JP6378497B2 - Information processing apparatus, information processing method, and program - Google Patents

Description

  This case relates to an information processing apparatus, an information processing method, a program, and a data structure.

  Conventionally, there has been proposed a multi-tenant architecture that reduces costs by sharing the resources of one system among a plurality of users. For example, it is determined whether or not each column in the data input from the tenant information processing apparatus is a target of serialization processing, and data of all columns determined to be serialization processing is serialized together. A technique of processing and storing in a column in a serialized column of a data schema has been proposed (Patent Document 1). According to this technique, the number of columns is not limited to a preset number. In addition, a function provided by a database system such as a search can be applied to a function field including columns for storing data that has not been serialized.

International Publication No. 2011-111532 Pamphlet

  In the above-described technique, data that may be a search target in the future needs to be determined in advance as “not a serialization target” and stored in an appropriate column of the function field. The determination as to whether or not to be serialized is made by an application developer or by analysis of program code or execution history. However, there is a problem that a design error may occur and it is difficult to design in anticipation of future specification changes.

  Accordingly, the present invention aims to eliminate the restriction on the number of columns in a general-purpose database used in a multi-tenant system and improve the usability of functions provided by a DBMS (DataBase Management System). To do.

  An information processing apparatus according to an aspect of the present invention performs data operations on a physical table shared by a plurality of tenants in a multi-tenant database. Specifically, based on the mapping information that defines the correspondence between the columns of the logical table defined for each tenant and the columns of the physical table for each tenant, the first record in the logical table is changed to the second record in the physical table. Based on a registration unit that disassembles a record and a third record and adds and registers information indicating a connection relationship between the second record and the third record, and mapping information and information indicating a connection relationship. An extraction unit that reads the second record and the third record from the physical table, and outputs the combined record.

In this way, the record of the logical table defined by the tenant can be decomposed into a plurality of records and registered in the physical table, and the restriction on the number of columns in the physical table can be eliminated. Further, since the columns of the logical table are registered in the columns of the associated physical table, functions such as search provided by the DBMS can be indirectly applied to each column of the logical table. That is, in the general-purpose database used in the multi-tenant system, it is possible to eliminate the restriction on the number of columns and improve the usability of the functions provided by the DBMS.

  The physical table may include a plurality of columns having different data types. In this way, the usability of functions provided by the DBMS can be further improved.

  Further, when one record in the logical table is associated with three or more records in the physical table in the mapping information, the registration unit converts one record in the logical table into three or more records in the physical table based on the mapping information. Break down into records, add and register information indicating the connection relationship of the three or more records, and the extraction unit reads out the three or more records from the physical table based on the mapping information, and outputs them by combining them Also good. As described above, if one record in the logical table is divided into three or more records in the physical table and held, the restriction on the number of columns is eliminated.

  The information processing apparatus may include a plurality of physical tables, and the mapping information may further define a correspondence between the logical table and any of the plurality of physical tables. By distributing the physical table in this way, it becomes possible to suppress the deterioration of the response due to the increase in the number of records.

  The data structure according to another aspect of the present invention is applied to a physical table shared by a plurality of tenants in a multi-tenant database. Specifically, one or more columns that hold information for uniquely identifying the logical table record defined for each tenant, and the logical table record are divided into multiple records in the physical table and registered. In this case, a column that holds information indicating a connection relation of the plurality of records and a plurality of columns that hold values held in the plurality of columns of the logical table independently.

  In this way, the record of the logical table defined by the tenant can be decomposed into a plurality of records and registered in the physical table, and the restriction on the number of columns in the physical table can be eliminated. In addition, since it includes a plurality of columns that independently hold values registered and held in a plurality of columns of the logical table, functions such as search provided by the DBMS can be indirectly applied to each column of the logical table. That is, in the general-purpose database used in the multi-tenant system, it is possible to eliminate the restriction on the number of columns and improve the usability of the functions provided by the DBMS.

