KR20000037515A - Method for composing b+ tree to manage history - Google Patents

Abstract

PURPOSE: A method for composing B+ tree to manage history is provided to construct a database easily by supporting a storage of item information structurally in a modified B+ tree. CONSTITUTION: Conventional record information is revised(S1). A current record is searched as a primary key using a recording searching progress of a conventional B+ tree. A revising record is searched according a record search. A current record is inserted into an item data structure using an item key. New information is added to the current record. Conventional record item information is searched(S2). An item database is searched through a pointer for indicating item information of a record. Necessary information is searched using the item key according to a kind of a corresponding item database. Conventional records are deleted(S3).

Description

How to configure a non-plus tree for history management

The present invention relates to a method of constructing a B + tree for history management. In particular, the present invention relates to a method of constructing a database using a data structure called a B + tree, which is widely used as a data storage and retrieval structure. The logical data structure is modified by modifying the B + tree so that it is easy to apply to the general information about each record of the type of the database and the database that needs to store and retrieve the details about the information that changes frequently. It is about how to present.

1 is a block diagram of a method for presenting a change data structure for managing history according to the prior art. As shown, the existing B + tree data structure is a logical data storage form of the tree structure. The tree structure consists of one unit of data, one record, which contains several child leaves, and each record is stored in a tree structure according to one or several key values. It is a structure that can insert and delete functions relatively easily and quickly. The B + tree data structure is a further development of the existing tree structure. Thus, one leaf 100 includes several records 110 and 120, and one leaf 100 has more '1' pointers than the number of records 110 and 120 that it contains. . Thus, the pointer is an advanced tree structure that points to its children. This variant has the advantage of making the search much faster by significantly reducing the depth of the tree, that is, the stage of generation. This advantage is even more pronounced as the number of records 110, 120, 210, 220, 310, 320, 410 and 420 of the database is huge. Therefore, it is widely used as the data structure of various database systems today. In general, however, this structure is generally used when the size of the records 110, 120, 210, 220, 310, 320, 410, and 420 in the database is static rather than dynamic. It is a suitable structure. In many databases, each record 110, 120, 210, 220, 310, 320, 410, 420 of the B + tree is used in a fixed size. The reason for this is that a pointer may be used to retrieve the child, but this fixed record 110, 120, 210, 220, 310, 320, 410, 420 can be retrieved by calculation. Because sometimes. This also reduces the memory space used for pointers. However, this static nature may not be appropriate for certain kinds of databases.

Existing B + tree data structures can be searched relatively quickly compared to other data structures, even if they are records 110, 120, 210, 220, 310, 320, 410 and 420 in large databases. have. And storing and inserting new records 110, 120, 210, 220, 310, 320, 410, 420 and current records 110, 120, 210, 220, 310 Deletion of (320) (410) 420 is not used in many databases because it is not very complicated. However, in the existing B + tree data structure, the information about each record 110, 120, 210, 220, 310, 320, 410, and 420 changes from time to time in the database. However, the changing information is not overwritten by the current records 110, 120, 210, 220, 310, 320, 410 and 420. If the previous information needs to be stored instead, the records 110, 120, 210, 220, 310, 320, 410 and 420 will continue to grow in size. Each record 110, 120, 210, 220, 310, 320, 410, 420 has a different size. This results in a static structure of the B + tree. Therefore, it is not suitable for a reason. Reasons for not being suitable are as follows. The records 110, 120, 210, 220, 310, 320, 410 and 420 of many B + trees are fixed. However, if one uses a method of dynamically allocating memory to increase the current records 110, 120, 210, 220, 310, 320, 410 and 420, one record ( There is a problem that the 110, 120, 210, 220, 310, 320, 410 and 420 may be too large.

The present invention has been devised to solve the problems of the prior art operating as described above, and suggests a logical data structure for data storage and retrieval of a database storing a record managing history, and a B + tree for history management. It is an object to provide a configuration method.

