JP2010044571A - Database device, data management method and program therefore - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a database device which prevents an increase in costs due to requirement for storing/searching large-scale data, and increases efficiency in storage/search processing. <P>SOLUTION: A data group to be less frequently accessed is usually stored in a backup database 106, and restored in a restoration database 108 for use as necessary. The backup database 106 is equipped with a storage medium whose storage capacity can be increased as a storage medium, while the restoration database 108 is equipped with only an expensive storage medium such as a high-speed disk device whose storage capacity cannot be easily increased. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a database apparatus, a data management method, and a program for storing a plurality of partitions in which a storage area is logically divided and storing different data groups in each of the plurality of partitions.

The amount of information handled by public institutions and companies is increasing dramatically year by year. A database is used as a system capable of storing and managing such information, and the techniques of Patent Documents 1 to 3 are disclosed as related database techniques. Here, Patent Document 1 discloses a technique for rearranging data in an added database storage area when adding a database storage area as the database capacity increases. A method of using a hash map for data division is shown. Patent Document 2 discloses a technique for reducing the recovery process at the time of abnormal termination while preventing the existing data from being affected when performing an addition process to a large-scale database. Patent Document 3 discloses a method of dividing / reorganizing a data storage area with high access frequency.
JP 2000-276382 A JP 2001-142752 A JP-A-6-139119

  Here, the problem in storing / retrieving a large amount of information is that when the amount of stored data is large, the data storage cost increases and the processing time of the storage / retrieval increases. In addition, when the amount of data is large, the backup and recovery processing time as a countermeasure against the failure also increases, and the operation cost increases.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a database apparatus, a data management method, and a program thereof that can prevent an increase in cost for a large-scale data storage / retrieval request and increase the efficiency of the storage / retrieval process.

  In order to achieve the above object, the present invention provides a database device having a plurality of partitions in which a storage area is logically divided and storing different data groups in each of the partitions, and the access frequency is a threshold value. A low access frequency storage means having a plurality of partitions each storing less than a data group and capable of increasing storage capacity, and a high access frequency having a plurality of partitions storing data groups each having an access frequency equal to or greater than the threshold value A storage means, a restored data storage area having a restored data storage area for temporarily storing data stored in a partition of the low access frequency storage means, a storage capacity limited storage data, identification information of the partition, and The high access frequency storage means of the partition or the low access frequency Identifying identification information of the partition that performs processing related to the new processing target data based on the data input of the new processing target based on the partition position storage unit that stores the storage position in association with any of the storage units Partition identification information specifying means, and when the specified partition identification information indicates the storage location of the low access frequency storage means, the data group in the corresponding partition of the low access frequency storage means based on the storage location Low access data group restoring means for temporarily restoring to the restored partition of the restored data storage area, and the high access data storage area of the high access frequency storage means or the restoration of the restored data storage means based on the specified partition identification information Any partition of the data storage area To a database system, characterized in that it comprises a storage data processing means for performing processing on the data of the new processing target.

  Further, according to the present invention, the database apparatus described above includes an access history storage unit that stores a partition that has been accessed based on the processing related to the new data to be processed, and data in the partition that has performed the access. Backup means for backing up only the group to a backup device.

  Further, the present invention is a database device that has a plurality of partitions in which the storage area is logically divided, and stores a different data group in each of the plurality of partitions, based on a new data input to be processed, Partition identification information specifying means for specifying identification information of a partition performing processing related to the new processing target data, and the new processing target data for the partition of the data storage means based on the specified partition identification information A storage data processing means for performing processing relating to, an access history storage means for storing a partition that has been accessed based on the processing relating to the new data to be processed, and a backup device that stores only a data group in the accessed partition Backup to backup The stage is a database system, characterized in that it comprises a.

  Further, the present invention provides an access period determination means for determining whether or not the elapsed period from the last access to the partition exceeds a threshold of the elapsed period, and the elapsed period exceeds the threshold. In this case, it is characterized by comprising: batch deleting means for collectively deleting the partitions whose elapsed period exceeds a threshold value.

  According to the present invention, in the above-described database device, the storage data processing means may be configured to execute parallel processing on the plurality of partitions when there are a plurality of pieces of identification information of the partition that performs processing related to the new processing target data. In addition, a process related to the new data to be processed is performed.

