JP5810982B2 - SEARCH DEVICE, SEARCH METHOD, AND SEARCH PROGRAM - Google Patents

Description

  The present invention relates to a search device, a search method, and a search program for searching for information.

  In recent years, with the expansion of the scale of systems and the diversification of business operations, it has been required to efficiently operate a wide variety of data using a storage for storing data. For example, as a technique related to a plurality of storages, there is one that facilitates monitoring of the usage status of a plurality of storage systems by integrating and collecting volume information of a plurality of storage systems (see, for example, Patent Document 1 below). .)

JP 2005-267501 A

  However, in the above-described prior art, when the system is operated using a plurality of data stores to which storage is allocated, the migration load increases when data stored in a specific data store is migrated to another data store. . For example, in one data store of a plurality of data stores, the storage position in the data stored in the remaining data store is stored, and the device that executes the migration process stores the data in the specific data store among the remaining data stores. Suppose you want to migrate stored data to another data store. At this time, as part of the migration process, there is a process for changing the storage position in the data stored in one data store, and the amount of change increases when there is a lot of data accumulated in a specific data store. Therefore, the transition load becomes large.

  An object of the present invention is to provide a search device, a search method, and a search program that can reduce the migration load in order to solve the above-described problems caused by the prior art.

  In order to solve the above-described problems and achieve the object, according to one aspect of the present invention, a search request for information that matches a search condition among information groups having a specific attribute is received and specified by the specific attribute. A first identifier indicating a storage location of the value or a second identifier for identifying a specified value, the identifier specified based on information having at least a specific attribute is stored under control of the first data store; The first data store is searched for information that matches the search condition with reference to the storage unit 1 and based on the information that matches the searched search condition. When an identifier used to access a second storage unit that stores a value specified by control of a different second data store is determined, and the determined identifier is the second identifier, the second identifier And an identifier indicating the storage location of the specified value based on the information indicating the relationship between the storage location of the specified value and the second identifier, and the identifier or the first identifier indicating the generated storage location A search device, a search method, and a search program for accessing a specified value using are proposed.

  According to one aspect of the present invention, it is possible to reduce the transition load.

FIG. 1 is an explanatory diagram illustrating an operation example of the search device. FIG. 2 is an explanatory diagram showing a connection example of the data store integrated system. FIG. 3 is an explanatory diagram illustrating a hardware configuration example of the search device. FIG. 4 is a block diagram illustrating an example of functions of the search device. FIG. 5 is an explanatory diagram showing an example of the contents stored in the data store group. FIG. 6 is an explanatory diagram showing an example of the contents stored in the storage method definition rule table. FIG. 7 is an explanatory diagram of an example of the contents stored in the identifier generation rule table. FIG. 8 is an explanatory diagram of an example of the contents stored in the search situation table. FIG. 9 is an explanatory diagram illustrating an operation example of registration processing. FIG. 10 is an explanatory diagram showing an example of data to be subjected to identifier deletion processing. FIG. 11 is an explanatory diagram illustrating an operation example of the search process. FIG. 12 is an explanatory diagram illustrating an operation example of the migration process. FIG. 13 is a flowchart illustrating an example of an identifier deletion processing procedure. FIG. 14 is a flowchart (part 1) illustrating an example of a search processing procedure. FIG. 15 is a flowchart (part 2) illustrating an example of a search processing procedure. FIG. 16 is a flowchart (part 1) illustrating an example of the migration processing procedure. FIG. 17 is a flowchart (part 2) illustrating an example of the migration processing procedure.

  Exemplary embodiments of a disclosed search device, search method, and search program will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is an explanatory diagram illustrating an operation example of the search device. A data store integration system 100 that integrates a plurality of data stores includes a search device 101 that searches for life log data, a data store that controls a storage area for storing life log data, data store # 0, and data store #. 1 and data store # 2. The data store integration system 100 also has an identifier generation rule table 111. The life log data is data in which human life, behavior, experience, etc. are recorded as digital data such as video / audio / position information. In the present embodiment, the data store integration system 100 accumulates four attributes of life log data: date, number of steps, lunch comment, and lunch photo.

  Data store # 0 controls the search index of life log data. The search index indicates an index for searching life log data. Specifically, the type of the data store # 0 is RDB (Relational DataBase). The storage area controlled by the data store # 0 stores records 112-1 to 112-7. Note that the records 112-1 to 112-3 are records with a high search frequency. A record with a high search frequency has a small number of records. Further, it is assumed that the records 112-4 to 112-7 are records with a low search frequency. A record with a low search frequency has a large number of records. The reason why the search index record has such a search frequency and the number of records will be described later with reference to FIG.

  The type of the data store # 1 is KVS (Key Value Store). The data stored in the storage area controlled by the data store # 1 will be described later with reference to FIG. Furthermore, the type of data store # 2 is a file system. The storage area controlled by the data store # 2 stores the image file file # 99, the image file file # 98, and the image file file # 88 in the data directory. Details of the type of data store will be described later with reference to FIG.

  Of the four attributes of the life log data, values specified by the date attribute and the step count attribute are stored in a storage area controlled by the data store # 0. The value designated by the lunch comment attribute is stored in the storage area controlled by the data store # 1. The value specified by the lunch photo attribute is stored in a storage area controlled by the data store # 2. Details of the storage contents of the storage areas controlled by the data store # 0 to the data store # 2 will be described later with reference to FIG.

  Further, the storage area controlled by the data store # 0 has a field for storing an identifier indicating a storage position of a value specified by the lunch comment attribute and a value specified by the lunch photo attribute. Hereinafter, this identifier is referred to as a data store dependent identifier. Further, the storage area controlled by the data store # 0 has a field for storing an identifier for identifying a value designated by the lunch comment attribute and a value designated by the lunch photo attribute. Hereinafter, this identifier is referred to as a data store independent identifier. The same identifier is used for the value specified by the lunch comment attribute and the value specified by the lunch photo attribute as the data store-independent identifier shown in the present embodiment, but different identifiers may be prepared.

  The identifier generation rule table 111 is information indicating the relationship between the data store-independent identifier and the storage position of the value specified by the attribute. In the example of FIG. 1, the identifier generation rule table 111 stores information indicating the relationship between the data store-independent identifier and the storage position of the value specified by the lunch photo attribute stored in the storage area controlled by the data store # 2. doing. Specifically, the identifier generation rule table 111 indicates that the value specified by the lunch photo attribute is stored at the position indicated by “/ data / file # {identifier}” using the identifier field. .

  First, the search device 101 receives a search request 121. The search condition of the search request 121 is that the number of steps is 5000 and the date is 2011/7/1 to 2011/8/31. Next, the search device 101 searches for a record that matches the search condition of the search request 121. In the example of FIG. 1, the records that match the search condition of the search request 121 are a record 112-1 and a record 112-4.