  The contents of the means for solving the above problems can be combined as much as possible without departing from the problems and technical ideas of the present invention. Also, a method for executing the above means by a computer may be implemented, or a program for causing the computer to execute the above means may be provided. The program may be provided by being recorded on a computer-readable recording medium. A computer-readable recording medium refers to a recording medium that stores information by electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer. Among such recording media, those removable from the computer include, for example, optical disks, magneto-optical disks, flexible disks, magnetic tapes, memory cards, and the like. Further, there are HDD (Hard Disk Drive), SSD (Solid State Drive), ROM (Read Only Memory) and the like as recording media fixed to the computer.

  According to the present invention, in a general-purpose database used in a multi-tenant system, it is possible to eliminate the restriction on the number of columns and improve the usability of functions provided by the DBMS.

It is a figure which shows schematic structure of the whole system. It is a functional block diagram which shows an example of a multi-tenant system. It is an apparatus block diagram which shows an example of a computer. It is a table | surface which shows an example of the logical table of tenant A. It is a table | surface which shows an example of the logical table of tenant A. It is a table | surface which shows an example of the logical table of the tenant B. It is a table | surface which shows an example of a physical table. It is a table | surface which shows an example of metadata. It is a processing flowchart which shows an example of a data operation process. It is a processing flowchart which shows an example of a setting process. It is a figure for demonstrating the correspondence of a data type. It is a table | surface which shows an example of an index table. It is a figure for demonstrating a table space.

  Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. The configuration of the following embodiment is an exemplification, and the present invention is not limited to the configuration of the embodiment.

<System configuration>
FIG. 1 is a diagram illustrating an example of a schematic configuration of a system. In the example of FIG. 1, a multi-tenant system 1 according to the present invention and a tenant terminal 2 (2 a and 2 b in FIG. 1) are connected via a network 3. The multi-tenant system 1 is a system that commonly uses hardware resources and software resources for a plurality of tenants (that is, a plurality of users), and uses a physical table having a data structure according to the present invention. That is, the system provides so-called SaaS (Software as a Service). In addition, the tenant terminal 2 requests a data operation from the multi-tenant system 1 via the network 3 and receives a result. For convenience, it is assumed that the tenant terminal 2a is a tenant A terminal and the tenant terminal 2b is a tenant B terminal. Although FIG. 1 shows two tenant terminals, the number of tenant terminals is not limited to two.

<Multi-tenant system>
FIG. 2 is a functional block diagram illustrating an example of the multi-tenant system 1. The multi-tenant system 1 of FIG. 2 includes a DBMS 11, a metadata storage unit 12, a setting unit 13, a request reception unit 14, a conversion unit 15, and a result response unit 16.

The DBMS 11 is a system that provides functions necessary for operation management of a database (for example, RDB (Relational Database)), and includes a physical table 111 and a data operation unit 112. The physical table 111 is a storage device that stores data records as physical files with a predetermined data structure (schema). In the present embodiment, a plurality of tenants share the physical table 111 having the same data structure. That is, data of different tenants is held in the same physical table 111. Further, the data operation unit 112 performs operations such as record insertion (registration), record selection (search), record update, and record deletion on the physical table 111. The DBMS 11 can use existing products provided by various vendors.

In addition, the metadata storage unit 12 stores metadata indicating a correspondence between the data structure of the logical table and the data structure of the physical table 111 that each tenant defines in accordance with the business or the like. In this embodiment, the physical table 111 is common to each tenant and has a general-purpose data structure. In the metadata, one column of the physical table 111 is assigned to one column (also referred to as “row”, “item”, or “field”) of the logical table. At this time, a column of a corresponding data type is assigned. The metadata is also referred to as “mapping information”.

  Here, one record of the logical table defined by each tenant may not be held by one record of the physical table 111. For example, if the number of columns in the physical table 111 is smaller than the number of columns in the logical table, a record in the logical table cannot be held by one record in the physical table. Further, even when the number of columns of a certain data type prepared in advance in the physical table 111 is smaller than the number defined in the logical table, the logical table record cannot be held by one record in the physical table. . In such a case, one record in the logical table is divided into two or more records in the physical table 111 and held. For this reason, the physical table 111 includes a column indicating a connection relationship (also referred to as “page”) of a plurality of records. The metadata associates the values in a one-to-one correspondence by associating the columns of the logical table with the columns and pages of the physical table 111.

