, if there is a volume identifier field 1106, the storage device management table 1100), and stores the backup data from the storage device management table 1100. The disk performance information is read out of the specifications of the storage device 132 . The data management program 111 calculates the time (read time) for reading the data of the corresponding size from the corresponding disk (data size/IOPS (Input/Output Per Second)). For example, the data management program 111 calculates the transfer time of the backup data from the storage device 132 in which the restoration data is stored to the management server 110 and adds it to the read time to obtain the restoration time. Further, for example, when restoring data to be stored in a predetermined server, the data management program 111 sets the transfer time of the backup data from the storage device 132 storing the restored data to the management server 110, and the predetermined time from the management server 110. The transfer time of the restored data to the server is calculated and added to the readout time to obtain the restoration time.

Note that the processor usage rate of the storage device 132 may be weighted in the calculation. Also, the specified backup data is calculated in the same way for the differential backup or the incremental backup.

Here, a method using past history may be used to calculate the restoration time. When the backup data is restored and provided, the execution time required for restoration is measured and managed as a history (stored in the secondary storage device 203). If the backup data specified as the target exists in the history, the measured execution time is read and set as the restoration time. If restoration is executed multiple times, all execution times may be read out and a representative value (for example, average value) may be calculated.

In step S1302, the data management program 111 acquires the possible waiting time from the usage request (use condition of the input data).

In step S1303, the data management program 111 determines whether or not the calculated restoration time is within the possible waiting time. The process moves to step S1305.

In step S1304, the data management program 111 determines that the restoration time determination result is "satisfactory", and terminates the restoration time confirmation process.

In step S1305, the data management program 111 sets the restoration time judgment result as "unsatisfactory", and terminates the restoration time confirmation process.

Here, if the required level is not satisfied, as additional processing, the data management program 111 may perform processing to determine whether or not another backup data can be used as an alternative. In the addition process, the data management program 111 determines whether or not the backup data initially identified as the target includes differential backup data or incremental backup data. When the data management program 111 determines that differential backup data or incremental backup data is included, it recalculates the time difference and restoration time with respect to the full backup data. The data management program 111 determines whether or not the recalculated restoration time satisfies the required level, and if it determines that it satisfies the requirement level, the data management program 111 excludes the differential backup data or the incremental backup data from the target data, and sets the determination result as "satisfactory". do.

FIG. 14 is a diagram showing an example of a flowchart relating to the provision form determination process (step S709) for determining the form of provision of data (form of provision). In the providing form determination process, the providing form is determined based on the size (capacity) of the provided data and the network performance.

In step S1401, the data management program 111 estimates the data size (size of restored data) when restored from the size of the backup data to be provided.

In step S1402, the data management program 111 estimates transfer time (for example, transfer time from a predetermined server to the client terminal 140) from the estimated data size and network performance.

In step S1403, the data management program 111 determines whether or not the estimated transfer time is equal to or less than the threshold. If it is determined that it exceeds, the process moves to step S1405.

In step S1404, the data management program 111 determines the form of provision as a file (provided in file format), and terminates the form of provision determination processing. In this case, the data management program 111 writes the data to a file and transfers it to a predetermined server (a server to which the user can access the file).

In step S1405, the data management program 111 determines that the provision form is a virtual machine (provided in the form of a virtual machine), and terminates the provision form determination process. In this case, the data management program 111 constructs an AP environment including the data restored to the virtual machine, accesses the AP of the virtual machine to make the data available, and outputs access information to the virtual machine in outputting the result. will do.

Also, the data management program 111 may determine the provision form according to the security policy of the provided data. A security policy is set by an administrator. When performing access control as a security policy, the administrator can set a form in which provided data is arranged and controlled in the administrator side environment, a form in which access to the AP is controlled in the form of a virtual machine, and the like.

Also, an interface may be provided for the user to specify the form of provision, in which case the data management program 111 will provide the data in the specified form.