  Subsequently, for the records 112-1 and 112-4 that match the search condition, the search apparatus 101 determines an identifier used for accessing the data store # 2 among the data store dependent identifier and the data store independent identifier. Since the search frequency of the record 112-1 is high, the search apparatus 101 accesses the storage area controlled by the data store # 2 using the data store dependent identifier “/ data / file # 99”, and the image file file # 99. To get.

  Further, since the search frequency of the record 112-4 is low, the search apparatus 101 uses the identifier “88” and “/ data / file # {identifier}” in the identifier generation rule table 111 to indicate the storage location. “/ Data / file # 88” is generated. Subsequently, using the identifier “/ data / file # 88”, the search device 101 accesses the storage area controlled by the data store # 2 and acquires the image file file # 88.

  As described above, the search apparatus 101 assigns the data store-dependent identifier indicating the storage position of the attribute in the data store to the information with high search frequency, and the data store independent that becomes the attribute identifier for the information with low search frequency. An identifier is assigned to generate a storage location. As a result, the search apparatus 101 only needs to change the data store-dependent identifier of a record with a high search frequency in the attribute migration process, and can reduce the load of the migration process. Hereinafter, the search device 101 will be described in detail with reference to FIGS.

  FIG. 2 is an explanatory diagram showing a connection example of the data store integrated system. The data store integration system 100 includes a client 201, a client 202, a search device 101, and a data store group 203. The client 201, the client 202, and the search device 101 are connected via a network 210.

  The client 201 may be a smartphone, a PHS (Personal Handyphone System), or a tablet terminal as a mobile phone. The client 202 is a desktop PC (Personal Computer). In response to a search request from application software running on the client 201 or the client 202, the search apparatus 101 acquires information corresponding to the search condition and returns it to the application software. The application software is hereinafter referred to as “application”.

  The data store group 203 includes data store # 0 to data store # 2. A data store is software that stores data in a storage area in a predetermined format. As shown in FIG. 1, the type of the data store # 0 is RDB. The type of data store # 1 is KVS. Furthermore, the type of data store # 2 is a file system.

  The RDB accumulates structured information in a table, specifies conditions regarding the value of each column, and reads the information from the table. RDB excels at complex search processing and result tabulation. The KVS associates a value with a key serving as an identifier, and acquires a corresponding value by designating the key. KVS can easily improve scalability by adding nodes. The file system reads and writes content by designating a file name as an identifier. The file system may have a large content size. In addition, the file system can easily improve scalability by adding nodes.

  As a data store other than RDB, KVS, and file system, for example, there is a full text search engine. The full-text search engine accumulates unstructured information and searches by specifying a field value or a character string included in the text. The unstructured information is, for example, a field that differs for each information, a plurality of values in one field, or text information.

  Further, since the data store is software, the device that executes the data store # 0 to the data store # 2 may be the search device 101 or an external device. If it is an external device, for example, the external device executes data store # 0 to data store # 2 and controls the storage devices managed by data store # 0 to data store # 2. The apparatuses that execute data store # 0 to data store # 2 may be the same or different.

  As shown in FIG. 2, the data store integration system 100 provides a service using life log data to the client 201 and the client 202 by combining a plurality of data stores. As a specific data store combination method, a storage area controlled by data store # 0, which is an RDB, is used as a search index. In addition, the storage areas controlled by the data store # 1 that is the KVS and the data store # 2 that is the file system store the optimum attribute of each data store among the attributes of the life log data. Which data store stores which attribute depends on the size of the value specified by the attribute, the accumulation amount of the entire attribute, the update frequency for the value, whether or not a search condition is specified, and the like. Will be determined.

  Although not shown in FIG. 2, the data store integration system 100 may include an administrator terminal operated by the administrator of the data store integration system 100. Alternatively, the administrator may log in to the data store integration system 100 from the client 201 or the client 202 using an administrator account.

(Hardware of search device 101)
FIG. 3 is an explanatory diagram illustrating a hardware configuration example of the search device. In FIG. 3, a search apparatus 101 includes a CPU (Central Processing Unit) 301, a ROM (Read-Only Memory) 302, a RAM (Random Access Memory) 303, a magnetic disk drive 304, a magnetic disk 305, an IF 306, and the like. ,including. Each unit is connected by a bus 307.

  Here, the CPU 301 is an arithmetic processing unit that controls the entire search apparatus 101. The ROM 302 is a nonvolatile memory that stores programs such as a boot program. A RAM 303 is a volatile memory used as a work area for the CPU 301. The magnetic disk drive 304 is a control device that controls reading / writing of data with respect to the magnetic disk 305 according to the control of the CPU 301. The magnetic disk 305 is a non-volatile memory that stores data written under the control of the magnetic disk drive 304.

  The IF 306 is connected to the network 210 through a communication line, and is connected to other devices via the network 210. The IF 306 controls an internal interface with the network 210 and controls data input / output from an external device. For example, a modem or a LAN adapter can be adopted as the IF 306. Note that the search device 101 may include an optical disc drive, a display, a keyboard, and a mouse when an administrator directly operates the search device 101.

  Although not shown in FIG. 3, the client 201 includes, for example, a CPU, a RAM, a ROM, a flash ROM, a display, and an IF. The flash ROM is a non-volatile memory in which stored contents can be rewritten. Further, the client 201 may include a camera device and a sensor. The client 202 may include, for example, a CPU, a RAM, a ROM, a magnetic disk drive, a magnetic disk, an IF, an optical disk drive, a display, a keyboard, and a mouse.

(Function of search device 101)
Next, functions of the search device 101 will be described. FIG. 4 is a block diagram illustrating an example of functions of the search device. The search apparatus 101 includes a reception unit 401, a search unit 402, a determination unit 403, a generation unit 404, an access unit 405, a deletion unit 406, an update unit 407, a change unit 408, an index adapter 409, A data store adapter 410 is included. The receiving unit 401 to the data store adapter 410 serving as the control unit realize their functions when the CPU 301 executes a program stored in the storage device. Specifically, the storage device is, for example, the ROM 302, the RAM 303, the magnetic disk 305, etc. shown in FIG. Alternatively, the function may be realized by being executed by another CPU via the IF 306.

  In addition, the search device 101 can access the data store group 203, the storage method definition rule table 411, the identifier generation rule table 111, and the search status table 412. The storage method definition rule table 411, the identifier generation rule table 111, and the search status table 412 are stored in a storage device such as the RAM 303 and the magnetic disk 305.

  In FIG. 4, the storage method definition rule table 411, the identifier generation rule table 111, and the search status table 412 are inside the search device 101, and the data store group 203 is outside the search device 101. The storage method definition rule table 411, the identifier generation rule table 111, and the search status table 412 may be outside the search device 101, and the data store group 203 may be inside the search device 101.

  The accepting unit 401 accepts a search request for information that matches a search condition among information groups having specific attributes. For example, the reception unit 401 receives a search request from an application that runs on the client 201.