  The setting unit 13 receives a command for setting metadata from the tenant terminal 2 via the network 3 and stores the command in the metadata storage unit 12. In addition, the request reception unit 14 receives a request for performing a data operation from the tenant terminal 2 via the network 3 and passes the request to the conversion unit 15. Based on the metadata stored in the metadata storage unit 12, the conversion unit 15 converts the data operation request received from the tenant terminal into a data operation command (also referred to as “query” or “query”) that can be executed in the DBMS. The data is transmitted to the data operation unit 112 of the DBMS 11. Similarly, the result of the data operation acquired from the data operation unit 112 is received, converted into an item in the logical table of each tenant, and passed to the result response unit 16. The result response unit 16 transmits the result of the data operation request received from the tenant terminal 2 to the tenant terminal 2.

<Device configuration>
FIG. 3 is an apparatus configuration diagram illustrating an example of a computer. The multi-tenant system 1 and the tenant terminal 2 are computers as shown in FIG. A computer 1000 shown in FIG. 3 includes a CPU (Central Processing Unit) 1001, a main storage device 1002, an auxiliary storage device 1003, a communication IF (Interface) 1004, an input / output IF (Interface) 1005, a drive device 1006, and a communication bus 1007. ing. The CPU 1001 performs processing according to the present embodiment by executing a program (also referred to as “software” or “application”). The main storage device 1002 caches programs and data read by the CPU 1001 and develops a work area of the CPU. Specifically, the main storage device is a RAM (Random Access Memory), a ROM (Read Only Memory), or the like. The auxiliary storage device 1003 stores programs executed by the CPU 1001, setting information used in the present embodiment, and the like. Specifically, the auxiliary storage device 1003 is an HDD (Hard-disk Drive), an SSD (Solid State Drive), a flash memory, or the like. The main storage device 1002 and the auxiliary storage device 1003 function as means for storing the physical table 111, the metadata storage unit 12, and other temporary data of the multi-tenant system 1. The communication IF 1004 transmits / receives data to / from other computers. The communication IF 1004 is specifically a wired or wireless network card or the like. The multi-tenant system 1 and the tenant terminal 2 are connected to a network 3 such as the Internet via a communication IF 1004. The input / output IF 1005 is connected to the input / output device and accepts an operation from the user or presents information to the user. Specifically, the input / output device is a keyboard, a mouse, a display, a touch panel, or the like. The drive device 1006 reads data recorded on a storage medium such as a magnetic disk, a magneto-optical disk, and an optical disk, and writes data to the storage medium. The above components are connected by a communication bus 1007. A plurality of these components may be provided, or some of the components (for example, the drive device 1006) may not be provided. Further, the input / output device may be integrated with the computer. In addition, the program executed in this embodiment is provided via a portable storage medium readable by the drive device 1006, a portable auxiliary storage device 1003 such as a flash memory, a communication IF 1004, and the like. It may be. Then, when the CPU 1001 executes a predetermined program, the computer shown in FIG.

<Data structure>
Next, an example of the correspondence between the data structure of the physical table 111 and the data structure of the logical table will be described. The tables shown in FIGS. 4 to 6 are examples of logical tables and stored data that the tenant A or the tenant B holds in the multi-tenant system 1. FIG. 4 is a table showing an example of column names and registered records of the “user” table managed by the tenant A. The “user” table shown in FIG. 4 has columns of “name”, “sex”, “height”, “weight”, and “birthday”. For example, “name” and “gender” are character string types, “height” and “weight” are fixed-point types, and “birthday” is a date type. FIG. 5 is a table showing an example of data registered in the “div” table managed by the tenant A. The “div” table illustrated in FIG. 5 includes columns of “department name” and “department type”. The data type is, for example, a character string type for both “department name” and “department type”. FIG. 6 is a table showing an example of data registered in the “item” table managed by the tenant B. The “item” table illustrated in FIG. 6 includes columns of “product name”, “product type”, “unit name”, “price”, and “release date”. The data type is, for example, “product name”, “product type”, “unit name” is a character string type, “price” is an integer type, and “release date” is a date type.