FIG. 15 is a diagram showing an example of an input/output interface (input/output screen 1500). An upper area 1501 of the input/output screen 1500 is an area for the user to input a usage request (use conditions), and a lower area 1502 is an area for outputting the result of the usage request.

Input information includes a data name 1503 , requested data target time (specified period 1504 or date and time specified 1505 ), request level 1506 , and possible waiting time 1507 . The user selects and sets a period designation 1504 or a date and time designation 1505 as the target time of the requested data. Here, as a case where the user sets the date and time designation 1505, for example, there is a case where the data at the time point is reproduced and the test is performed. Further, as a case where the user sets the period designation 1504, for example, there is a case of performing analysis using data updated within the period.

Output information includes a provided data name 1508, a providing form 1509, an access destination 1510 indicating an access method, and a requirement satisfaction level 1511 indicating whether or not requirements are satisfied. As additional information on the requirement satisfaction level, the difference in time and the time until provision are output.

The requirement satisfaction level consists of a process of determining whether the required level is satisfied from the time difference of the provided data for the target period (Fig. 12) and a process of determining whether the calculated restoration time satisfies the possible waiting time ( It is determined based on the result of FIG. 13). Here, for example, three levels are provided for the requirement satisfaction level. If both judgment results are satisfied, the level is "high". If both judgment results are unsatisfactory, the level is set to "low".

In response to a usage request, the requirement satisfaction level (difference in time, time until provision) of the specified backup data before restoring the data may be output, and the restoration may be performed after confirmation by the user.

As described above, according to the configuration described above, in response to a request from the user, it is possible to specify the optimum backup data from the stored backup data and provide the data to the user. For example, even if the time section specified in the data usage request is different from the date and time of backup acquisition, data that meets the usage request can be extracted from the existing backup data. Also, for example, as auxiliary information indicating whether the usage request is satisfied, difference information from the specified date and time (e.g., time difference) and estimated preparation time (e.g., time until provision) are added, and the result is output. make it possible to Such time difference allows the user to grasp the accuracy of the data provided. According to the time until provision, the user can grasp the time until the data is provided.

According to the present embodiment, by managing time information, storage location information, etc., for storage data stored in various types of storage devices at various timings, appropriate storage data that matches the unit of use can be obtained. It allows us to provide in small units. In addition, in response to a usage request for data generated by a business application, stored data stored in a separate storage device can be used to specify and provide stored data that meets the usage requirements. It becomes possible to avoid the influence on the operation.

(2) Second Embodiment In this embodiment, a case will be described in which a plurality of pieces of data are included in a unit requesting data. In this embodiment, a configuration different from that of the first embodiment is mainly described.

FIG. 16 shows an example of business data management information (business data management table 1600) managed in this embodiment in addition to the management information of the first embodiment. A business data management table 1600 is included in the data management DB 206 . The business data management table 1600 includes a business identifier field 1601 , data item name field 1602 , AP field 1603 , and intra-AP data identifier field 1604 .

The business identifier field 1601 stores information indicating a business identifier that can identify a business. The data item name field 1602 stores information indicating the item name of data. AP field 1603 stores information identifying the AP and its type. The intra-AP data identifier field 1604 stores information that indicates an identifier that identifies data within the AP (data identifier that identifies business-related data).

In this embodiment, the user designates a business name (business identifier) when entering usage conditions. The data management program 111 searches the business data management table 1600 to identify a matching business identifier. The data management program 111 identifies the data item name corresponding to the business identifier, the AP, and the intra-AP data identifier. The data management program 111 searches the data management table 400 using the identified intra-AP data identifier as a data name (data identifier) (step S702), and similarly performs subsequent processing. Designation of the time requirement is the same as in the first embodiment.

In this way, the data identification unit 304 identifies the data identifier corresponding to the business identifier included in the request based on the business data management information, and stores the data corresponding to the identified data identifier based on the data management information. Determine whether or not there is data.

As a result, even if the data request unit is business and multiple data are included, the user can specify the business name to provide the data at the required time for each of the multiple data. Become.