  The accepting unit 401 may accept a search request for information whose value specified by the attribute as the search condition matches the search condition to be transferred. As a search condition to be transferred, for example, there may be a search condition that there is a value specified by the lunch photo attribute to be transferred, or there is a value specified by the lunch photo attribute, and the administrator The search conditions up to the past several years specified from may be used. With the function of the reception unit 401, the search device 101 can detect the trigger of search processing and migration processing. The accepted search request is stored in a storage area such as the RAM 303 and the magnetic disk 305.

  When the search unit 402 receives a search request by the receiving unit 401, the search unit 402 refers to the first storage unit and searches for information that matches the search condition. The first storage unit is, for example, a storage area controlled by data store # 0. For example, the search unit 402 searches for information that matches the search condition of the search request received from the application. With the function of the search unit 402, the search apparatus 101 can acquire records that match the search conditions. The search result is stored in a storage area such as the RAM 303 and the magnetic disk 305.

  The first storage unit is specified based on information having at least a specific attribute of the first identifier indicating the storage position of the value specified by the specific attribute or the second identifier for identifying the specified value. The identifier is stored under the control of the first data store. The first identifier is, for example, a data store dependent identifier. The storage location of the value is an identifier that can be interpreted by the data store that controls the first storage unit. For example, if the first data store is KVS, the first identifier is a key. In the case of a file system, the first identifier is a file path.

  In addition, since the 1st identifier which shows a storage position becomes unique according to a value, it changes for every value. In order to make the storage location different for each value, for example, the storage location may include a second identifier that identifies the value. Alternatively, the first identifier may include information obtained by converting the second identifier.

  In addition, when the first identifier is specified, the first storage unit stores the first identifier, and may or may not store the second identifier. In addition, when the second identifier is specified, the first storage unit stores the second identifier, and may or may not store the first identifier. Moreover, a specific identification method specifies, for example according to the date when the information which has a specific attribute was registered into the 1st memory | storage part, and the search frequency of the information which has a specific attribute.

  Further, the first storage unit is specified based on a comparison result between the date when the information having at least a specific attribute of the first identifier or the second identifier is registered in the first storage unit and the current date. The identifier is stored under the control of the first data store. For example, if the registered date and the current date are within one month, the first storage unit stores at least the first identifier. For example, if the registered date and the current date exceed one month, the first storage unit stores at least the second identifier. Specifically, if the current date is “2011/8/31” and the date when the information having the specific attribute is registered in the first storage unit is “2011/8/1”, the first date The storage unit stores at least the first identifier.

  Further, the first storage unit stores an identifier specified based on information having at least a specific attribute of the first identifier or the second identifier and a history of search conditions updated by the update unit 407. . The search condition history is stored in the search status table 412. For example, it is assumed that the number of times that a search condition for acquiring records within the past month is accepted as the search condition history is greater than a threshold specified by the administrator. At this time, when the information having the specific attribute is a record within the past one month, the first storage unit stores at least the first identifier. In addition, when the number of times the search condition for acquiring a record within the past month is received is equal to or less than a specified threshold and information having a specific attribute is a record within the past month, the first storage unit Stores at least the second identifier.

  The determination unit 403 is designated by control of a second data store different from the first data store among the first identifier and the second identifier based on information that matches the search condition searched by the search unit 402. An identifier used to access the second storage unit that stores the value to be stored is determined. The second storage unit is, for example, a storage area controlled by data store # 1, or a storage area controlled by data store # 2. As for the determination method as to which one of the first identifier and the second identifier is used, for example, the date registered in the first storage unit of the information that matches the search condition or the search condition is matched. It is specified according to the search frequency for information or the presence or absence of the first identifier.

  The determining unit 403 also determines the first identifier and the second based on the comparison result between the date when the information that matches the search condition searched by the search unit 402 is registered in the first storage unit and the current date. Of the identifiers, an identifier used for accessing the second storage unit may be determined. For example, if the registered date and the current date of the retrieved record are within one month, the determination unit 403 determines the identifier used for accessing the second storage unit as the first identifier. Further, for example, if the registered date and the current date of the retrieved record exceed one month, the determination unit 403 determines the identifier used for accessing the second storage unit as the second identifier. To do.

  In addition, when the information that matches the search condition searched by the search unit 402 has the first identifier, the determination unit 403 may determine the identifier used for accessing the second storage unit as the first identifier. Good. For example, if the retrieved record has the first identifier, the determination unit 403 determines the identifier used for accessing the second storage unit as the first identifier.

  Further, the determination unit 403 determines the second of the first identifier and the second identifier based on the information that matches the search condition searched by the search unit 402 and the history of the search condition updated by the update unit 407. An identifier for use in accessing the storage unit may be determined.

  For example, it is assumed that the number of times that a search condition for acquiring records within the past month is accepted as the search condition history is greater than a threshold specified by the administrator. At this time, if the retrieved record is a record within the past one month, the determination unit 403 determines the identifier used for accessing the second storage unit as the first identifier. In addition, if the number of times the search condition for acquiring records within the past month is received is equal to or less than the threshold and the searched record is a record within the past month, the determination unit 403 includes the second storage unit The identifier used for access to is determined as the second identifier.

  With the function of the determination unit 403, the search device 101 can use the first identifier that does not need to generate an identifier indicating the storage position if the search frequency is high, and if the search frequency is low, A second identifier that does not need to be changed at the time of migration can be used. The determination result is stored in a storage area such as the RAM 303 and the magnetic disk 305.

  When the identifier determined by the determination unit 403 is the second identifier, the generation unit 404 specifies based on the second identifier and information indicating the relationship between the storage location of the specified value and the specified value. An identifier indicating the storage location of the value to be generated is generated. Information indicating the relationship between the storage location of the second identifier and the specified value is information stored in the identifier generation rule table 111.

  For example, the information indicating the relationship between the storage positions of the specified values is specific character information, and an identifier indicating the storage position of the specified value is obtained by replacing a part of the character information with the second identifier. Information. Specifically, the information indicating the relationship between the storage positions of designated values is “/ data / file # {identifier}”. By replacing {identifier} with the second identifier, the generation unit 404 generates an identifier indicating the storage position of the designated value. The information indicating the relationship between the storage positions of the specified values is specified by combining the second identifier with the information indicating the relationship between the storage positions of the specified values, for example, in addition to the replacement. It may be information that can generate an identifier indicating the storage location of a value. In this case, information indicating the relationship between storage positions of designated values is “/ data / file #”.

  Further, the generation unit 404 may generate an identifier indicating a storage position in the third storage unit of a value specified by control of a third data store different from the second data store. Specifically, the generation unit 404 stores information in the third storage unit from the second identifier and information indicating the relationship between the storage position of the designated value in the third storage unit and the second identifier. An identifier indicating the position may be generated.