  Further, according to the present invention, the above-described database device temporarily records the data in the processing data storage unit based on the new data to be processed, and the stored data processing unit stores the processing data storage The processing relating to the new data to be processed temporarily recorded in the means is performed.

  The present invention is also a data management method in a database apparatus that has a plurality of partitions in which a storage area is logically divided, and stores different data groups in each of the plurality of partitions, the storage capacity of the database apparatus being The low access frequency storage means that can increase the volume has a plurality of the partitions each storing a data group whose access frequency is less than a threshold value, and the high access frequency storage means of the database device is a data group whose access frequency is the threshold value or more. A plurality of partitions for storing data, and a restoration data storage means of the database device has a restoration data storage area for temporarily storing data stored in the partition of the low access frequency storage means, The partition location storage means of the database device is The partition identification information and the storage location of the partition in either the high access frequency storage means or the low access frequency storage means are stored in association with each other. Based on the data input of the processing target, identification information of the partition that performs processing related to the new processing target data is specified, and the low-access data group restoration unit of the database device determines that the specified partition identification information is the low When the storage location of the access frequency storage means is indicated, the data group in the corresponding partition of the low access frequency storage means is temporarily restored to the restoration partition of the restoration data storage area based on the storage location, and the storage of the database device The data processing means Based on the partition identification information, a process related to the new data to be processed is performed on any partition of the high access data storage area of the high access frequency storage means or the restored data storage area of the restored data storage means This is a database management method characterized by the above.

  According to the present invention, in the database management method described above, the access history recording unit of the database device records, in the access history storage unit, a partition that is accessed based on the process related to the new processing target data. The backup unit of the database apparatus backs up only the data group in the accessed partition to the backup apparatus.

  The present invention also includes a low access frequency storage means having a plurality of partitions each storing a data group having an access frequency less than a threshold and capable of increasing the storage capacity, and a plurality of data groups each storing the data group having the access frequency equal to or greater than the threshold A high-access-frequency storage unit having a partition, a restored-data storage unit having a limited storage capacity, and a restored-data storage area that temporarily stores data stored in the partition of the low-access frequency storage unit, Each of the partitions, comprising partition identification information and partition location storage means for storing the partition in association with the storage location of either the high access frequency storage means or the low access frequency storage means of the partition, A database device that stores different data groups. A partition identification information identifying means for identifying identification information of the partition that performs processing related to the new data to be processed based on data input of the new processing target, and the identified partition identification information is the low access frequency A low access data group restoring means for temporarily restoring a data group in a corresponding partition of the low access frequency storage means to a restored partition of the restored data storage area based on the storage position when indicating a storage position of the storage means; Based on the identified partition identification information, the processing related to the new data to be processed for any partition in the high access data storage area of the high access frequency storage means or the restored data storage area of the restored data storage means Storage data processing means for performing Is a program to function.

  In addition to the above-described means, the present invention further includes an access history storage means for storing the computer of the database device, and further storing, among the partitions, a partition that has been accessed based on a process related to the new data to be processed. This is a program for causing only the data group in the accessed partition to function as backup means for backing up to a backup device.

  According to the present invention, the data group accessed at low frequency is normally stored in the low access frequency storage means, and is restored to the restored data storage means and used only when necessary. The low access frequency storage means is a database that can be provided with a storage medium that can increase the storage capacity as a storage medium, and the restored data storage means cannot be increased easily because of high cost such as a high-speed disk device. Only a storage medium is provided. As a result, a large amount of infrequently accessed data is recorded in a low access frequency storage means having a recording medium that can be easily increased at a low cost, and is restored to the restored data storage means only when necessary. Since the capacity of the access frequency storage means is kept below a certain capacity, it is not necessary to increase the amount of expensive recording media, and the cost for managing enormous data can be reduced.

  Further, according to the present invention, the access history storage means stores the access presence / absence information for the partition storing the accessed data (data update or the like) by changing the access status from “No” to “Yes”. . Then, the backup / recovery processing unit reads the partition name for which access status information is registered as “Yes” from the partition management table held by the partition management unit, and stores the data in the partition corresponding to the partition name to the backup device. Write process. As a result, it is not necessary to back up the data of all the partitions stored in the high access frequency storage unit, so that the time and capacity related to the backup processing can be reduced.

