JP2002041567A - Database managing method, device for executing the same, and recording medium on which processing program therefor is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology capable of improving the processing efficiency in database retrieval. SOLUTION: The database managing method for managing a database to perform retrieving processing, has a step for generating an identification number for identifying data, which are the target of retrieving processing, a step for reading the contents of additional information corresponding to the generated identification number while referring to an additional information bit map storing the additional information showing whether the data exist in the database or not, and a step for retrieving the relevant data when the read additional information shows the presence of these data, and suppressing the retrieval of the relevant data when the additional information does not show the presence of these data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management system for managing a database, and more particularly to a B-Tre.
The present invention relates to a technology effective when applied to a database management system for improving a search speed at the time of an index search using an e-index.

[0002]

2. Description of the Related Art Conventionally, "Lehman, P., Yao, SB: Efficient
Locking for Concurrent Operationson B-trees, ACM
Transactions on Database System Vol 6, No4 (1981), P.
As described in 650 to P.670 ", in the B-Tree search, the search is started from the root node, and it is checked whether or not the data to be searched is stored in the node of interest. If not, the child node expected to store the data to be searched is checked, and the above operation is repeatedly performed.
If the data does not exist in the e-index, that is, the data does not exist in the database, the search is performed up to the leaf node.

[0003]

In the above prior art, B
-Even if there is data that does not exist in the Tree index, the search is performed up to the leaf page. Therefore, if data that does not exist in the B-Tree index is searched, the cost for searching the B-Tree index increases, and the search efficiency decreases. There is a problem. An object of the present invention is to solve the above problem and provide a technique capable of improving the processing efficiency of database search.

[0004]

SUMMARY OF THE INVENTION The present invention relates to a database management system for managing a database in which a search process is performed, when a data to be searched exists in the database. Is what you do.

[0005] In the database management system of the present invention,
When creating an index, an identification number for identifying data to be indexed is generated, and additional information indicating that the data exists in the database corresponds to the generated identification number. To create the additional information bitmap.

When performing an index search, an identification number for identifying data to be searched is generated, and the data is referred to by referring to an additional information bit map corresponding to the generated identification number. The contents of the additional information indicating whether or not the information exists in the database are read. If the read additional information indicates the existence of the data, the data is searched. If the additional information does not indicate the presence of the data, the search of the data is suppressed.

As described above, in the present invention, data that does not exist in the index is not searched until the leaf page, but data that needs to be searched is distinguished from unnecessary data, and only the data that needs to be searched is indexed. Search for. As a result, it is possible to reduce the cost at the time of the conventional index search in which even the data that does not exist in the index is searched up to the leaf page.

As described above, according to the database management system of the present invention, when data to be searched is present in the database, the data is searched, so that the processing efficiency of the database search is improved. It is possible to improve.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A database management system according to an embodiment for managing a database in which a search process using a B-Tree index is performed will be described below.

FIG. 1 is a diagram showing an outline of a process of searching for corresponding data by accessing a B-Tree index from an additional information bit map for certain data according to the present embodiment. As shown in FIG. 1, in the additional information bitmap 111 corresponding to the data stored in the index of the present embodiment on a one-to-one basis, when the corresponding data is stored in the B-Tree index 112, the bit corresponding to the data is stored. Is set to "1", and when the data is not stored, "0" is set to a bit corresponding to the data. When the corresponding bit of the additional information bit map 111 is “1”, the B-Tree index 112 is searched, and the corresponding data is searched from the table 113 in the database.

FIG. 2 is a diagram showing a schematic configuration of the database management system of the present embodiment. As shown in FIG. 2, the database management device 211 of the present embodiment includes an index creation processing unit 200, a data search processing unit 201, a data insertion processing unit 202, a data deletion processing unit 203, and an identification number generation processing unit 204. And additional information setting processing unit 205
And an additional information reference processing unit 206.

[0012] The index creation processing section 200 is a BT
When creating the REE index 216, the processing unit creates an additional information bitmap 214 storing additional information indicating that the data exists in the database.

When performing a data search, if the additional information corresponding to the data to be searched indicates that the data is present, the data search processing unit 201 searches the data. When the additional information does not indicate the existence of the data, the processing unit suppresses the search for the data.