  For example, it is assumed that the information indicating the relationship between the storage positions in the third storage unit is “/ data / usr1 / file- {identifier}” and the second identifier is “99”. The generation unit 404 generates an identifier indicating a storage position in the third storage unit, such as “/ data / usr1 / file-99”.

  In addition, there may be a case where the information that matches the search condition searched by the search unit 402 does not include the second identifier. However, in this case, the information that matches the search condition has the first identifier. Therefore, the generation unit 404 extracts the second identifier from the first identifier, information indicating the relationship between the extracted value and the storage position of the second identifier and the specified value in the third storage unit, From the above, an identifier indicating the storage position in the third storage unit may be generated. With the function of the generation unit 404, the search device 101 can access a specified value.

  The access unit 405 accesses the designated value using the identifier indicating the storage position generated by the generation unit 404 or the first identifier. For example, when the generation unit 404 generates “/ data / file # 88”, which is an identifier indicating the storage position, the access unit 405 accesses the value stored in “/ data / file # 88”.

  In addition, the access unit 405 reads a value designated from the second storage unit using the identifier indicating the storage position in the second storage unit generated by the generation unit 404 or the first identifier. Subsequently, the access unit 405 writes the read value to the third storage unit using the identifier indicating the storage position in the third storage unit generated by the generation unit 404. For example, the access unit 405 reads a value designated by using the identifier “/ data / file # 99” from the second storage unit. Subsequently, the access unit 405 writes the value designated by using the identifier “/ data / usr1 / file-99” in the third storage unit. With the function of the access unit 405, the search device 101 can access the values that are the targets of the search process or the migration process.

  The deletion unit 406 deletes the first identifier included in the information having the specific attribute based on the information having the specific attribute stored in the first storage unit. For example, the deletion unit 406 deletes the first identifier according to the date registered in the first storage unit of information having a specific attribute or the search frequency for information having a specific attribute. Specifically, if the date on which the information having the specific attribute is registered and the current date are separated from one month, the deletion unit 406 deletes the first identifier of the information having the specific attribute.

  When deleting the first identifier in a state where there is no second identifier in the information having the specific attribute, the deletion unit 406 extracts the second identifier included in the first identifier, and extracts the extracted value. After writing to the first storage unit, the first identifier is deleted. With the function of the deletion unit 406, the search device 101 can reduce the storage amount of the first storage unit.

  When the receiving unit 401 receives a search request, the updating unit 407 updates the search condition history. In the search condition history, the first identifier is deleted. For example, when the reception unit 401 receives a search request having a search condition of within the past month, the update unit 407 increments the number of times of reception of the search condition within one month. With the function of the update unit 407, the search device 101 can use the search status actually searched.

  When the identifier determined by the determination unit 403 is the first identifier, the changing unit 408 transmits the first identifier stored in the first storage unit to the third storage unit generated by the generation unit 404. The identifier is changed to indicate the storage location. For example, the changing unit 408 changes the first identifier “/ data / file # 99” stored in the first storage unit to the identifier “/ data / usr1 / file-99” generated by the generating unit 404. . With the function of the changing unit 408, the search device 101 can access the value written in the migration destination data store.

  The index adapter 409 is software that absorbs differences in protocols and data formats for each search index, and enables access to the data store # 0 that is the search index with the same API (Application Programming Interface). The data store adapter 410 is software that absorbs the difference in the protocol and data format for each data store and enables access to the data store # 1 and the data store # 2 with the same API.

  FIG. 5 is an explanatory diagram showing an example of the contents stored in the data store group. The data store integrated system 100 shown in the present embodiment accumulates life log data. The storage area controlled by the data store # 0 shown in FIG. 5 stores records 112-1 to 112-7. Further, the storage area controlled by the data store # 1 shown in FIG. 5 stores records 501-1 to 501-3. Further, the storage area controlled by the data store # 2 shown in FIG. 5 stores the image file file # 99 and the image file file # 98. Furthermore, it is assumed that the records 112-1 to 112-3 are records having a high search frequency. Further, it is assumed that the records 112-4 to 112-7 are records with a low search frequency.

  The storage area controlled by the data store # 0 has five fields: date, number of steps, identifier, lunch comment, and lunch photo. The date field stores the date when the target record was registered by the application by the user's operation or the date when the sensor registered the information. Note that the date field may include a time. The number of steps field stores the number of steps of the user on the date specified in the date field. The identifier field is a data store-independent identifier that identifies the values specified by the lunch comment field and the lunch photo field. Specifically, the identifier field is an identifier stored in the lunch comment field and a serial number of the identifier stored in the lunch photo field. Specifically, the identifier field is a numerical value numbered from 1.

  The lunch comment field is an identifier indicating a storage position of character information that becomes a user's comment on lunch on the day specified in the date field. Specifically, the character information is stored in a storage area controlled by the data store # 1. The lunch photo field is an identifier indicating storage information of image information showing a lunch on the date specified in the date field. Specifically, the image information is stored in the data store # 2.

  The storage area controlled by the data store # 1 has two fields: a key serving as an identifier and a lunch comment corresponding to the key. In the KVS specification, an arbitrary character string can be set in the key field. In the example of the present embodiment, it is assumed that the same character string as the lunch comment field in the corresponding record in the storage area controlled by the data store # 0 is set in the key field. In the lunch comment field, character information that becomes a user's comment on lunch is set. For example, the record 501-1 indicates that the key is “key # 99” and the lunch comment is “delicious”.

  The storage area controlled by the data store # 2 has a directory structure. The storage area controlled by the data store # 2 shown in FIG. 5 stores image information showing the user's lunch in the “data” directory. The image information is, for example, an image file file # 99 and an image file file # 98. The identifier of the image file file # 99 is “/ data / file # 99” which is a file path. The identifier of the image file file # 98 is “/ data / file # 98” which is a file path.

  For example, the record 112-1 indicates that the step count attribute of the life log data in “2011/8/31” is 5000 steps. Further, the record 112-1 indicates that the character information that is the value specified by the lunch comment attribute is “delicious” than the record 501-1 indicated by the lunch comment field “key # 99” that is a data store-dependent identifier. Show. The record 112-1 indicates that the image information having a value specified by the lunch photo attribute is an image file file # 99 having a file path indicated by the lunch photo field “/ data / file # 99”.

  FIG. 6 is an explanatory diagram showing an example of the contents stored in the storage method definition rule table. The storage method definition rule table 411 illustrated in FIG. 6 stores records 411-1 to 411-3. Note that the information for one record in the storage method definition rule table 411 is referred to as storage method definition rule information. The storage method definition rule table 411 includes two fields: an attribute name and a storage destination. The attribute name field stores the attribute name of life log data. The storage field stores the identification information of the data store that controls the storage area in which the value specified by the attribute is stored.