Next, a process when a plurality of data backups for a usage unit are stored in storage devices 132 with different specifications will be described.

In the example of FIG. 16, it is assumed that the data corresponding to the same business identifier are from different APs and the backup data are stored in different storage devices 132 . Here, refer to the specifications of the storage device 132 registered in the storage device management table 1100 shown in FIG. 11 of the first embodiment. For example, it is assumed that the backup data of the customer DB is stored in the storage A, and the backup data of the inventory DB is stored in the storage B. In this case, the SDS system 130 performs protocol conversion for different protocols when retrieving backup data, and retrieves it in a common protocol (same format). For example, the backup data of the customer DB stored in the block storage is taken out as a file, and the backup data of the stock DB stored in the file storage is taken out as a file as it is. By retrieving with a common protocol, it is possible to collectively provide multiple pieces of data required for each business.

(3) Third Embodiment In the present embodiment, a method for controlling placement of backup data in the storage device 132 will be described. In this embodiment, a configuration different from that of the above-described embodiment will be mainly described.

The SDS system 130 virtualizes and manages various storage devices 132, provides tiers according to disk performance of the storage devices 132, and provides a tier control function (hierarchy control unit) for migrating data between tiers. The management server 110 (data migration unit not shown in the data management program 111) requests the hierarchical control function of the SDS system 130 to migrate data to the storage destination of the backup data based on the usage history of the backup data.

FIG. 17 is a diagram showing an example of storage tier management information (storage tier management table 1700). A storage tier management table 1700 is included in the data management DB 206 . The storage tier management table 1700 includes a level field 1701 , a storage identifier field 1702 , a disk field 1703 , a performance field 1704 and a free space field 1705 .

A level field 1701 stores information indicating the level of the hierarchy. The storage identifier field 1702 stores information indicating a storage identifier that identifies the storage device 132 to be managed. The disk field 1703 stores information indicating the type of disk. The performance field 1704 stores information indicating disk performance (eg, transfer rate (MB/s)). The free capacity field 1705 stores information indicating the free capacity of the disk.

The assignment of storage devices 132 and disks to levels is set in advance by an administrator or system. Regarding performance and free space, the management server 110 periodically acquires and registers information from the SDS system 130 or each storage device 132, and for example, average transfer speed and free space at the time of acquisition are managed. Hierarchies are, for example, three hierarchies of high, medium, and low in descending order of performance, which are managed in the level field 1701 as levels 1, 2, and 3, respectively.

The management server 110 (not shown storage unit of the data management program 111) measures the execution time until restoration when providing data using backup data, and manages (stores in the secondary storage device) as a history. In this embodiment, the management server 110 also records (stores) the possible waiting time, which is a usage condition when using backup data, as a history. The management server 110 manages the history used for providing each piece of backup data. Even if the size of the provided data differs according to the period request, the management server 110 records the time taken to restore the entire backup data from the backup. Note that the record is registered as use history management information in the present embodiment.

FIG. 18 is a diagram showing an example of usage history management information (usage history management table 1800). A usage history management table 1800 is included in the data management DB 206 . The usage history management table 1800 includes a backup identifier field 1801 , a usage date/time field 1802 , a restoration time field 1803 and a possible waiting time field 1804 . Note that the recording is not limited to the output (registration) to the table, and may be the output to a log file or the like.

The backup identifier field 1801 stores information indicating a backup identifier (storage identifier) that identifies backup data. The date and time of use field 1802 stores information indicating the date and time used for restoration. The restore time field 1803 stores information indicating the time required to restore the backup data. Possible wait time field 1804 stores information indicating the possible wait time specified in the request.