The data insertion processing unit 202 converts the data into a table R
The processing unit sets the additional information indicating that the data is present in the database in the additional information bitmap 214 when inserting the data into the database 217. Data deletion processing unit 2
03, when the data is deleted from the table R217 and the identification number of the data is a number for uniquely identifying the data within a range that the data can take, the data exists in the database. This is a processing unit that sets additional information indicating that no additional information is stored in the additional information bitmap 214.

The identification number generation processing unit 204 uniquely identifies the data within a range that the data can take as an identification number for identifying data to be subjected to database processing such as index creation and index search. This is a processing unit that generates a number for performing the operation or a number indicating the hash value of the data.

An additional information setting processing unit 205 sets additional information indicating whether or not certain data exists in a database in an additional information bitmap 214 corresponding to an identification number for identifying the data. It is.

The additional information reference processing unit 206 refers to an additional information bitmap 214 storing additional information indicating whether or not the data exists in the database, and generates additional information corresponding to an identification number generated for a certain data. Is a processing unit that reads out the contents of.

The database management device 211 includes an index creation processing unit 200, a data search processing unit 201, a data insertion processing unit 202, a data deletion processing unit 203, an identification number generation processing unit 204, an additional information setting processing unit 205, and an additional information reference process. A program for functioning as the unit 206 is recorded on a recording medium such as a CD-ROM, stored on a magnetic disk or the like, and then loaded into a memory and executed. The recording medium for recording the program is C
A recording medium other than the D-ROM may be used.

In FIG. 2, a database management device 211 comprising a CPU 212, a memory 213, and an external storage device 215
2 shows a configuration of a database management system that stores a table R217 and a B-Tree index 216 and performs a search by adding an additional information bitmap 214 to the B-Tree index 216.

In FIG. 2, the B-Tree index 216 corresponds to the column Rk of the table R217 (the total number of rows | R | of the table R217; hereinafter, the total number of rows of the table is represented as | table |).
Is the index of the definition that has at the beginning of the constituent sequence.
At this time, certain data is stored in the B-Tree index 21.
An additional information bitmap 214 that is used as a constituent column of No. 6 and indicates whether it exists in the column data of the database is added. Hereinafter, a process of creating the additional information bitmap 214 at the time of creating an index in the database management system of the present embodiment will be described.

FIG. 3 is a flowchart showing a processing procedure when creating the additional information bitmap according to the present embodiment. As shown in FIG. 3, when creating the B-Tree index 216, the index creation processing unit 200 of the database management device 211 according to the present embodiment determines that the data constituting the B-Tree index 216 exists in the database. To create an additional information bitmap 214 storing additional information indicating that the

When the index creation processing is started in step 311, the index creation processing unit 200 of the database management device 211 determines in step 312 that the B-Tre
The e index 216 is created, and the process proceeds to step 313.

In step 313, it is determined whether or not designation of creating the additional information bitmap 214 has been performed by the database administrator.
If creation of the file No. 14 is not specified, step 31
In step 4, the creation of the B-Tree index 216 is completed.
If the creation of the additional information bitmap 214 is specified, the process proceeds to step 321.

At step 321, it is determined whether the number of possible combinations of data stored in the B-Tree index 216 is the number of bits that can be expanded in the memory 213, and the number of possible combinations of the data is determined as the number of bits that can be expanded in the memory 213. If not, the process proceeds to step 331.

In step 331, the identification number generation processing unit 204 and the additional information setting processing unit 205
The additional information bitmap 214 corresponding to the data that the ee index 216 can take one-to-one is created. That is, the identification number generation processing unit 204 generates, as an identification number for identifying the data for which the index is to be created, a number for uniquely identifying the data within a range that the data can take. For example, when the data type of the data is SMALLINT (signed integer in a 2-byte binary format), there are 65536 kinds of possible values, and therefore, from "1" to 655 in order from the minimum value of the data.
A process for generating identification numbers up to No. 36 is performed. Then, the additional information setting processing unit 205 sets “1” as additional information indicating that the data exists in the database,
It is set in the additional information bitmap 214 corresponding to the generated identification number.