  Further, according to the present invention, in the data search, when a plurality of search target information is recorded in the data search start request, a plurality of partition key values are specified, and thereby search processing is performed from a plurality of different partitions. You may need to do it. In such a case, the data search processing unit searches each of the plurality of partitions in parallel. As a result, when searching from an enormous amount of data, the search is performed in parallel from each partition divided into multiple data, compared to when searching sequentially from one partition. The search process can be performed in a short time.

Hereinafter, a database apparatus according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of the database apparatus according to the embodiment.
In this figure, reference numeral 1 denotes a database device. The database apparatus 1 includes a data update unit 101, a partition management unit 102 (partition location storage unit, access history storage unit), an update work data processing unit 103 (temporary recording unit), and a high access frequency database 104 (high access frequency). Storage means), a backup / recovery processing unit 105 (backup means), a backup database 106 (low access frequency storage means), a data restoration processing unit 107, a restoration database 108 (restored data storage means), and a data search processing unit 109. Yes.

  Then, the data update unit 101 performs a process of updating the newly received data to the database. Further, the partition management unit 102 stores information for specifying a database having a partition that is a data storage destination, information on a partition that has been updated such as data writing, and the like. The update work data processing unit 103 temporarily stores data to be processed by the data update unit 101 in the database, and stores the data in the database. The high access frequency database 104 stores data that is frequently accessed, the backup database 106 stores both data that is accessed frequently or infrequently, and the restoration database 108 includes: It has a storage area where data accessed at low frequency is temporarily restored from the backup database 106.

  Each database in the database apparatus 1 according to the present embodiment has a plurality of partitions whose storage areas are logically divided, and stores a different data group in each of the plurality of partitions. The database device 1 is configured such that the backup database 106 mainly stores a plurality of the partitions storing data groups whose access frequencies are less than a threshold, and the storage capacity can be increased. The backup database is also used for backing up partitions that store frequently accessed data. In addition, the high access frequency database 104 has a plurality of partitions for storing data groups whose access frequencies are equal to or higher than a threshold value. Further, the restoration database 108 has a restoration data storage area for temporarily storing data stored in the partition of the backup database 106, and includes a storage device having a limited storage capacity. Further, the partition management unit 102 stores the partition identification information (partition ID) and the storage location of the partition in either the high access frequency database 104 or the restoration database 108 in association with each other.

  The update work data processing unit 103 (partition identification information specifying unit) or the data search processing unit 109 (partition identification information specifying unit) relates to the new processing target data based on the new processing target data input. Identify identification information (partition key value) of the partition to be processed. When the data restoration processing unit 107 (low access data group restoring means) determines that the specified partition identification information indicates the storage location of the backup database 106, the corresponding partition of the backup database 106 is based on the storage location. Is temporarily restored to the restoration partition of the data restoration processing unit 107. Further, the update work data processing unit 103 (stored data processing unit) or the data search processing unit 109 (stored data processing unit) performs the high access data storage area or restoration of the high access frequency database 104 based on the specified partition identification information. For any partition in the restored data storage area of the database 108, processing related to new data to be processed is performed.

Next, a data table stored in the partition management unit will be described.
FIG. 2 is a diagram illustrating an example of data in the partition management table.
The partition management table shown in FIG. 2 is stored by the partition management unit 102, and stores information such as partition name, partition key value, database type, and update presence / absence. Here, the partition key value is information for specifying a partition for storing data. For example, when the partition is divided as an area for storing data of the years 2008, 2007, and 2006. The AD information about the input data is acquired as a partition key value, and the partition is specified based on the partition key value. The database type is information for specifying a database (the high access frequency database 104 or the restoration database 108) having the storage area of the partition indicated by the corresponding partition name. The partition storing data with low access frequency is normally stored in the backup database 106. However, when accessing the data, the entire data in the partition storing the data is stored in the backup database 106. Are temporarily restored in the restoration database 108.