  The table shown in FIG. 7 is an example of the data structure of the physical table 111 and stored data. The physical table of FIG. 7 includes columns such as “TenantID”, “TypeID”, “DataID”, “PageID”, “Char1”, “Char2”, “Num1”, “Num2”, “Date1”, “Date2”, etc. Have. The data types are, for example, “TenantID”, “TypeID”, “Char1” and “Char2” are character string types, “Num1” and “Num2” are numeric types, and “Date1” and “Date2” are date types. is there. The data size of each column (also referred to as “data length”) may be a variable length or a fixed length. Although not explicitly shown in FIG. 7, a plurality of columns having the same data type and different data sizes may be provided.

  In the “TenantID” column, identification information for identifying a tenant who is a user of the multi-tenant system 1 is registered. In this embodiment, when the user of the tenant terminal 2 uses the multi-tenant system 1, first, authentication processing is performed using tenant identification information (for example, “TenantID”) and a password. Basically, each tenant can perform operations such as selection, update, and deletion of a record whose identification information is registered in the “TenantID” column among the records registered in the physical table 111. Shall. Further, when each tenant inserts a new record into the physical table 111, his / her identification information is registered in the “TenantID” column. By doing in this way, SaaS which shared from a hardware resource to an application and a database structure is realizable.

  Identification information for identifying a logical table in each tenant is registered in the “TypeID” column. In the column “DataID”, identification information for uniquely specifying a record in each logical table is registered. By assigning (numbering) the value of “DataID” so that the record can be uniquely identified in each logical table, the record in each tenant's logical table is uniquely identified by the composite key of “TenantID”, “TypeID”, and “DataID” Can be specified. Note that, as shown in FIG. 7, a value of “DataID” may be assigned so that a record can be uniquely specified in all logical tables.

  In the “PageID” column, information representing a connection relationship (connection order) when one record in the logical table is divided into a plurality of records and registered in the physical table 111 is registered. In the example of FIG. 7, a serial number is registered as information representing the connection relationship. In the example of FIG. 7, since the column of the logical table can be specified by “DataID” and “PageID”, one record in the logical table can be restored from a plurality of records in the physical table 111. Note that the records of the physical table 111 may be uniquely identified using the “TenantID”, “TypeID”, “DataID”, and “PageID” columns as composite keys.

  In addition, the values of each column of the logical table are registered in the columns “Char1”, “Char2”, “Num1”, “Num2”, “Date1”, “Date2”,. That is, these are columns that independently hold values held in a plurality of columns of the logical table. For example, “Char1” and “Char2” are character string type columns. “Num1” and “Num2” are numeric type columns. “Date1” and “Date2” are date type columns. 7 is an example in which the column configuration is simplified. The physical table 111 may further include columns of other data types, and may include three or more columns of each data type. Also good. The number of columns of each data type provided in the physical table 111 may not be the same. For example, the physical table 111 can be defined to have 30 character string columns, 10 numeric columns, and 5 date columns. However, since multi-tenant systems require versatility, the number of columns for each data type may be determined on average, or the number of columns for each data type may be determined based on tenant needs, actual usage, etc. . The association between the data structure of the logical table and the data structure of the physical table 111 is defined in the metadata in advance for each tenant.

FIG. 8 is a table showing an example of contents registered as metadata. The metadata is stored as a table on the DBMS or a file on the file system for each tenant, for example. In the present embodiment, the physical table 111 has a combination of a tenant (eg, “tenant A”, “tenant B”, etc.) and a logical table (eg, “user” table, “div” table, “item” table, etc.). “TypeID” is assigned uniquely. Further, the metadata indicates the correspondence between the columns of the logical tables shown in FIGS. 4 to 6 and the columns of the physical table shown in FIG. In the present embodiment, one record of the logical table may be decomposed into a plurality of records in the physical table 111 and registered. Therefore, the “column name” of the logical table is associated with the “corresponding column name” of the physical table 111 and “PageID” for uniquely identifying each decomposed record. When the “PageID” is only 1 for a certain “DataID” like the records of “001” to “004” in FIG. 7, the logical table record is not decomposed and the physical table 111 is not decomposed. It can be seen that On the other hand, when there is a record having the same “DataID” value and a “PageID” value of 2 or more, such as the “DataID” value “005” and “006” in FIG. It can be seen that the table records are divided into a plurality of pieces and stored in the physical table 111. Thus, the “PageID” column can be said to be data indicating a connection relationship between a plurality of records in the physical table 111.