As a result of the determination at step 321, B-Tr
If the number of possible combinations of data stored in the ee index 216 is larger than the number of bits that can be expanded in the memory 213, the process proceeds to step 332.

In step 332, the data is classified by the identification number generation processing unit 204 and the additional information setting processing unit 205 by an appropriate hash function whose hash value is within the number of bits that can be expanded in the memory 213, and the hash value An additional information bitmap 214 having a bit size is created. That is, in step 332, the identification number generation processing unit 204 of the database management device 211
A number indicating a hash value of the data is generated as an identification number for identifying the data to be indexed, and the additional information setting processing unit 205 determines that the data exists in the database. “1” is set as the additional information to be set in the additional information bitmap 214 corresponding to the generated identification number.

The additional information bit map 2 of the present embodiment
14 is set with a flag for distinguishing between an additional information bitmap corresponding to data and an additional information bitmap based on a hash value.

FIG. 4 is a diagram showing an example of the additional information bitmap 214 according to the present embodiment. FIG. 4 shows an example of an additional information bitmap 411 corresponding to data and an additional information bitmap 421 based on a hash value of a hash function.

In FIG. 4, the additional information bitmap 421 based on the hash value indicates that one of a plurality of different data having the same hash value is currently stored in the B-Tree index 216 or stored in the past. If so, the bit corresponding to the hash value becomes “1”. Further, if any one of a plurality of different data having the same hash value is not currently stored and has not been stored in the past, the bit corresponding to the hash value becomes “0”.

A process for performing a join search of two tables using the additional information bitmap 214 added to the B-Tree index 216 in the database management system of the present embodiment will be described below.

FIG. 5 is a diagram showing an outline of a join search of two tables stored in the database management system of this embodiment. In FIG. 5, table S511 (the total number of rows | S |)
Is stored in the database management system, and the table R2
When performing a join search between Table 17 and Table S511, first, Table S5
11, the condition for searching the table S511, that is, the condition including only the column of the table S511 (for example, the table S51)
As conditions for the column Sn of 1, Sn = 10 or Sn> 10
Table S′512 (the total number of rows | S ′ |), which is data corresponding to the above conditions).

Then, the retrieved data S 'and the table R21
Then, the data of No. 7 is matched, and a row satisfying the condition of the join search is searched from the database management system. Among the conditions for the join search, the conditions for matching table R217 with table S'512 include column Sk of table S'512 and column Rk of table R217 (for example, conditions such as Sk = Rk and Sk = Rk + 10). Shall be included.

FIG. 6 is a flowchart showing a processing procedure of a search process using the additional information bitmap 214 according to the present embodiment. In FIG. 6, a set of search rows including the column Sk of the table S511, that is, the table S'512 is added to the column Rk of the table R217.
B-Tree index 21 having k at the beginning of the constituent column
6 shows a process of performing a search using the additional information bitmap 214.

At step 612, the database management device 2
The eleventh data search processing unit 201 first extracts the data of S ′, and in step 613, determines whether the B-Tree index 216 can be used to search the table R217. If the B-Tree index 216 cannot be used, the data of the table R217 in the database is searched in step 614. If the B-Tree index 216 can be used, the process proceeds to step 621.

In step 621, the additional information bitmap 214 is assigned to the B-Tree index 216.
Is checked, and if the additional information bitmap 214 does not exist for the B-Tree index 216, the process proceeds to step 633.

At step 633, a search is made as to whether the data extracted from S 'exists in the B-Tree index 216. In step 634, as a result of the search in step 633, it is checked whether or not the data extracted from S ′ exists in the B-Tree index 216.
If the data does not exist in the B-Tree index 216, the process returns to step 612 to retrieve the data from S '.
Proceed to.

In step 635, it is determined whether only the search for the B-Tree index 216 is sufficient. B-Tre
The condition determination for the search is completed only with the columns constituting the e-index 216, and the data to be searched is B-Tree.
If there is only the constituent column of the index 216, B-Tr
A search using only the e-index 216 is sufficient. Therefore, B-
If only the search of the tree index 216 is sufficient, the process proceeds to step 615 and ends the search. If data that does not exist in the B-Tree index 216 is to be searched, the table R217 in the database is searched in step 614, and the search is terminated.