  For example, the record 411-1 indicates that the value specified by the date attribute and the step count attribute among the attributes of the life log data is stored in the storage area controlled by the data store # 0. The record 411-2 indicates that the value specified by the lunch comment attribute among the life log data attributes is stored in the storage area controlled by the data store # 1. Similarly, the record 411-3 indicates that the value specified by the lunch photo attribute is stored in the storage area controlled by the data store # 2.

  FIG. 7 is an explanatory diagram of an example of the contents stored in the identifier generation rule table. The identifier generation rule table 111 illustrated in FIG. 7 stores a record 111-1 and a record 111-2. The information for one record in the identifier generation rule table 111 is referred to as an identifier generation rule. The identifier generation rule table 111 includes two fields: a storage destination and an identifier generation rule. The storage destination field stores the name of the data store that controls the storage area that stores the specified value according to the life log data attribute. The identifier generation rule field stores information indicating the relationship between the data store-independent identifier and the storage position of the specified value.

  For example, the record 111-1 indicates that the identifier generation rule of the data store # 1 is “key # {identifier}”. Specifically, if the data store independent identifier is 1, the search device 101 generates a data store dependent identifier such as “key # 1”.

  FIG. 8 is an explanatory diagram of an example of the contents stored in the search situation table. The search status table 412 illustrated in FIG. 8 stores records 412-1 to 412-3. Note that the information for one record in the search status table 412 is referred to as search status information. The search status table 412 includes two fields: a period and the number of searches. The period in the search condition is stored in the period field. The number of times of searching for the corresponding period is stored in the search number field.

  For example, the record 412-1 indicates that the number of times that the search condition from the application is within one month is 100 times. The record 412-2 indicates that the number of times that the search condition from the application is within half a year is 50 times. Subsequently, operation examples of registration processing, search processing, identifier deletion processing, and migration processing performed by the search device 101 using the storage contents shown in FIGS. 5 to 8 are shown in FIGS.

  FIG. 9 is an explanatory diagram illustrating an operation example of registration processing. In FIG. 9, a process of registering life log data in the data store group 203 will be described. The life log data 901 is generated by, for example, an application that operates on the client 201 operated by a user or a sensor in the client 201. Specifically, the value stored in the date field of the storage area controlled by the data store # 0 is the operation date of the application by the user. Further, the value stored in the step count field is a value counted by the sensor. The information stored in the lunch comment in the storage area controlled by the data store # 1 is information generated by the user operating the application. The image file in the storage area controlled by the data store # 2 is a file generated by the camera device of the search apparatus 101 by a user operation.

  For example, in FIG. 9, the life log data 901 has the date “2011/8/31”, the number of steps is 5000 steps, the lunch comment is “delicious”, and the lunch photo is the image file 902. .

  The application transmits life log data 901 to the search device 101. The search apparatus 101 that has received the life log data 901 adds a record 112-1 that becomes a new record to the storage area controlled by the data store # 0. Next, the search device 101 stores “2011/8/31” in the date field of the record 112-1. In addition, the search apparatus 101 stores “5000” in the step count field of the record 112-1. Further, the search device 101 stores “99” numbered by the search device 101 in the identifier field of the record 112-1.

  Subsequently, using the identifier “99”, the search device 101 generates “key # 99”, which is a data store-dependent identifier, and stores the generated data store-dependent identifier in the launch comment field of the record 112-1. To do. Similarly, the search apparatus 101 generates “/ data / file # 99”, which is a data store-dependent identifier, using the identifier “99”, and generates the data store-dependent in the lunch photo field of the record 112-1. Stores an identifier.

  Further, the search apparatus 101 adds a record 501-1 in which the key field is “key # 99” and the lunch comment field is “delicious” to the data store # 1. Subsequently, the search apparatus 101 stores the image file 902 in the file path “/ data / file # 99” of the data store # 2. The stored image file is the same as the image file file # 99 shown in FIG.

  FIG. 10 is an explanatory diagram showing an example of data to be subjected to identifier deletion processing. In FIG. 10, a target for deleting a data store dependent identifier in the data store group 203 will be described. A graph 1001 indicates the number of searches to the data store group 203. The horizontal axis of the graph 1001 indicates the date. The vertical axis of the graph 1001 indicates the number of searches.

  As the graph 1001 shows, among the life log data stored in the data store group 203, the life log data included in the most frequent search period shown in FIG. There is a tendency. The most frequent search period is a date that is a certain period before the current date. The fixed period is, for example, one week, one month, or the like. As described above, the reason why the number of searches is biased by life log data is that information displayed by the application on the client 201 or information to be searched is often life log data from the latest life log data to a certain period. It is. Note that the number of life log data included in the most frequent search period is small, and the number of life log data after the most frequent search period is large.

  Therefore, the search apparatus 101 stores the data store-dependent identifier and the data store-independent identifier for the records included in the most frequent search period among the record groups in the storage area controlled by the data store # 0. For the records after the most frequent search period, the search apparatus 101 stores the data store independent identifier and deletes the data store dependent identifier.

  As a result, the record included in the most frequent search period that is likely to be a search target has a data store-dependent identifier and does not have to generate a data store-dependent identifier. Can be improved. Further, at the time of migration, the search apparatus 101 changes the data store dependent identifier of the record included in the most frequent search period. However, since the number of records included in the most frequent search period is small, the migration load is reduced. can do.

  Further, records after the most frequent search period have data store-independent identifiers, and do not have data store-dependent identifiers that need to be changed during migration. Since there are a large number of records after the most frequent search period, the search apparatus 101 can reduce the migration load during migration. Further, at the time of search, the search device 101 generates a data store-dependent identifier for records after the most frequent search period, but since records after the most frequent search period have few opportunities to be searched, the search performance The influence on can be reduced.

  FIG. 11 is an explanatory diagram illustrating an operation example of the search process. In FIG. 11, a search process for the data store group 203 will be described. When receiving the search request, the search device 101 notifies the data store # 0 of the search request. Data store # 0 extracts records that match the search conditions of the search request from the record group stored in the storage area controlled by the data store # 0. In FIG. 11, it is assumed that records 112-1 to 112-7 are records that match the search conditions.

  The search apparatus 101 that has acquired the records 112-1 to 112-7 is designated by the value designated by the lunch comment attribute and the lunch photo attribute by using either the lunch comment field and the lunch photo field or the identifier field. Get the value. In FIGS. 11 and 12, a case where a value designated by a lunch photo is acquired will be described for the sake of simplicity. The method for acquiring the value specified by the lunch comment attribute is also the same as the method for acquiring the value specified by the lunch photo attribute, so the notation in FIG. 11 is omitted.