Other objects and advantages of the present invention will become more apparent from the following detailed description of the invention and the accompanying drawings.

1 is a block diagram showing a method of presenting a change data structure for history management according to the prior art;

2 is a block diagram of a method of presenting a changed data structure for managing details according to the present invention;

3 is a flowchart illustrating a method of presenting a change data structure for managing history according to the present invention.

<Explanation of symbols for main parts of drawing>

Leaves: 100, 200, 300, 400, 500, 600, 700, 800

Records: 110, 120, 210, 220, 310, 320, 410, 420, 510, 520, 610, 620, 710, 720, 810, 820

515, 525, 615, 625, 715, 725, 815, 825: record history information

The method for constructing a history management B + tree according to the present invention, which is designed to achieve the above object, in the method for improving the structure of a B + tree, which is a logical structure for data storage,

Retrieving additional structures relating to the management portion of the history information of each record and corresponding history information;

Modifying current information;

Retrieving, inserting, or deleting general information about one record; And

In the case of changing the information about one existing record, a step of storing and retrieving the existing information is made in addition to the new information to be changed.

In a preferred embodiment of the present invention, the step of modifying the existing record information,

Retrieving the current record with the primary key using a record retrieval procedure of an existing B + tree;

Searching for the record to be modified by the record search,

The retrieving of the record to be modified by the record retrieval includes: newly inserting a current record into a history data structure using a history key;

Inserting the record and overwriting new information in the current record,

The step of retrieving the existing record history information includes: retrieving a history database through a pointer indicating the record information of the record;

And after searching the database, searching for necessary information using a history key according to the type of the history database.

The deleting of the existing record comprises deleting not only the searched record but the corresponding record after deleting the history information indicated by the record.

In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. Terms to be described later are terms defined in consideration of functions in the present invention, and may be changed according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the specification.

2 is a block diagram of a method for presenting a change data structure for managing details according to the present invention. As shown, in the modified B + tree data structure, an additional structure relating to the management portion of the historical information of each record 510, 520, 610, 620, 710, 720, 810, 820 And the retrieval procedure of the historical information and the correction procedure of the current information. The unique structure of the rest of the B + tree is a well known structure. In addition, the process of searching for, inserting, or deleting general information about one record 510, 520, 610, 620, 710, 720, 810, 820 is identical. . However, when changing information on one existing record, the structure and procedure of storing the existing information in addition to the new information to be changed, and the procedure of retrieving such information become elements of the invention.

In the modified B + tree data structure of the present invention, each record 510, 520, 610, 620, 710, 720, 810, and 820 also has one pointer (hereinafter referred to as B pointer). Have more Each record 510, 520, 610, 620, 710, 720, 810, 820 is sequentially stored based on a certain field value of 'key' like the existing B + tree structure. The A pointer (an existing pointer) acts like an existing pointer. That is, a pointer to a child leaf. The B pointer is a pointer to another data structure that stores historical information of the particular record, that is, history information 515, 525, 615, 625, 715, 725, 815, and 825. Another data structure is based on a second key that can easily store data. The list information may be a linked list according to the size of the details information 515, 525, 615, 625, 715, 725, 815, and 825. Or it could be another B + tree structure. However, it must be a dynamic data structure in which the number of information can grow, not a static structure.

Hereinafter, with reference to the accompanying drawings will be described in detail the operating principle of the preferred embodiment of the present invention.