In the management server 110, when the frequency of use of each piece of backup data reaches a predetermined level (for example, once/week) or more, the average requested waiting time and the average restoration time are calculated. Calculate the time difference of the restoration time with respect to the waiting time. If the difference is equal to or greater than a preset upper limit reference value (the restoration time is longer than the possible waiting time), the management server 110 selects the tier in which the backup data is stored. Refer to the table 1700 to check the free space. The management server 110 instructs the SDS system 130 to migrate the data to the tier when there is free space equal to or greater than the size of the backup data (see the data management table 400).

If the calculated difference is equal to or less than a preset lower limit reference value (restoration time is shorter than the possible waiting time), the management server 110 selects the tier in which the backup data is stored. The tier management table 1700 is referenced to check the free space of the storage device 132 (disk) of the tier. In addition, the management server 110 refers to the performance (transfer speed) of the storage device 132 of the relevant tier, calculates the restoration time in the case of restoration from the size of the backup data (see the data management table 400), and calculates the average possible waiting time. Make sure it fits in time. The management server 110 migrates the data to the storage device 132 of the tier if there is free space and the restoration time is within the time limit.

In this way, the data migration unit, for example, based on the storage device management information and the information (usage history management information) stored in the storage unit, restores the stored data whose restoration time is longer than the possible waiting time. The process of migrating the saved data to a disk with higher performance than the disk that stores the saved data, and/or the time required for restoration is shorter than the waiting time. Perform migration processing.

In this embodiment, storage data is arranged appropriately so as to satisfy the user's request, and efficient use of the storage device 132 is made possible.

(4) Fourth Embodiment In this embodiment, a method of controlling backup acquisition based on the usage history of backup data will be described. In this embodiment, a configuration different from that of the above-described embodiment will be mainly described.

The SDS system 130 provides a volume snapshot function and a copy function for copying a volume to a local device or another device. The backup execution frequency, storage destination, etc. are set in the backup management means (software etc.) by the administrator, and the backup management means executes the snapshot function or copy function of the SDS system 130 based on the settings. The management server 110 (not shown processing unit of the data management program 111) periodically determines the backup data storage destination and full backup acquisition interval from the usage history, and sends a request to the backup execution means. By doing so, control related to backup acquisition is performed.

The management server 110 calculates the usage frequency and the average restoration time from the usage history of the backup data shown in the third embodiment in determining the storage destination. The management server 110 selects the highest tier (level 1) as the storage destination when the usage frequency is higher than the reference value and the average restoration time is higher than the reference value. In addition, if the frequency of use is equal to or greater than the reference value and the average restoration time is less than the reference value, or if the frequency of use is less than the reference value and the average restoration time is equal to or greater than the reference value, the management server 110 hierarchy (level 2). Also, the management server 110 selects the lowest tier (level 3) when the frequency of use is less than the reference value and the average restoration time is less than the reference value. If there are multiple storage devices 132 in the selected tier (level), the management server 110 selects, for example, the storage device 132 with the largest free space that is larger than the size of the backup data as the storage destination. If there is no storage device 132 whose free space is larger than the size of the backup data, the management server 110 may select from the storage device 132 in the lower tier.

Note that the management server 110 may calculate the average restoration time from the usage history of the backup data and determine the storage destination without using the usage frequency.

Furthermore, the management server 110 determines the backup data acquisition interval based on the usage history. The management server 110 calculates the frequency of use for each piece of backup data. Here, differential backup data and incremental backup data related to full backup data are identified based on the unit of backup data acquisition.

When the frequency of use of full backup data is equal to or less than the reference value, the management server 110 lengthens the interval between full backup data, differential backup data, or incremental backup data. As the lengthening rate, a predetermined value may be set, or the rate may be determined according to the frequency of use.

The management server 110 shortens the full backup acquisition interval when the frequency of use of the differential backup data or the incremental backup data is equal to or higher than the reference value and the average restoration time from the differential backup data or the incremental backup is equal to or higher than the reference value. A predetermined value may be set for the rate of shortening, or the differential backup or incremental backup interval may be shortened together as the differential backup or incremental backup interval. In comparing the usage frequencies, the usage frequency of only full backup data and the usage frequency of differential backup data or incremental backup data may be compared. If differential backup data or incremental backup data is used more frequently than full backup only, set the above.