If it is determined in step 621 that the additional information bitmap 214 exists for the B-Tree index 216, the process proceeds to step 622, where the additional information bitmap 214 is added to the additional information bitmap based on the hash value of the data in the index. 421 is determined.

At step 622, additional information bit map 2
If it is determined that 14 is the additional information bitmap 421 based on the hash value, the process proceeds to step 631. In step 631, the identification number generation processing unit 204 obtains a hash value using a column corresponding to the index configuration column of the table R217 among the data extracted from S ′.

At step 632, the additional information reference processor 20
No. 6 refers to the additional information bitmap 214 and checks whether the corresponding bit of the hash value-th additional information bitmap 421 is “1”. If the corresponding bit is "0", there is no corresponding data in the B-Tree index 216, so the process returns to step 612 to extract the data from S '. If the corresponding bit is "1", the data is included in the B-Tree index 216. Since there is a possibility that the corresponding data exists, the process proceeds to step 633.

In step 633, the data search processing unit 201
Performs a search of the B-Tree index 216. In step 634, if the same value as the data extracted from S ′ does not exist as a result of the search, step 612
Return to step 635 if the same value exists.
Proceed to. In step 635, it is determined whether only the search of the B-Tree index 216 is sufficient. If the search is sufficient, the search is terminated. If the other data is to be searched, the table R21 in the database is determined in step 614.
After searching the data No. 7, the search is terminated.

At step 622, additional information bit map 2
If it is determined that 14 is not the additional information bitmap 421 based on the hash value, but the additional information bitmap 411 corresponding to the data on a one-to-one basis, the process proceeds to step 641.

In step 641, the identification number generation processing section 20
4 generates, as an identification number for identifying the data to be subjected to the search processing, a number for uniquely identifying the data within a range that the data can take. The number of the value of the data of the B-Tree index 216 in the table R217 is obtained. This is the position of the corresponding bit in the additional information bit map 411.
For example, when the data type of Rk is SMALLINT (a 2-byte binary signed integer) and the data is ordered from the minimum value, if the value of the data extracted from S ′ is “0”, the data is It is 32,969th.

In step 642, the additional information reference processor 20
6 refers to the additional information bitmap 411 and checks whether the corresponding bit is “1”. The corresponding bit is "0"
If there is no corresponding data in the B-Tree index 216, the process returns to step 612 to retrieve data from S ′. If the corresponding bit is “1”, the B-Tree
Since the corresponding data exists in the e index 216, the process proceeds to step 643.

In step 643, the data search processing unit 201
Performs a search of the B-Tree index 216. In step 644, it is determined whether or not only the search of the B-Tree index 216 is sufficient. If the search is sufficient, the search is terminated.
After searching the data of the table R217 in the database in step 614, the search is terminated.

FIG. 7 is a flowchart showing a processing procedure for inserting a row into a table in which the additional information bitmap 214 exists in the B-Tree index 216 according to the present embodiment. As shown in FIG. 7, the data insertion processing unit 202 of the database management device 211 of the present embodiment stores the data in the table R21.
7, the additional information indicating that the data is present in the database is added to the additional information bitmap 2.
14 is performed.

In FIG. 7, when inserting a row into the table R217, the data to be inserted is passed to the database management unit 211. In step 712, the data insertion processing unit 202 of the database management device 211 inserts the passed data into the database of the table R217.

Next, in step 713, B is added to the table R217.
It is checked whether or not the Tree index 216 is defined. If not defined, the insertion processing is terminated in step 714. If the B-tree index 216 is defined, the data is stored in the B-tree index 216 in step 721. Add.

Next, at step 722, the BT
It is checked whether or not the additional information bitmap 214 exists for the REE index 216.
If 14 does not exist, the insertion process ends at step 714, and if the additional information bitmap 214 exists, the process proceeds to step 731.

In step 731, it is determined whether or not the additional information bitmap 214 is an additional information bitmap 421 based on the hash value of the data in the B-Tree index 216. If it is 421, the process proceeds to step 741.