FIG. 3 is a diagram showing an example of data temporarily stored in the update work data processing unit.
As shown in the figure, the update work data processing unit 103 temporarily records new processing target data received by the data update unit 101. The example shown in FIG. 3 shows an example in which the new data to be processed is sales data. For example, the processing number (No), transaction date, supplier, sales representative, sold The product number, sales quantity, sales amount, and data updated flag are recorded in association with each other. Here, in the sales data as shown in FIG. 3, for example, the transaction date is used as the partition key value. When the partition recorded in each database is separated by the transaction date, the transaction date is read from the data to be processed as a partition key value, and the year is specified from the information. Then, processing (writing, updating, reading, etc.) is performed on the partition corresponding to the year. In addition, the partition key value may be the product number of the product, and the partition may be divided for each product number, or the partition key value may be information on the business partner, and the partition may be divided for each business partner. Note that because the data to be processed is temporarily stored in the database of the update work data processing unit 103, even if a failure occurs in processing such as writing or updating the database based on the data received as the processing target. The data can be easily recovered.

FIG. 4 is a diagram showing an outline of processing at the time of data access.
As shown in this figure, the update work data processing unit 103 performs processing related to data temporarily stored in a database held by itself. At that time, the partition key value is identified from the stored data, and the partition name recorded in the partition management table corresponding to the partition key value and the database having the storage area of the partition (high access frequency database) 104 or the restoration database 108). Then, the update work data processing unit 103 specifies a partition to be processed for the new data to be processed from the database destination having the specified partition name and the storage area of the partition, and the new work data is processed for the partition. Processing for new writing, updating, or reading of data to be processed is performed. The logical position in the database storage area corresponding to the partition name is stored in, for example, the partition management unit 102. Based on the information, the logical position in the database storage area is determined as the start position of the partition storage area. That's fine.

  In addition, the update work data processing unit 103, when the database having the partition name corresponding to the partition key value of the new data to be processed and the storage area of the partition is not recorded, Is stored in the backup database 106 as a data group with low access frequency, and a data restoration request is made to the data restoration unit 107. This data restoration request stores the partition key value of the new data to be processed. The data restoration processing unit 107 identifies a partition in the backup database 106 from the partition key value stored in the data restoration request, decompresses the compressed data group stored in the partition, and restores the data. By writing to the database 108, the data in the partition stored in the backup database 106 is restored in the restoration database 108.

  That is, the data group accessed at a low frequency is normally stored in the backup database 106 and is restored to the restoration database 108 and used only when necessary. Here, the backup database 106 is a database that can include a storage medium that can increase the storage capacity as a storage medium. For example, a plurality of magnetic tapes that can be increased in price at a low price can be provided. On the other hand, the high access frequency database 104 and the restoration database 108 include only a storage medium such as a high-speed disk device that is expensive and cannot be easily increased. A large amount of infrequently accessed data is recorded in the backup database 106 having a recording medium that can be easily increased at a low price as described above, and is restored to the restoration database 108 only when necessary. Since the capacity of the access frequency database 104 is kept below a certain capacity, it is not necessary to increase the amount of expensive recording media, thereby reducing the cost in managing enormous data.

  The high access frequency database 104 stores frequently accessed data. However, in order to avoid data damage due to a database failure, it is necessary to back up these data. Therefore, the backup / recovery processing unit 105 performs a process of writing the data stored in the partition in the high access frequency database 104 to the backup database 106. Here, the update work data processing unit 103 registers the update presence / absence of the partition storing the data subjected to the update process for the high access frequency database 104, so that the partition management table (FIG. 2) is updated. Update the information from “No” to “Yes”. Then, the backup / recovery processing unit 105 reads the partition name in which the update presence / absence information is registered as “present” from the partition management table (FIG. 2) held by the partition management unit 102, and stores the high name corresponding to the partition name. A process of writing data in the partition of the access frequency database to the backup database 106 is performed. As a result, it is not necessary to back up the data of all the partitions stored in the high access frequency database 104, so that the time and capacity related to the backup processing can be reduced.