When the frequency of use of differential backup data or incremental backup data is equal to or greater than the reference value and the average restoration time from the differential backup or incremental backup is less than the reference value, the management server 110 shortens the differential backup or incremental backup acquisition interval. . In this case, do not change the full backup interval.

Otherwise, do not change the acquisition interval for full backup, differential backup, or incremental backup.

Here, for the reference value, a value set in advance by an administrator or system is used. For the reference value for the average recovery time, the average possible waiting time requested from the usage history may be used.

In this way, the processing unit changes the storage destination of the stored data based on, for example, the representative value of the time required to restore the stored data calculated from the information stored in the storage unit (for example, the average restoration time). and/or change the interval for acquiring the stored data based on the frequency of use of the stored data calculated from the information (usage history management information) stored in the storage unit.

According to the present embodiment, it is possible to control the acquisition of stored data so that the accuracy of data can be improved in consideration of the restoration time required for provision in accordance with the usage status of stored data.

It should be noted that in the third and fourth embodiments, it is necessary to satisfy the required restoration time even for data recovery in the event of a failure, which is the original purpose of the backup data. Therefore, the administrator may set for each AP whether or not the management server 110 may control the arrangement and acquisition frequency of backup data. In that case, the above control is executed only when the control is permitted.

(5) Other Embodiments In the above-described embodiments, the case of applying the present invention to a data management system was described, but the present invention is not limited to this, and various other systems and devices can be used. , methods and programs.

In the above-described embodiment, for convenience of explanation, XX tables and XX files have been used to describe various types of data, but the data structure is not limited and may be expressed as XX information or the like.

In the above description, information such as programs, tables, and files that implement each function are stored in memory, hard disks, SSD (Solid State Drives), and other storage devices, or IC cards, SD cards, DVDs, and other recording media. can be placed in

Moreover, the above-described configurations may be appropriately changed, rearranged, combined, or omitted within the scope of the present invention.

100... Data management system, 110... Management server, 120... Host computer, 130... SDS system, 140... Client terminal.

Claims (10)