  In FIG. 11, it is assumed that the search apparatus 101 determines that the search frequency is high for the records 112-1 to 112-3. Further, regarding the records 112-4 to 112-7, the search device 101 determines that the search frequency is low. Here, as a specific determination criterion, for example, the search apparatus 101 may determine that a record in which the date field is included in the most frequent search period is a record having a high search frequency. The example shown in FIG. 11 is an example in which the most frequent search period is set to one month. Alternatively, with reference to the search status table 412, a record included in a period in which the number of searches is equal to or greater than a specific threshold value may be determined as a record having a high search frequency.

  As another determination method for determining whether or not the search frequency is high, the search apparatus 101 may determine whether there is a value of the data store dependent identifier. When this determination method is used, the search apparatus 101 determines that the records 112-1 to 112-4 have a high search frequency. Also, the search apparatus 101 determines that the search frequency is low for the records 112-5 to 112-7 in which the value “(deleted)” is stored in the lunch comment field and the lunch photo field that are data store-dependent identifiers. To do.

  Since the search frequency for the records 112-1 to 112-3 is high, the search apparatus 101 uses the lunch comment field and the lunch photo field to specify a value specified by the lunch comment attribute and a value specified by the lunch photo attribute. get. For example, for a lunch photo, the search device 101 uses the file path stored in the lunch photo field to acquire an image file that has a value specified by the lunch photo attribute.

  Since the search frequency is low with respect to the records 112-4 to 112-7, the search apparatus 101 uses the identifier field to acquire the value specified by the lunch comment attribute and the value specified by the lunch photo attribute. For example, for a lunch photo, the search apparatus 101 uses the identifier field and the identifier generation rule of the data store # 2 in which the lunch photo is stored to acquire an image file that has a value specified by the lunch photo attribute. Specifically, for the record 112-4, the search apparatus 101 replaces {identifier} in the identifier generation rule “/ data / file # {identifier}” with “88” to identify the identifier “indicating the storage location of the image file”. / Data / file # 88 "is generated. Next, the search device 101 uses the generated identifier “/ data / file # 88” to acquire an image file having a value specified by the lunch photo attribute.

  FIG. 12 is an explanatory diagram illustrating an operation example of the migration process. In FIG. 12, the migration process of the data store group 203 will be described. In the migration process performed in FIG. 12, it is assumed that the data store that stores the image file having the value specified by the lunch photo attribute is migrated from the data store # 2 to the data store # 3. It is assumed that the data store # 3 is a file system as the data store type.

  Further, as an opportunity to perform migration processing, for example, when the operation of the data store integrated system 100 is started, a value specified by a specific attribute of life log data is stored as a file system in a local file system with low development cost and operation cost Suppose you were. However, as the storage amount of the specified value increases as the operation continues, the value specified by a specific attribute is transferred to a highly scalable distributed file system. Another opportunity for performing the migration process is when the number of users using the data store integration system 100 has increased, so that the hierarchy of the file system is added and the directory is divided into user groups.

  In FIG. 12, the storage method definition rule table 411 in which the information of the migration source is described is shown as a storage method definition rule table 411_S. Further, the storage method definition rule table 411 in which the information of the migration destination is described is shown as a storage method definition rule table 411_D. Further, the identifier generation rule table 111 in which the migration source information is described is shown as an identifier generation rule table 111_S. Further, the identifier generation rule table 111 in which the information of the migration destination is described is shown as an identifier generation rule table 111_D.

  Specifically, the record 411-3_S of the storage method definition rule table 411_S indicates that the value specified by the lunch photo attribute is stored in the data store # 2 that is the migration source. The record 411-3_D of the storage method definition rule table 411_D indicates that the value specified by the lunch photo attribute is stored in the data store # 3 that is the migration destination. Furthermore, the record 111-2_S of the identifier generation rule table 111_S indicates that the identifier generation rule in the data store # 2 is “/ data / file # {identifier}”. In addition, the record 111-2_D of the identifier generation rule table 111_D indicates that the identifier generation rule in the data store # 3 is “/ data / usr1 / file- {identifier}”.

  In FIG. 12, it is assumed that the search device 101 determines that the search frequency is high for the records 112-1 to 112-3. Further, regarding the records 112-4 to 112-7, the search device 101 determines that the search frequency is low. The search frequency determination criteria are the same as the determination criteria shown in FIG.

  Since the search frequency is high for the records 112-1 to 112-3, the search apparatus 101 determines the storage location in the data store # 3 from the value indicating the storage location in the data store # 2. Change to the value shown. Specifically, for the record 112-1, the search device 101 changes the value of the lunch photo field from “/ data / file # 99” to “/ data / usr1 / file-99”. Further, the search apparatus 101 moves the image file having the value specified by the lunch photo attribute from the data store # 2 to the data store # 3. Next, flowcharts for performing the operations shown in FIGS. 10 to 12 will be described with reference to FIGS.

  FIG. 13 is a flowchart illustrating an example of an identifier deletion processing procedure. FIG. 13 illustrates an example of identifier deletion processing for the data store # 0. The identifier deletion process may be periodically executed when a timer included in the search apparatus 101 expires. Alternatively, the identifier deletion process may be executed by an instruction from the administrator. Further, the search device 101 records the latest date of the target data when the identifier deletion process is executed.

  The search device 101 acquires search status information from the search status table 412 (step S1301). Next, the search device 101 reads the latest date of the target data for the previous identifier deletion process (step S1302). There may be a case where the latest date of the target data for the previous identifier deletion process is not stored. In this case, the search device 101 sets the oldest date among the date fields of the record group in the storage area stored in the data store # 0 as the latest date of the target data for the previous identifier deletion process.

  Subsequently, the search device 101 acquires storage method definition rule information from the storage method definition rule table 411 (step S1303). Next, the search device 101 searches for a record that is not included in the most frequent search period from a record group in which the date field of the storage area controlled by the data store # 0 is newer than the latest date of the target data for the previous identifier deletion process. (Step S1304).

  Subsequently, the search device 101 determines whether the search result is empty (step S1305). When the search result is not empty (step S1305: No), the search device 101 deletes the data store dependent identifier of each record of the search result (step S1306). Next, the search device 101 stores the latest date of the target data for the identifier deletion process (step S1307). After the execution of step S1307 ends, or when the search result is empty (step S1305: Yes), the search device 101 ends the identifier deletion process. As described above, the identifier deletion process shown in FIG. 13 deletes the data store-dependent identifier that is not read out, and can reduce the storage amount of the storage area controlled by the data store # 0.

  FIG. 14 is a flowchart (part 1) illustrating an example of a search processing procedure. 14 and 15, an example of a search process for the data store group 203 will be described. The search device 101 receives a search request from the application (step S1401). Next, the search device 101 acquires storage method definition rule information from the storage method definition rule table 411 (step S1402). Subsequently, the search device 101 determines whether the storage method definition rule information has been acquired (step S1403). When acquisition is not possible (step S1403: No), the search device 101 returns an error to the application (step S1404). After completing the process of step S1404, the search device 101 ends the search process.