FIG. 5 is a first diagram showing a processing flow of the database apparatus.
As shown in this figure, first, in the database apparatus 1, the data updating unit 101 receives an input of new data to be processed (step S501), and temporarily stores it in the database held by the update work data processing unit 103. (Step S502). Then, the update work data processing unit 103 identifies the partition key value from the processing target data that has received the input (step S503), and the partition management recorded in the partition management unit 102 in association with the partition key value. A partition name and a database type are acquired from the table (step S504). Then, the update work data processing unit 103 specifies the logical position of the storage area of the partition where the data to be processed is written or read based on the partition name and the database type. Based on the partition key value of the data to be processed, it is determined whether or not the partition has been specified (step S505). If the partition has been specified, the processing requested for the data to be processed is requested. (New writing, updating, reading, etc. of processing target data) is performed (step S506).

  If the partition cannot be identified based on the information recorded in the partition management table, it is determined that the data is the low access frequency data stored in the backup database 106, and the partition key value is stored in the data restoration processing unit 107. A request for restoring the stored data is made (step S507). The update work data processing unit 103 restores the data in the restoration database 108, and then stores the requested data in the restored partition (the new write / update of the processing target data, Reading etc.) (step S508). This processing includes new writing, updating, reading of processing target data in the partition. Then, when updating the data, the update work data processing unit 103 changes the information indicating whether or not the update is recorded in the partition management table in association with the name of the partition storing the updated data from “No” to “Yes”. To "" (step S509).

  Next, it is determined whether the processing related to the data to be processed has been completed (step S510). If the processing has been completed, the update work database 103 requests the backup / recovery processing unit 105 for backup processing (step S511). Then, the backup / recovery processing unit 105 reads the partition name and database type in which the update presence / absence information is “present” in the partition management table, and identifies the backup target partition based on the partition name and database type. (Step S512). Then, the backup / recovery processing unit 105 reads the data in the partition identified as the backup target from the high access frequency database 104 or the restoration database 108 and stores it in the backup database 106 (step S513). The backup / recovery processing unit 105 compresses data to be backed up.

  When the backup process is completed, the update work data processing unit 103 determines whether there is a request to delete all data in the partition from the data update unit 101 (step S514). The data in the partitioned partition is deleted from the high access frequency database 104 or the restoration database 108 for each partition (step S515). Then, the update work data processing unit 103 deletes the name of the deleted partition and the corresponding columns from the partition management table, and updates the partition management table stored in the partition management unit 102. If there is no data deletion request, the partition management unit 102 next determines whether the backup database 106 has a partition whose retention period has passed (step S516). If there is a partition whose retention period has passed in the backup database 106, the partition management unit 102 deletes the data in the partition for each partition from the backup database 106 (step S517). Note that the retention period may be determined by comparing whether the elapsed period from the date and time when the partition was created to the current date and time exceeds the partition valid period stored in the partition management unit 102.

FIG. 6 is a second diagram showing the processing flow of the database device.
Next, a processing flow of the database apparatus at the time of data search will be described using FIG.
First, the data search processing unit 109 receives a data search start request (step S601). Then, the data search processing unit 109 identifies the partition key value from the search target data that has received the input (step S602), and from the partition management table recorded in the partition management unit 102 in association with the partition key value. The partition name and the database type are acquired (step S603). Then, the update work data processing unit 103 identifies the logical position of the storage area of the partition in which the search target data is recorded, based on the partition name and the database type. Further, the data search processing unit 109 determines whether or not the partition can be specified based on the partition key value of the search target data (step S604), and when the partition can be specified, the search target data The search process is performed (step S605). If the partition cannot be identified based on the information recorded in the partition management table, it is determined that the data is the low access frequency data stored in the backup database 106, and the partition key value is stored in the data restoration processing unit 107. A request to restore the stored data is made (step S606).

  Then, the work data processing unit for update 103 restores the data in the restoration database 108, and then searches the data stored in the restored partition (step S607). Here, in the data search, when a plurality of pieces of search target information are recorded in the data search start request, there are a plurality of partition key values to be specified, which makes it necessary to perform a search process from a plurality of different partitions. There is a case. In such a case, the data search processing unit 109 searches each of the plurality of partitions in parallel. As a result, when searching from an enormous amount of data, the search is performed in parallel from each partition divided into multiple data, compared to when searching sequentially from one partition. The search process can be performed in a short time. Then, the search processing unit 109 outputs the search result (step S608).