The operation retrieves the latest information of the records 510, 520, 610, 620, 710, 720, 810, 820. In addition, insertion and deletion of new records 510, 520, 610, 620, 710, 720, 810 and 820 are performed by the same procedure as the existing B + tree. When modifying existing record information and when retrieving existing record history information 515, 525, 615, 625, 715, 725, 815, 825, and existing records ( 510, 520, 610, 620, 710, 720, 810, and 820 will be described by dividing them into steps of deleting. When modifying existing record information, the primary key can be retrieved using the records 510, 520, 610, 620, 710, 720, 810, and 820 retrieval procedures of the existing B + tree. It searches for the current record 510, 520, 610, 620, 710, 720, 810, 820. When the record 510, 520, 610, 620, 710, 720, 810, and 820 to be modified is found, the record is newly stored in the history data structure using the history key. Then overwrite the current record with the new information. When retrieving historical information of an existing record (515) (525) (615) (625) (715) (725) (815) (825), the records (510) (520) are processed in the same manner as the existing B + tree. Search for (610) 620, 710, 720, 810, and 820. Then, the history database is searched through a pointer indicating the history information 515 (525) 615 (625) 715 (725) 815 (825) 825 of the record. After retrieving, the information 515, 525, 615, 625, 715, 725, 815, and 825 are retrieved using the history key according to the type of the history database. Finally, when deleting existing records 510, 520, 610, 620, 710, 720, 810 and 820, records 510 and 520 like the existing B + tree are likewise deleted. Search for 610, 620, 710, 720, 810, and 820. Then, the searched records 510, 520, 610, 620, 710, 720, 810, 820, as well as the searched records 510, 520, 610, 620, 710 ( The history information 515, 525, 615, 625, 715, 725, 815, 825 indicated by 720, 810, and 820 is deleted. After this step, delete the record.

This procedure results in historical information 515, 525, 615, 625, 715, 725 of the current record 510, 520, 610, 620, 710, 720, 810, 820. 815, 825, the historical information 515, 525, 615, 625, 715, 725, 815, 825, which is generally passed by, is invariant. Thus, historical databases can be stored on less expensive, slower access mass storage devices, such as read-only optical file systems, rather than hard disks. This is a time consuming task of finding the corresponding records 510, 520, 610, 620, 710, 720, 810, and 820, and in general, historical information 515, 525. 615, 625, 715, 725, 815, 825 compared to the number of records 510, 520, 610, 620, 710, 720, 810, 820 in the database. It is assumed that the search cost is insignificant. 3 is a flowchart illustrating a method of presenting a change data structure for managing details according to the present invention. The method modifies existing record information (S1). The existing record detail information is retrieved (S2). In addition, the existing record is deleted (S3).

The present invention can be variously modified and can take various forms and only the specific embodiments thereof are described in the detailed description of the invention. It is to be understood, however, that the present invention is not limited to the specific forms mentioned in the detailed description of the invention, but rather includes all modifications, equivalents, and substitutions within the spirit and scope of the invention as defined by the appended claims. It should be understood to do.

In the present invention operating as described above, the effects obtained by the representative ones of the disclosed inventions will be briefly described as follows.

If the data structure of the present invention is used to store the record information together, the existing B + tree can be used. However, the modified B + tree proposed by the present invention can be used to structurally store the information. The base is much easier to build.

Claims (5)

In the structure improvement method of B + Tree, which is a logical structure for data storage, Retrieving additional structures relating to the management portion of the history information of each record and corresponding history information; Modifying current information; Retrieving, inserting, or deleting general information about one record; And storing and retrieving existing information in addition to the new information that is changed when changing information about one existing record. The method of claim 1, wherein the modifying the existing record information, Retrieving the current record with the primary key using a record retrieval procedure of an existing B + tree; And retrieving the record to be modified by said record retrieval. 3. The method of claim 2, wherein searching for the record to be modified by the record search comprises: inserting a current record into a history data structure using a history key; And overwriting new information in a current record after inserting the record. 2. The method of claim 1, wherein retrieving the existing record history information comprises: retrieving a history database through a pointer indicating the record information of the record; Retrieving the necessary information using a history key according to the type of the corresponding history database after searching the database. The method of claim 1, wherein the deleting of the existing record comprises deleting the record after deleting the history information indicated by the record, not just the searched record. Way.