At step 741, the identification number generation processing section 20
No. 4 generates a number indicating a hash value of the data as an identification number for identifying the data to be subjected to the insertion processing. In step 742, the additional information setting processing unit 20
5 sets the bit of the generated hash value-th additional information bitmap 421 to “1”. Then, the insertion process ends in step 714.

If it is determined in step 731 that the additional information bitmap 214 is not the additional information bitmap 421 based on the hash value but the additional information bitmap 411 corresponding to the data on a one-to-one basis, step 7 is executed.
Go to 51.

In step 751, the identification number generation processing section 20
No. 4 generates, as an identification number for identifying the data to be subjected to the insertion processing, a number for uniquely identifying the data within a range that the data can take, and the data is stored in B in Table R217. -Determine the number of data of the tree index 216. In step 752, the additional information setting processing unit 205 sets the bit of the additional information bitmap 411 corresponding to the generated number to “1”. Then, in step 714, the insertion process ends.

FIG. 8 is a flowchart showing a processing procedure for deleting a row from a table in which the additional information bitmap 214 exists in the B-Tree index 216 according to the present embodiment. As shown in FIG. 8, the data deletion processing unit 203 of the database management device 211 of the present embodiment stores the data in the table R2.
17, when the identification number of the data is a number for uniquely identifying the data within a range that the data can take, an addition indicating that the data does not exist in the database. A process for setting information in the additional information bitmap 214 is performed.

In FIG. 8, when a row is deleted from the table R217, first, at step 812, the data deletion processing unit 203 of the database management apparatus 211 stores the data in the table R2.
17 from the database. Next, step 813
Then, it is checked whether or not the B-Tree index 216 is defined in the table R217. If the B-Tree index 216 is not defined, the deletion process ends in step 814. If the B-Tree index 216 is defined, -Maintenance of the Tree index 216 is performed.

Next, in step 822, it is checked whether or not there is a line of the same data as the deleted data in the B-Tree index 216. If there is a line of the same data as the deleted data, it is deleted in step 814. When the process is terminated and there is no row having the same data as the deleted data, the process proceeds to step 823.

At step 823, it is checked whether or not the additional information bitmap 214 exists for the B-Tree index 216. If the additional information bitmap 214 does not exist, the deletion process is terminated at step 814, and if the additional information bitmap 214 exists. Go to step 824.

In step 824, it is determined whether or not the additional information bitmap 214 is an additional information bitmap 421 based on the hash value of the data in the B-Tree index 216.

If it is determined in step 824 that the additional information bitmap 214 is the additional information bitmap 421 based on the hash value, there is a possibility that there is different data that can take the same hash value. The process of setting the corresponding bit of the bitmap 421 to “0” is not performed, and the deletion process ends. In this case, B-Tr
Even though there is no data in the ee index 216, the corresponding bit of the additional information bit map 421 may be "1".
The re-creation of the additional information bitmap 421 is performed at a predetermined timing, for example, when optimizing e.

As a result of the determination in step 824, if the additional information bitmap 214 is not the additional information bitmap 421 based on the hash value but the additional information bitmap 411 corresponding to the data on a one-to-one basis, the flow proceeds to step 831.

At step 831, the identification number generation processing section 20
No. 4 generates, as an identification number for identifying the data to be deleted, a number for uniquely identifying the data within a range that the data can take. -The number of possible types of data of the tree index 216 is determined. Step 83
In step 2, the additional information setting processing unit 205 sets the bit of the additional information bitmap 411 corresponding to the generated number to “0”.
Set to. Then, in step 814, the deletion process ends.

FIG. 9 is a flowchart showing a procedure of updating a row in a table in which the additional information bitmap 214 exists in the B-Tree index 216 according to the present embodiment. In FIG. 9, when a row is updated in the table R217, the data is updated in the database management device 211. In this case, first, data deletion processing is performed in step 912, and then data insertion processing is performed in step 913. At this time, the data deletion processing is performed according to the processing procedure illustrated in FIG. 8, and the data insertion processing is performed according to the processing procedure illustrated in FIG.