In the present embodiment, logical tables and metadata are basically defined for each tenant. However, master data shared by each tenant may be prepared. For example, a postal code master that stores a postal code and a part of an address in association with each other may be held with metadata common to each tenant. Since the contents of such a zip code master are determined by the standard of the postal system, it is more efficient that each tenant uses it in common. The common metadata may be centrally managed by the administrator of the multi-tenant system 1 as a file on the file system, and the metadata of each tenant may be managed in association with “TenantID” as a record on the DBMS.

<Data manipulation processing>
FIG. 9 is a process flow showing an example of the data operation process. Note that when the multi-tenant system 1 receives a connection request from the tenant terminal 2, it first identifies the tenant by performing a predetermined authentication process. Further, it is assumed that the metadata shown in FIG. 8 is held in advance in the metadata storage unit.

  The request accepting unit 14 of the multi-tenant system 1 accepts a request for data operation from the tenant terminal 2 (FIG. 9: S1). The request is accepted as, for example, a query (for example, a query language such as SQL) that requests the logical table to insert, select, update, and delete records. Then, the request reception unit 14 transmits the request to the conversion unit 15. The request may be issued from a so-called business application of each tenant. For example, when a product purchase request is received from a customer terminal connected to the network 3, the application of the multi-tenant system 1 may generate a data operation request.

  The conversion unit 15 of the multi-tenant system 1 converts the request received from the request reception unit 14 into a request for the physical table 111 based on the metadata stored in the metadata storage unit 12 (S2). Here, the content of the query is corrected based on the metadata shown in FIG. That is, one or two or more records are virtually handled as one record based on the value held in the column “PageID” of the physical table, which is information indicating the connection relationship between records. More specifically, a condition is added in which a record whose value of “TenantID” matches the identification information of the requesting tenant is targeted. Also, the operation target table is changed to the physical table 111, and the specification of the operation target logical table in the request before conversion is replaced with the specification of the value of the column “TypeID”. Further, the column specification in the request before conversion is replaced with the column of the physical table 111 and the specification of “PageID”.

For example, an example in which the tenant B transmits the following query 1A for searching will be described.
(Query 1A)
SELECT product name, product type, unit name, price, release date FROM item WHERE product type = 'food';

In this example, a record whose value of “product type” is “food” is extracted from the logical table “item”, and “product name”, “product type”, “unit name”, “price”, and “release date” are extracted. This is a request to display the value of each item. Then, the query 1A is converted into the following query 1B by the conversion unit 15. It is assumed that the physical name of the physical table 111 is “DataTable”.
(Query 1B)
SELECT p1.Char1, p1.Char2, p2.Char1, p1.Num1, p1.Date1
FROM DataTable p1 LEFT OUTER JOIN DataTable p2
ON p1.DataID = p2.DataID and p2.PageID = 2
WHERE p1.Char2 = 'food'
AND p1.TenantID = 'B' AND p1.TypeID = 'item' AND p1.PageID = 1;

In this example, a record having a “PageID” value of 1 and a record having a “PageID” value of 2 are self-joined. Further, based on the metadata in FIG. 8, the “product type” in the logical table is converted to “Char2” (that is, p1.Char2) whose “PageID” is 1 in the physical table 111. Note that the DBMS 11 may extract a record with “PageID” of 1 and a record of 2, respectively, and the conversion unit 15 may be combined into one virtual record. Further, when there is a record having a “PageID” value of 3 or more, it is possible to self-join three or more records in the converted query.

Next, an example will be described in which the tenant B transmits the following query 2A in order to count the number of records.
(Query 2A)
SELECT COUNT (*) FROM item WHERE product type = 'food';

The query 2A is converted into the following query 2B by the converter 15.
(Query 2B)
SELECT COUNT (*)
FROM DataTable p1 LEFT OUTER JOIN DataTable p2
ON p1.DataID = p2.DataID and p2.PageID = 2
WHERE p1.Char2 = 'food'
AND p1.TenantID = 'B' AND p1.TypeID = 'item' AND p1.PageID = 1;

Also in this example, a record with “PageID” of 2 is self-joined, and a record with “PageID” of 1 is selected (SELECT). Further, based on the metadata in FIG. 8, the “product type” in the logical table is converted to “Char2” (that is, p1.Char2) whose “PageID” is 1 in the physical table 111. Note that the DBMS 11 may extract a record with “PageID” of 1 and a record of 2, respectively, and the conversion unit 15 may combine and count to a virtual record.