A data management system comprising a plurality of storage devices,
a request receiving unit that receives a request containing information indicating the requested time, information indicating a data identifier for identifying the requested data, and request level information regarding the accuracy of the requested target ;
based on data management information in which information indicating a time when storage data for storing data is acquired and information indicating a storage device in which the storage data is stored are associated with each other; a data identification unit that identifies, from among the stored data stored, storage data matching the requested time included in the request received by the request reception unit;
calculating a time difference between the storage data specified by the data specifying unit and the requested time included in the request based on the data management information, and calculating the request level included in the request based on the calculated time difference; a data evaluation unit that evaluates whether or not the
an output unit that outputs a result of identification by the data identification unit;
A data management system comprising: The data management information further includes information indicating an identifier that identifies data in association with it,
Information indicating a date and time for requesting data is specified as information indicating the time of the request target included in the request,
The request further includes information indicating an identifier that identifies the requested data,
When the data identification unit determines that there is storage data of data corresponding to the data identifier included in the request based on the data management information, the storage data matching the requested time included in the request does not exist. Determine whether or not it exists, and if it is determined that it exists, specify the stored data, and if it is determined that it does not exist, specify the stored data that has the closest date and time to the requested time included in the request. ,
The data management system according to claim 1, characterized by: The data management information further includes information indicating an identifier for identifying data, and information of an initial time stamp that is the first time stamp in the stored data and a final time stamp that is the last time stamp in the stored data. is associated with
Information of a start date and time and an end date and time indicating a period for requesting data is specified as information indicating the requested time included in the request,
The request further includes information indicating an identifier that identifies the requested data,
The data identification unit determines whether or not the stored data of the data has time stamp information when determining that there is stored data of the data corresponding to the data identifier included in the request based on the data management information. is determined, and when it is determined that there is time stamp information, the first stored data whose initial time stamp is before the start date and time included in the request and is the latest is specified, and Identifying second stored data whose final time stamp is later than the end date and time included in the request, and identifying all stored data from the first stored data to the second stored data;
The data management system according to claim 1, characterized by: In the data management information, acquisition indicating a full backup for backing up all data, a differential backup for backing up the difference from the full backup, or an incremental backup for backing up the changed portion from the previous backup, as a unit for acquiring stored data. Unit information is further associated and included,
Based on the data management information, the data specifying unit determines whether or not there exists backup data matching the requested time included in the request. when it is a backup, full backup data of the most recent full backup of the backup data is specified, and when the acquisition unit of the backup data is an incremental backup, full backup data of the most recent full backup of the backup data; identifying incremental backup data from said full backup data to said backup data;
The data management system according to claim 1, characterized by: The request further includes information indicating a data identifier that identifies the data to be requested and information on possible waiting time of the requested data,
a data evaluation unit that calculates the time required to restore the stored data identified by the data identification unit and evaluates whether the calculated restoration time is within the possible waiting time included in the request;
The data management system according to claim 1, characterized by: Information indicating an identifier for identifying data is further associated with the data management information,
The request further includes information indicating a business identifier that can identify the business,
The data identification unit is configured to, based on business data management information in which information indicating a business identifier capable of identifying a business and information indicating a data identifier identifying data related to the business are associated with each other, identifying a data identifier corresponding to the business identifier that is stored, and determining whether or not there is stored data corresponding to the identified data identifier based on the data management information;
The data management system according to claim 1, characterized by: The output unit uses a different protocol for the storage data specified by the data specifying unit based on storage device management information in which information that can identify a storage device and information indicating a protocol of the storage device are associated. If it is identified from the storage device of , convert the protocol and output as stored data in the same format
The data management system according to claim 1, characterized by: The request includes information indicating a data identifier that identifies requested data and information indicating a possible waiting time for the requested object,
information indicating a storage identifier that identifies the storage data specified by the data specifying unit; information indicating the time required for the output unit to restore the storage data; information indicating a possible waiting time included in the request; a storage unit that associates and stores
Storage device management information in which information that can identify a storage device, information that indicates a disk that stores storage data provided in the storage device, and information that indicates the performance of the disk are associated, and stored in the storage unit. Based on this information, the time required for restoration is longer than the possible waiting time, and/or the process of migrating the saved data to a disk with higher performance than the disk storing the saved data, and/or the time required for restoration. a data migration unit that performs a process of migrating the stored data to a disk having performance lower than the disk storing the stored data whose time is shorter than the possible waiting time;
The data management system according to claim 1, characterized by: Information indicating a storage identifier that identifies the storage data identified by the data identification unit, information indicating the time required for the output unit to restore the storage data, and information indicating the time when the storage data was acquired a storage unit that associates and stores
a process of changing the storage destination of the stored data based on a representative value of the time required to restore the stored data calculated from the information stored in the storage unit, and/or A processing unit that performs a process of changing the interval for acquiring stored data based on the frequency of use of stored data calculated from information,
The data management system according to claim 1, characterized by: A data management method in a data management system including a plurality of storage devices, comprising:
a first step in which the request receiving unit receives a request containing information indicating the requested time, information indicating a data identifier identifying the requested data, and request level information regarding the accuracy of the requested target ;
based on the data management information in which the information indicating the time when the storage data for storing the data was acquired and the information indicating the storage device in which the storage data is stored are associated with each other; a second step of identifying, from among stored data stored in a plurality of storage devices, stored data that matches the requested time included in the request received by the request receiving unit;
A data evaluation unit calculates, based on the data management information, a time difference between the storage data identified by the data identification unit and a requested time included in the request, and the calculated time difference is included in the request. a third step of evaluating whether or not the included requirement level is met;
a fourth step in which an output unit outputs a result of identification by the data identification unit;
A data management method comprising:

Images