  In the database device 1 described above, an example in which one high access frequency database 104 and one restoration database 108 are provided has been described. However, depending on the amount of data, the high access frequency database 104 and A plurality of restoration databases 108 may be provided. In this case, the update work data processing unit 103 performs processing on any one of the high access frequency database 104 or the restoration database 108 stored in the partition, based on the specified partition key value. Do. Also, when data with low access frequency is restored from the backup database 106, the data restoration processing unit 107 performs partition restoration processing on any one of the restoration databases 108.

  Although the embodiment of the present invention has been described above, according to the above-described processing, the data group accessed at low frequency is normally stored in the backup database 106 and restored to the restoration database 108 only when necessary. Used. The backup database 106 is a database that can be provided with a storage medium that can increase the storage capacity as a storage medium, and the restoration database 108 is only a storage medium that is expensive and cannot be easily increased, such as a high-speed disk device. It is equipped with. As a result, a large amount of infrequently accessed data is recorded in the backup database 106 having a recording medium that can be increased easily at a low cost, and is restored to the restoration database 108 only when necessary. Since the capacity of 104 is kept below a certain capacity, it is not necessary to increase the amount of expensive recording media, and the cost for managing enormous data can be reduced.

  Further, according to the present invention, the update work data processing unit 103 changes the update management information in the partition management table from update “No” to “Yes” for the partition that stores the updated data, and stores it. is doing. Then, the backup / recovery processing unit 105 reads the partition name in which the update presence / absence information is registered as “present” from the partition management table (FIG. 2) held by the partition management unit 102, and stores the partition name corresponding to the partition name. Is written into the backup database 106. As a result, it is not necessary to back up the data of all the partitions stored in the high access frequency database 104, so that the time and capacity related to the backup processing can be reduced.

  Further, according to the present invention, in the data search, when a plurality of search target information is recorded in the data search start request, a plurality of partition key values are specified, and thereby search processing is performed from a plurality of different partitions. You may need to do it. In such a case, the data search processing unit 109 searches each of the plurality of partitions in parallel. As a result, when searching from an enormous amount of data, the search is performed in parallel from each partition divided into multiple data, compared to when searching sequentially from one partition. The search process can be performed in a short time.

  The database apparatus described above has a computer system inside. Each process described above is stored in a computer-readable recording medium in the form of a program, and the above process is performed by the computer reading and executing the program. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

  The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

It is a block diagram which shows the structure of a database apparatus. It is a figure which shows the example of data of a partition management table. It is a figure which shows the example of data temporarily stored in the work data processing part for an update. It is a figure which shows the process outline at the time of data access. It is a 1st figure which shows the processing flow of a database apparatus. It is a 2nd figure which shows the processing flow of a database apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Database apparatus 101 ... Data update part 102 ... Partition management part 103 ... Update work data processing part 104 ... High access frequency database 105 ... Backup / recovery processing part 106 ... Backup database 107: data restoration processing unit 108 ... restoration database 109 ... data search processing unit

Claims (10)