Next, an example in which the tenant B transmits the following query 3A to insert a record (INSERT) will be described.
(Query 3A)
INSERT INTO item VALUES ('TV', 'Household appliances', 'Taiwan', 30000, 7/29);

The query 3A is converted into the following query 3B and query 3C by the converter 15. It is assumed that at the time when the request is received, records up to “DataID” 004 are registered in the physical table 111, and “005” is assigned to “DataID” of the newly inserted record.
(Query 3B)
INSERT INTO DataTable (TenantID, TypeID, DataID, PageID, Char1, Char2, Num1, Date1)
VALUES ('B', 'item', 005, 1, 'TV', 'Household appliances', 30000, 7/29);
(Query 3C)
INSERT INTO DataTable (TenantID, TypeID, DataID, PageID, Char1)
VALUES ('B', 'item', 005, 2, 'unit');

  In the case of record insertion, the conversion unit 15 refers to the metadata, and generates a query for insertion for each “PageID” when one record in the logical table is associated with a plurality of records in the physical table 111. In the above example, a query 3B for inserting a record (first record) with “PageID” “1” and a query 3C for inserting a record (second record) with “PageID” “2” are generated. ing.

  In the foregoing, an example of the data operation in the case where the record of the logical table is decomposed and associated with the plurality of records of the physical table 111 in the metadata has been described. When the record in the logical table is registered without being disassembled in the physical table 111 as in the case of the tenant A, the conditions of “TenantID” and “TypeID” are added and the column name is converted. “PageID” is only “1”, and it is not necessary to combine or decompose.

  As for the record update, the request is converted in the same manner as the selection, and a query for updating the record corresponding to the condition is generated. As for the record deletion, for example, a query for deleting a record having the same “DataID” value as that of the record corresponding to the condition is generated.

  Thereafter, the DBMS 11 of the multi-tenant system 1 performs data operation on the physical table 111 (S3). In this step, the data operation unit 112 of the DBMS 11 issues a converted query to the physical table 111 to perform data operations such as insertion, selection, update, and deletion, and outputs the execution result to the conversion unit 15. As described above, the conversion unit 15 and the DBMS 11 function as a “registration unit”, “extraction unit”, “update unit”, “deletion unit”, and the like according to the received request. Further, S2 to S4 of the data operation process function as a “registration step”, “extraction step”, “update step”, and “deletion step” according to the received request.

  Next, the conversion unit 15 converts the output result based on the metadata in the metadata storage unit 12, and outputs the result to the result response unit (S4). Here, for example, the column name of the physical table 111 is replaced with the column name of the logical table.

  Then, the result response unit 16 of the multi-tenant system 1 transmits the result to the requesting tenant terminal 2 via the network 3 (S5). Note that the result response unit 16 may pass the result to an application (not shown) of the multi-tenant system 1.

  As described above, in the multi-tenant system according to the present embodiment, the records of the logical table defined by each tenant are decomposed into a plurality of records as necessary and registered in the physical table. Therefore, even when a general-purpose physical table is prepared in the multi-tenant system, the restriction on the number of columns can be eliminated. In addition, the multi-tenant system according to the present embodiment registers each column in the logical table as an independent column in the physical table. Therefore, when using a function such as search provided by the DBMS, any column can be designated as a condition. As described above, in the general-purpose database used in the multi-tenant system, it is possible to eliminate the restriction on the number of columns and improve the usability of the functions provided by the DBMS.

<Setting process>
FIG. 10 is a process flow diagram illustrating an example of a setting process for setting metadata and the like. Prior to the data operation process described above, a setting process as shown in FIG. 10 is performed.

  The setting unit 13 of the multi-tenant system 1 receives a request from the tenant terminal 2 and sets the correspondence between the logical table column and the physical table 111 column in the metadata storage unit 12 (FIG. 10: S11). In this step, metadata as shown in FIG. 8 is registered.