FIG. 10 shows a table of this embodiment and a B-Tree.
FIG. 11 is a diagram illustrating a configuration of a database management system that stores an index 216 divided into a plurality of database management apparatuses and performs a search by adding an additional information bitmap 214 to the B-Tree index 216. In the configuration shown in FIG. 10, it is assumed that the data of the table R217 (the total number of rows | R |) is divided and stored in the n database management apparatuses 211 as the table R1012.

Here, a column used to divide the table R217 is denoted by Rm, and a B-Tr having a column Rk different from the column Rm used to divide the data in the table R217 at the head of the constituent columns is used.
The ee index 216 is defined. At this time, an additional information bitmap 214 indicating whether or not data exists in the B-Tree index 216 (local index) managed by each database
-Add to the Tree index 216.

FIG. 11 shows a table R1012 and a table S511 divided into a plurality of database management apparatuses 211 according to the present embodiment.
It is a figure which shows the outline | summary of the joint search of. In FIG. 11, it is assumed that table S511 (the total number of rows | S |) is stored in the database management system.

When performing a join search between the table R1012 and the table S511, first, of the data in the table S511, the table S511
Table S'51 which is data corresponding to the condition for retrieving
Search for 2. Then, the retrieved data S 'and table R
The data of 1012 is matched, and a row that satisfies the condition of the join search is searched from the database management system.

Among the conditions for the join search, Table R1012 and Table S
The condition for matching 511 is column Sk in table S511.
And the column Rk of the table R1012, and the table R10
It is assumed that the column Rm used for dividing into 12 is not included. At this time, the column Rm used for dividing the table R1012
Since the first Rk of the index configuration columns defined in the table R1012 and the table R1012 are different columns, the entire database management apparatus 211 having the table R1012 which is a partial table obtained by dividing the table R217
To all the data in S '.

Attention is paid to a certain database management device 211 having a partial table of Table R217. Additional information bitmap 2
Reference numeral 14 denotes information closed in each database management device 211 and does not depend on the state of other database management devices 211. Therefore, the noted database management device 211
In this case, the same processing as when stored in one database management device 211 can be performed. As described above, in this embodiment, each database management device 211 does not rely on the processing of another device, and
4 can be performed.

FIG. 11 shows additional information bit map 2
14 is stored in the memory 213,
Additional information bitmap 21 if the size does not fit inside
4 may be stored in the external storage device 215. In this case, the additional information bitmap 214 is divided into several sections, and the memory 21 is divided into divided sections.
3 and the external storage device 215
The above section additional information bitmap is stored in LRU (Least Rece
(ntly Used) method. A part or all of the additional information bitmap 214 is stored in the external storage device 2
15 is a B-Tree index 2
At the time of the definition of 16 or the definition of the schema, it is assumed that the operation is performed in a part that defines the operation of the database management system or in a manner that the database management system can recognize.

Further, when the number of possible combinations of the data constituting the B-Tree index 216 is large, the range of the data for creating the additional information bitmap 214 is limited so that the frequently referenced data can be reduced. Only the additional information bitmap 214 can be created to reduce the size of the memory 213 or the external storage device 215. At this time, instead of classification by the hash function, the data and the B-Tree index 216 are 1
Since the additional information bitmap 411 corresponding to one-to-one correspondence can be created, the search speed can be improved. The specification for limiting the data range of the additional information bitmap 214 is performed when the B-Tree index 216 is defined or when the schema is defined, a part that defines the operation of the database management system, or any other method that can be recognized by the database management system. It shall be performed in.

In addition to the B-Tree index, it is possible to reduce the search cost of the index by using the additional information bitmap for the index of another method. The specification indicating that the additional information bitmap is to be used is performed at the time of defining an index or a schema, at a part that defines the operation of the database management system, or in a manner that can be recognized by the database management system. .

Further, even for a table having no index, by using an additional information bitmap indicating whether or not data exists in the table, the table is not searched when data does not exist. Can be reduced. The specification indicating that the additional information bitmap is used for a table where no index exists can be specified when defining an index or schema, a part that regulates the operation of the database management system, or other parts that can be recognized by the database management system. It shall be performed in a simple manner. Similarly, the additional information bitmap can be applied to an index of a database management system that does not hold data in a table format.