A database device having a plurality of partitions in which a storage area is logically divided, and storing different data groups in each of the plurality of partitions,
A low access frequency storage means having a plurality of partitions each storing a data group whose access frequency is less than a threshold, and capable of increasing the storage capacity;
High access frequency storage means having a plurality of the partitions each storing a data group having the access frequency equal to or higher than the threshold;
A restored data storage means having a limited storage capacity, having a restored data storage area for temporarily storing data stored in the partition of the low access frequency storage means;
Partition position storage means for storing the partition identification information and the storage position of the partition in either the high access frequency storage means or the low access frequency storage means in association with each other;
Partition identification information specifying means for specifying identification information of the partition for performing processing related to the new processing target data based on the new processing target data input;
When the specified partition identification information indicates the storage location of the low access frequency storage means, the data group in the corresponding partition of the low access frequency storage means is used as the restoration partition of the restoration data storage area based on the storage location. A low-access data group restoring means for temporarily restoring;
On the basis of the specified partition identification information, with respect to any partition of the high access data storage area of the high access frequency storage means or the restored data storage area of the restored data storage means, the new processing target data Stored data processing means for performing processing;
A database apparatus comprising: Among the partitions, an access history storage unit that stores a partition that has been accessed based on the processing related to the new processing target data;
Backup means for backing up only the data group in the accessed partition to a backup device;
The database apparatus according to claim 1, further comprising: A database device having a plurality of partitions in which a storage area is logically divided, and storing different data groups in each of the plurality of partitions,
Partition identification information specifying means for specifying identification information of a partition that performs processing related to the new processing target data based on the new processing target data input;
Based on the specified partition identification information, storage data processing means for performing processing related to the new data to be processed for the partition of the data storage means;
Access history storage means for storing a partition that has been accessed based on the processing related to the new processing target data;
Backup means for backing up only the data group in the accessed partition to a backup device;
A database apparatus comprising: Access period determination means for determining whether or not the elapsed period from the last access to the partition exceeds a threshold of the elapsed period;
When the elapsed period exceeds the threshold, batch deletion means for collectively deleting the partitions whose elapsed period exceeds the threshold;
The database apparatus according to any one of claims 1 to 3, further comprising: The storage data processing means, when there are a plurality of pieces of identification information of the partition that performs processing related to the new processing target data, processes related to the new processing target data in parallel to the plurality of partitions. The database apparatus according to any one of claims 1 to 4, wherein: Temporary recording means for temporarily recording the data in the processing data storage means based on the new processing target data input;
The database apparatus according to any one of claims 1 to 5, wherein the storage data processing unit performs processing related to the new processing target data temporarily recorded in the processing data storage unit. A data management method in a database apparatus that has a plurality of partitions in which a storage area is logically divided and stores different data groups in each of the plurality of partitions,
The low access frequency storage means capable of increasing the storage capacity of the database device has a plurality of the partitions each storing a data group whose access frequency is less than a threshold value,
The high access frequency storage means of the database device has a plurality of the partitions each storing a data group having the access frequency equal to or higher than the threshold,
The restoration data storage means of the database device has a restoration data storage area for temporarily storing data stored in the partition of the low access frequency storage means,
The partition position storage means of the database device stores the identification information of the partition in association with the storage position of either the high access frequency storage means or the low access frequency storage means of the partition,
The partition identification information specifying means of the database device specifies the identification information of the partition that performs processing related to the new processing target data based on the new processing target data input,
When the specified partition identification information indicates the storage location of the low access frequency storage unit, the low access data group restoration unit of the database device stores the location in the corresponding partition of the low access frequency storage unit based on the storage location. Temporarily restore the data group to the restoration partition of the restoration data storage area,
Based on the specified partition identification information, the storage data processing means of the database device is assigned to any partition in the high access data storage area of the high access frequency storage means or the restored data storage area of the restored data storage means. On the other hand, the database management method characterized by performing the process regarding the said new process target data. The access history recording means of the database device records, in the access history storage means, a partition that is accessed based on the processing related to the new processing target data among the partitions,
The database management method according to claim 7, wherein the backup unit of the database device backs up only the data group in the accessed partition to the backup device. A low access frequency storage means having a plurality of partitions each storing a data group whose access frequency is less than a threshold, and capable of increasing the storage capacity;
High access frequency storage means having a plurality of partitions each storing a data group with the access frequency equal to or higher than the threshold;
A restored data storage means having a limited storage capacity, having a restored data storage area for temporarily storing data stored in the partition of the low access frequency storage means;
Each partition comprising: partition identification information; and partition location storage means for storing the partition in association with either the high access frequency storage means or the low access frequency storage means of the partition. A database device computer for storing different data groups,
Partition identification information specifying means for specifying identification information of the partition that performs processing related to the new processing target data based on the data input of the new processing target;
When the specified partition identification information indicates the storage location of the low access frequency storage means, the data group in the corresponding partition of the low access frequency storage means is used as the restoration partition of the restoration data storage area based on the storage location. Low access data group restoration means for temporary restoration,
On the basis of the specified partition identification information, with respect to any partition of the high access data storage area of the high access frequency storage means or the restored data storage area of the restored data storage means, the new processing target data Stored data processing means for processing,
Program to function as. In addition to each means of claim 9, the database device computer further comprises:
An access history storage means for storing a partition that has been accessed based on the processing related to the new data to be processed among the partitions;
Backup means for backing up only the data group in the accessed partition to a backup device,
Program to function as.

Images