In this embodiment, one of the features is that columns of various data types are prepared in the physical table 111. In this way, functions such as search and tabulation provided by the DBMS can be applied according to the data type of the column. Note that the data types set in the columns of the logical table and the data types set in the columns of the physical table do not necessarily have a one-to-one correspondence. For example, the data type between columns may be set as shown in FIG. Here, the “option type” in the logical table is a data type that can specify any one of a plurality of predefined options. For example, in the item indicating the priority, any one of predefined options such as “highest”, “high”, “medium”, “low”, etc. can be designated. The “reference type” in the logical table is a data type that represents a reference to another data. The “binary data type” in the logical table is a type that handles binary data such as the BLOB type prepared in the DBMS. In the multi-tenant system 1, a binary data storage unit (for example, a storage service on a file system or a cloud service) is separately prepared, and a LobID serving as a pointer to the storage unit is stored in a column in the physical table. It may be. Then, the storage means may be switchable to a DB or a file system depending on the setting. The “long text type” in the logical table is a data type that can store a character string that exceeds the upper limit of the data size of the character string type column of the DBMS. For example, it is possible to use CLOB, which is a DBMS native type, and make it switchable in the same manner as the binary data type.

  Note that all the columns of the physical table 111 may be of a character string type. In this case, the DBMS 11 stores all numerical values and dates in the physical table 111 as character strings. For example, processing such as aggregation is performed by an application (not shown) of the multi-tenant system 1 instead of using the DBMS function. If multiple columns of various data types are prepared in advance in the physical table 111, there is a possibility that the number of columns of unused data types will increase. For example, if all the columns are character string types, the columns of the physical table will be It can be used effectively. Even in such a case, if each column in the logical table is registered in an independent column in the physical table, a function provided by the DBMS such as search for each column can be used.

  Further, as described above, a plurality of columns having the same data type and different data sizes may be provided. For example, the physical table 111 is provided with a character string type (maximum length: 256 characters) column and a character string type (maximum length: 2000 characters) column. For example, the short character string type column of the logical table is mapped to the character string type (maximum length: 256 characters) column of the physical table 111, and the long character string type column of the logical table is mapped to the character string of the physical table 111. Map to column of type (maximum length: 2000 characters). By providing a plurality of columns having the same data type and different data sizes in the physical table 111, the user can select a column having the maximum length according to the stored data and is stored in the table. The amount of data can be reduced, and the buffer memory used during data processing can be saved.

  In the physical table 111, an index (INDEX) may be assigned to some columns in advance. The index can be assigned by using the function of the DBMS 11. In this case, in S <b> 11, a column that is assumed to be used for many of the logical table columns is mapped to an indexed column in the physical table 111. In this way, the search process can be speeded up in the data manipulation process shown in FIG.

  Depending on the DBMS method used, an index may be assigned to the null value to occupy the data area. Although the columns of the logical table are not necessarily mapped to the columns of the physical table according to the present embodiment, it is not preferable to provide an index in the case of such a DBMS. In such a case, an index table as shown in FIG. 12 may be provided separately. As the index table, for example, an index table for a character string type, an index table for a numeric type, and the like are provided according to the type of data type. Then, for the column to which index assignment is set in the metadata, the column value is copied (that is, synchronized) to the index table and used for the search.

Further, when receiving a request from the tenant terminal 2, the setting unit 13 sets a table space in the metadata storage unit 12 (S12). Here, the table space is a concept indicating each of a plurality of prepared physical tables. By preparing a plurality of table spaces, physical tables can be distributed like so-called partitioning.

FIG. 13 is a diagram for explaining the table space. As shown in FIG. 13, in the metadata, an association is set which table space each logical table uses. In addition, as many physical tables as the number of table spaces are prepared, and records of logical tables associated with metadata are registered. In the example of FIG. 13, “Default”
A physical table “DataTable_Default” is associated with the table space, and partitioning is performed using “TenantID” and “TypeID” as keys. In addition, a physical table called “DataTable_DateBase” is associated with the “DateBase” tablespace, and “Date1”
And partition with “TenantID” as a key.

  In such an example, it is assumed that the “DateBase” table space stores transaction data having items such as order dates and records increasing daily. In general, the response decreases as the number of records in the physical table increases. However, logical tables that are expected to increase in the number of records to be held can be managed separately on the physical table.