  If it can be obtained (step S1403: Yes), the search device 101 updates the search status table 412 according to the search condition (step S1405). Regarding the processing in step S1405, for example, when the search condition is within one month, the search device 101 increments the value of the search count field of the record 412-1. If the search condition is within half a year, the search apparatus 101 increments the value of the search count field of the record 412-2.

  Subsequently, the search device 101 searches the search index for a record that matches the search condition (step S1406). Next, the search device 101 determines whether the search result is empty (step S1407). When the search result is empty (step S1407: Yes), the search device 101 returns the empty search result to the application (step S1408). After completing the process of step S1408, the search device 101 ends the search process.

  If the search result is not empty (step S1407: No), the search device 101 selects the first record from the record group of the search result (step S1409). After completing the process of step S1409, the search apparatus 101 proceeds to the process of step S1501 shown in FIG.

  FIG. 15 is a flowchart (part 2) illustrating an example of a search processing procedure. After completing the process of step S1409, the search device 101 determines whether to use the data store-dependent identifier in the search index for the value specified by the attribute in the selected record (step S1501). Note that the method of determining whether to use the data store-dependent identifier in the search index, which is the processing of step S1501, is, for example, a method of referring to the date field of the selected record shown in FIG.

  When the data store dependent identifier in the search index is not used (step S1501: No), the search device 101 acquires identifier generation rule information from the identifier generation rule table 111 (step S1502). Next, the search device 101 generates a data store-dependent identifier using the acquired identifier generation rule information (step S1503).

  Next, the search device 101 uses the data store dependency identifier generated in the process of step S1503 or the data store dependency identifier in the search index (step S1501: Yes), and the value specified by the attribute from the data store group 203. Is acquired (step S1504). Subsequently, the search device 101 determines whether or not the values designated by the attributes have been acquired for all the record groups of the search results (step S1505). When there is a record that has not acquired the value specified by the attribute (step S1505: No), the search device 101 selects the next record from the record group of the search result (step S1506). After step S1506 ends, the search apparatus 101 proceeds to the process of step S1501.

  When the values specified by the attributes are acquired for all the search result record groups (step S1505: Yes), the search apparatus 101 aggregates the acquired values (step S1507). Next, the search device 101 returns the aggregated value to the application (step S1508). After completing the process of step S1508, the search device 101 ends the search process. As described above, the search processing shown in FIGS. 14 and 15 can return search results corresponding to the search request requested by the application.

  FIG. 16 is a flowchart (part 1) illustrating an example of the migration processing procedure. An example of the migration process of the data store group 203 will be described with reference to FIGS. The search device 101 receives a migration request from the administrator terminal (step S1601). Next, the search device 101 acquires the storage method definition rule information of the migration source and the migration destination from the storage method definition rule table 411 (step S1602). Subsequently, the search apparatus 101 determines whether or not the migration source and migration destination storage method definition rule information has been acquired (step S1603).

  When acquisition is not possible (step S1603: No), the search device 101 returns an error to the administrator terminal (step S1604). After the process of step S1604 ends, the search device 101 ends the migration process. If it can be obtained (step S1603: Yes), the search device 101 searches the search index for a record that matches the search condition (step S1605). Note that the search condition in step S1605 may be, for example, a search condition having a value specified by the attribute to be migrated among the search index record group, or a search condition of all the search index record groups. It may be.

  Next, the search device 101 selects the first record from the record group of search results (step S1606). After executing the process of step S1606, the search device 101 proceeds to the process of step S1701 shown in FIG.

  FIG. 17 is a flowchart (part 2) illustrating an example of the migration processing procedure. The search apparatus 101 determines whether to use the data store dependent identifier in the search index for the value specified by the attribute in the selected record (step S1701). Note that the method of determining whether to use the data store-dependent identifier in the search index, which is the processing of step S1701, is a method of referring to the date field of the selected record shown in FIG. 10, for example.

  When the data store dependency identifier in the search index is not used (step S1701: No), the search device 101 generates a data store dependency identifier of the migration source (step S1702). Subsequently, the search device 101 generates a migration destination data store dependency identifier (step S1703). Next, the search device 101 uses the migration source and migration destination data store dependent identifiers to migrate the value specified by the attribute from the migration source data store to the migration destination data store (step S1704). As a specific process of step S1704, the search device 101 reads a specified value from the storage area controlled by the migration source data store using the migration source data store dependent identifier. Subsequently, the search device 101 writes the value read into the storage area controlled by the migration destination data store using the migration destination data store dependent identifier.

  When the data store dependent identifier in the search index is used (step S1701: Yes), the search device 101 generates a migration destination data store dependent identifier (step S1705). Next, the search device 101 uses the data store dependency identifier and the transfer destination data store dependency identifier in the search index to transfer the value specified by the attribute from the transfer source data store to the transfer destination data store (step S1706). ). As a specific process of step S1706, the search device 101 reads a specified value from the storage area controlled by the migration source data store using the data store dependent identifier in the search index. Subsequently, the search device 101 writes the value read into the storage area controlled by the migration destination data store using the migration destination data store dependent identifier. Subsequently, the search device 101 changes the data store dependency identifier in the search index to the generated data store dependency identifier of the migration destination (step S1707).

  After executing the processing of step S1704 or step S1707, the search apparatus 101 determines whether or not the values designated by the attributes have been transferred for all the record groups of the search results (step S1708). If there is a record that does not migrate the value specified by the attribute (step S1708: No), the search device 101 selects the next record (step S1709). After completing the execution of step S1709, the search apparatus 101 proceeds to the process of step S1701.

  When the values designated by the attributes are transferred for all the record groups of the search results (step S1708: Yes), the search device 101 returns the transfer results to the administrator terminal (step S1710). After the process of step S1710 ends, the search device 101 ends the migration process. In this way, the migration process shown in FIGS. 16 and 17 can perform the migration process of the value specified by the attribute corresponding to the migration request requested by the administrator terminal.

  As described above, according to the search device, the search method, and the search program, the storage location of the value specified by the attribute in the data store is assigned to the information with high search frequency, and the attribute is specified to the low information with the attribute The storage position is obtained by assigning the identifier of the value to be stored.

  As a result, when the search device migrates data stored in a specific data store to another data store, the number of records for changing the data store-dependent identifier in the search index is less than the number of records in all the search indexes. , Migration load can be reduced. Since the migration load can be reduced, for example, the search device can suppress the performance degradation of the registration process or the search process during the migration process. Further, since the search device has a data store dependent identifier for a record with a high search frequency, it is not necessary to generate a data store dependent identifier, and search performance can be improved. Thus, the search device can achieve both a reduction in search index migration load and an improvement in search performance.

  In addition, the search device controls the data store using either the data store-dependent identifier or the data store-independent identifier based on the comparison result between the date when the information matching the search condition is registered and the current date. It may be determined whether to access the storage area. Thereby, the search device can identify a record having a high search frequency by a simple process of comparing dates without acquiring an actual search status.