As described above, according to the database management system of the present embodiment, when data to be searched is present in the database, the data is searched. It is possible to improve efficiency.

[0075]

According to the present invention, when data to be searched is present in the database, the data is searched, so that the processing efficiency of the database search can be improved. .

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an outline of a process of accessing a B-Tree index from an additional information bit map and searching for corresponding data according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a schematic configuration of a database management system according to the embodiment;

FIG. 3 is a flowchart illustrating a processing procedure when creating an additional information bitmap according to the embodiment;

FIG. 4 is a diagram illustrating an example of an additional information bitmap 214 according to the present embodiment.

FIG. 5 is a diagram showing an outline of a join search of two tables stored in the database management system of the embodiment.

FIG. 6 is a flowchart illustrating a processing procedure of a search process using the additional information bitmap 214 according to the present embodiment.

FIG. 7 illustrates a B-Tree index 216 according to the present embodiment.
9 is a flowchart showing a processing procedure for inserting a row into a table in which an additional information bitmap 214 exists.

FIG. 8 shows a B-Tree index 216 according to the present embodiment.
13 is a flowchart showing a processing procedure for deleting a row from a table in which an additional information bitmap 214 exists.

FIG. 9 shows a B-Tree index 216 according to the present embodiment.
9 is a flowchart showing a processing procedure for updating a row in a table in which an additional information bitmap 214 exists.

FIG. 10 is a diagram illustrating a configuration of a database management system in which a table and a B-Tree index 216 according to the present embodiment are divided and stored in a plurality of database management apparatuses, and a search is performed by adding an additional information bitmap 214 to the B-Tree index 216; FIG.

FIG. 11 shows a plurality of database management apparatuses 2 of the present embodiment.
11 is a diagram illustrating an outline of a join search of a table R1012 and a table S511 divided into 11; FIG.

[Explanation of symbols]

111: additional information bitmap, 112: B-Tree
Index, 113 ... Table in database, 211 ...
Database management device, 212 CPU, 213 memory, 214 additional information bitmap, 215 external storage device, 216 B-Tree index, 217 table R, 200 index creation processing unit, 201 data search processing unit , 202: data insertion processing unit, 203: data deletion processing unit, 204: identification number generation processing unit, 205:
Additional information setting processing unit, 206... Additional information reference processing unit, 4
11 additional information bitmap, 421 additional information bitmap, 511 table S, 512 table S ', 1011 network, 1012 table R.

Claims (5)

[Claims] 1. A database management method for managing a database on which a search process is performed, comprising: a step of generating an identification number for identifying data to be searched; and whether the data exists in the database. Referring to an additional information bitmap storing additional information indicating whether or not the content of the additional information corresponding to the generated identification number is read; if the read additional information indicates the presence of the data, Performing a data search, and, when the additional information does not indicate the existence of the data, suppressing the data search. 2. A step of generating an identification number for identifying data to be indexed, and adding additional information indicating that the data is present in a database to the generated identification number. Setting a corresponding additional information bitmap into a corresponding additional information bitmap. 3. The identification information according to claim 1, wherein the identification information is a number for uniquely identifying the data within a range that the data can take or a number indicating a hash value of the data. 2. The database management method according to any one of 2. 4. A database management system for managing a database on which a search process is performed, wherein: an identification number generation processing unit for generating an identification number for identifying data to be searched; An additional information reference processing unit for reading the contents of the additional information corresponding to the generated identification number by referring to the additional information bitmap storing the additional information indicating whether the data is present; A data search processing unit that searches for the data when the data indicates presence, and suppresses the data search when the additional information does not indicate the data. . 5. A computer-readable recording medium storing a program for causing a computer to function as a database management system for managing a database on which a search process is performed, for identifying data to be searched. An identification number generation processing unit that generates an identification number, and an additional information bitmap that stores additional information indicating whether data is present in the database, refer to the contents of the additional information corresponding to the generated identification number. The additional information reference processing unit to be read, and if the read additional information indicates the existence of the data, search the data. If the additional information does not indicate the existence of the data, search the data. A program for causing a computer to function as a data search Recording medium characterized.