  Further, for each table space, the number of columns of each data type may be differentiated. For example, a table space giving priority to character storage (character storage priority type) and a table space giving priority to numerical value storage (numerical value storage priority type) are defined. In the character storage priority type physical table, for example, 80 character string columns, 10 numeric columns, and 10 date columns are provided. Also, the numeric storage priority type physical table is provided with, for example, 10 character string columns, 80 numeric columns, and 10 date polymorphic columns. When a user defines a logical table, the stored data has a clear view, so it is possible to select an appropriate table space, reduce the amount of data stored in the table, and use a buffer for data processing Memory can be saved.

<Others>
The present invention is not limited to the above-described examples, and various modifications can be made without departing from the gist of the present invention.

  For example, as the physical table 111, a main physical table and an auxiliary physical table may be provided. In this case, for example, the column of the first page (PageID is 1) is stored in the main physical table (main table), and the column of the second page and later (PageID is 2 or more) is an auxiliary physical table (auxiliary table). Remember me. In this way, the record of the main table and the record of the logical table are associated with each other one-to-one, and a unique index can be set for the column “DataID” of the main table. Further, for example, when the user directly refers to the DBMS by using SQL or the like without using the conversion unit 15 in FIG. 2, the number of records in the main table becomes the number of records in the logical table, so that maintenance and the like can be easily performed.

1 Multi-tenant system 11 DBMS
111 Physical table 112 Data operation unit 12 Metadata storage unit 13 Setting unit 14 Request reception unit 15 Conversion unit 16 Result response unit 2 Tenant terminal 3 Network

Claims (5)

An information processing apparatus that performs data operations on a single physical table shared by multiple tenants in a multi-tenant database,
Based correspondence between columns of logical tables and columns in the single physical tables defined for each tenant to mapping information defining for each of the tenant, the records in the logical table before SL single physical table The first record and the first record are decomposed into second records having different combinations of the data types of the columns, and the data of each column of the decomposed records in the logical table is divided into the first record and the first record. in the second record, and registers in the column of the same data type as the column of the decomposed records, registration and registers the grant information indicating the connection relationship between said first record and the second record And
Based on the mapping information and the information indicating the connection relationship, an extraction unit that reads the first record and the second record from the single physical table, combines them, and outputs them,
An information processing apparatus comprising: The information processing apparatus according to claim 1, wherein the single physical table includes a plurality of columns having different data types. When one record in the logical table is associated with three or more records in the single physical table in the mapping information,
The registration unit disassembles one record in the logical table into three or more records in the single physical table based on the mapping information, and gives information indicating a connection relation of the three or more records. Register,
The information processing apparatus according to claim 1, wherein the extraction unit reads the three or more records from the single physical table based on the mapping information, and outputs the combined records. An information processing method for performing data operations on a single physical table shared by multiple tenants in a multi-tenant database,
Based correspondence between columns of logical tables and columns in the single physical tables defined for each tenant to mapping information defining for each of the tenant, a record in the logic table, the in the single physical table The first record and the first record are decomposed into second records having different combinations of the data types of the columns, and the data of each column of the decomposed records in the logical table is divided into the first record and the first record. in the second record, and registers in the column of the same data type as the column of the decomposed records, registration and registers the grant information indicating the connection relationship between said first record and the second record Steps,
Based on the mapping information and the information indicating the connection relationship, an extraction step of reading the first record and the second record from the single physical table, combining them, and outputting them,
An information processing method in which a computer executes. A program that performs data operations on a single physical table shared by multiple tenants in a multi-tenant database,
Based correspondence between columns of logical tables and columns in the single physical tables defined for each tenant to mapping information defining for each of the tenant, a record in the logic table, the in the single physical table The first record and the first record are decomposed into second records having different combinations of the data types of the columns, and the data of each column of the decomposed records in the logical table is divided into the first record and the first record. in the second record, and registers in the column of the same data type as the column of the decomposed records, registration and registers the grant information indicating the connection relationship between said first record and the second record Steps,
Based on the mapping information and the information indicating the connection relationship, an extraction step of reading the first record and the second record from the single physical table, combining them, and outputting them,
A program that causes a computer to execute.

Images