  In addition, if there is a data store dependency identifier in the information that matches the search condition, the search device may determine whether to access the storage area controlled by the data store using the data store dependency identifier. In this case, data store independent identifiers are not assigned to information with low search frequency. Thereby, the search device can specify a record having a high search frequency without performing a comparison process.

  In addition, the search device updates the search condition history, and uses the search condition history to access the storage area controlled by the data store using either the data store-dependent identifier or the data store-independent identifier. May be determined. Thereby, the search device can specify a record with a high search frequency according to the actual search situation.

  In addition, the search device may store an identifier specified based on a comparison result between a date on which information having at least a specific attribute is registered and a current date among the data store dependent identifier and the data store independent identifier. . As a result, the search device can be set to access information with a high search frequency using the data store-dependent identifier by a simple process of date comparison without acquiring the actual search status.

  In addition, the search device updates the search condition history, and uses the search condition history to register the date and the current date when the information having at least a specific attribute among the data store dependent identifier and the data store independent identifier is registered. An identifier specified based on the comparison result may be stored. Accordingly, the search device can be set to access information with a high search frequency using the data store-dependent identifier according to the actual search situation.

  Further, the search device may delete the data store dependent identifier based on information having a specific attribute. Thereby, the search device can reduce the storage amount of the search index for storing the data store dependent identifier.

  Further, the search device searches for information that matches the search condition to be migrated, and changes the data store dependent identifier from the migration source to the migration destination for information accessed using the data store dependent identifier. Since it is not necessary to change the information accessed using the data store independent identifier, the search device can reduce the load of the migration process compared to the conventional search device.

  Note that the search method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. The search program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The search program may be distributed via a network such as the Internet.

# 0, # 1, # 2 Data store 101 Search device 111 Identifier generation rule table 401 Reception unit 402 Search unit 403 Determination unit 404 Generation unit 405 Access unit 406 Delete unit 407 Update unit 408 Change unit 411 Storage method definition rule table 412 Search Status table

Claims (11)

A receiving unit that receives a search request for information that matches a search condition among information groups having a specific attribute;
When the search request is received by the reception unit, at least the specific attribute of the first identifier indicating the storage position of the value specified by the specific attribute or the second identifier for identifying the specified value A search unit that searches for information that matches the search condition with reference to a first storage unit that stores an identifier specified based on information having
The designation is made by controlling a second data store different from the first data store among the first identifier and the second identifier based on the information that matches the search condition searched by the search unit. A determination unit that determines an identifier used to access the second storage unit that stores the value to be stored;
When the identifier determined by the determination unit is the second identifier, the designation is made based on the second identifier and information indicating the relationship between the storage location of the designated value and the designated value. A generation unit that generates an identifier indicating a storage position of the value to be stored;
Using the identifier indicating the storage location generated by the generation unit or the first identifier, the access unit accessing the specified value;
A search device comprising: The determination unit
Of the first identifier and the second identifier, the information that matches the search condition searched by the search unit is based on a comparison result between the date registered in the first storage unit and the current date. The search device according to claim 1, wherein an identifier used for accessing the second storage unit is determined. The determination unit
When the information that matches the search condition searched by the search unit includes the first identifier, an identifier used to access the second storage unit is determined as the first identifier. The search device according to claim 1. When the search request is received by the reception unit, the update unit updates the history of the search condition,
The determination unit
Based on the information that matches the search condition searched by the search unit and the history of the search condition updated by the update unit, the second storage of the first identifier and the second identifier. The search apparatus according to claim 1, wherein an identifier used for accessing a copy unit is determined. The first storage unit
The identifier specified based on the comparison result between the date when the information having at least the specific attribute of the first identifier or the second identifier is registered in the first storage unit and the current date is a first identifier. The search device according to claim 1, wherein the search device stores the data by controlling the data store. When the search request is received by the reception unit, the update unit updates the history of the search condition,
The first storage unit
Control of the first data store with the identifier specified based on at least the information having the specific attribute of the first identifier or the second identifier and the history of the search condition updated by the update unit The search device according to claim 1, wherein the search device is stored according to claim 1. The first storage unit
Control of the first data store with the identifier specified based on at least the information having the specific attribute of the first identifier or the second identifier and the history of the search condition updated by the update unit The retrieval apparatus according to claim 4, wherein the retrieval apparatus stores the retrieval information.   2. The information processing apparatus according to claim 1, further comprising: a deletion unit that deletes the first identifier included in the information having the specific attribute based on the information having the specific attribute stored in the first storage unit. The search apparatus as described in any one of -7. The reception unit
Accepting a search request for information that matches the search condition with the specified value as the search condition,
The generator is
Further, an identifier indicating a storage position in the third storage unit to which the designated value is transferred is generated by controlling a third data store different from the second data store,
The access unit is
The specified value is read from the second storage unit using the identifier indicating the storage position in the second storage unit generated by the generation unit or the first identifier, and the read value is read from the second storage unit. Write to the third storage unit using an identifier generated by the generation unit and indicating the storage position in the third storage unit,
When the identifier determined by the determination unit is the first identifier, the first identifier stored in the first storage unit is changed to an identifier indicating a storage position in the third storage unit The search device according to claim 1, further comprising: a changing unit that performs the operation. Computer
Accept a search request for information that matches the search criteria among information groups with specific attributes,
When the search request is received, information having at least the specific attribute of the first identifier indicating the storage position of the value specified by the specific attribute or the second identifier for identifying the specified value is included. A first storage unit that stores the identifier identified based on the control of the first data store, and searches for information that matches the search condition;
Based on the searched information that matches the search condition, the value specified by the control of the second data store different from the first data store among the first identifier and the second identifier is stored. Determining an identifier used to access the second storage unit,
When the determined identifier is the second identifier, the storage of the specified value is based on the second identifier and the information indicating the relationship between the storage position of the specified value and the specified value. Generate an identifier that indicates the location,
Using the identifier indicating the generated storage location or the first identifier to access the specified value;
A search method characterized by executing processing. On the computer,
Accept a search request for information that matches the search criteria among information groups with specific attributes,
When the search request is received, information having at least the specific attribute of the first identifier indicating the storage position of the value specified by the specific attribute or the second identifier for identifying the specified value is included. A first storage unit that stores the identifier identified based on the control of the first data store, and searches for information that matches the search condition;
Based on the searched information that matches the search condition, the value specified by the control of the second data store different from the first data store among the first identifier and the second identifier is stored. Determining an identifier used to access the second storage unit,
When the determined identifier is the second identifier, the storage of the specified value is based on the second identifier and the information indicating the relationship between the storage position of the specified value and the specified value. Generate an identifier that indicates the location,
Using the identifier indicating the generated storage location or the first identifier to access the specified value;
A search program characterized by causing processing to be executed.

Images