JPH02259942A - Data base generation management processing system - Google Patents

Abstract

PURPOSE:To efficiently execute the retrieval of information at every generation by storing a bit with a prescribed value representing the validity of data held in a record at every generation in the bit masking field of each record when the addition, the updating, and the erasure of the data are preformed. CONSTITUTION:The bit masking field 2 having prescribed bit length and in which each bit corresponds to the generation number of a prescribed generation, respectively is provided on the record 1. A data base managing means 4 that is a managing means preserves the record 1 corresponding to the updating and erasure when the addition, the updating, and the erasure of the data are performed, and stores the bit with the prescribed value representing the validity of the bit masking field at every generation in the bit mask field 2 of each record 1. In such a manner, since it is possible to select the record of the generation only by the value of the bit corresponding to the requested generation of the bit masking field, a processing to make access to the requested generation selectively can be performed with high efficiency.

Description

[Detailed Description of the Invention] [Summary] Regarding generation management of a database that stores change history in an information processing system, it is possible to efficiently search for information for each generation, obtain difference information between generations, etc. A database that stores historical data that has one or more records for storing data and that is managed by generations that are updated every time the data is added, updated, or deleted. In the system, for the record,
A bit mask field is provided that has a predetermined bit length and each bit corresponds to a predetermined generation number, and the management means is configured to add, update, or delete data corresponding to the update or deletion. Save the record to be added, add a record with the contents after the addition and update, and store each bit of a predetermined value indicating the validity of the data held in the record for each generation in the bit mask field of each record. Configure it to do so.

[Industrial application field]

The present invention relates to a generation management processing method for a database that stores change history in an information processing system.

In databases for program development, etc., it is desirable to be able to refer to the history of changes at any time, and in such cases, generation management of update history is necessary.

[Prior Art and Problems to be Solved by the Invention] A database manages a collection of required records as a palette, for example, with a record holding data of a predetermined length as the minimum unit, and generally consists of a plurality of palettes.

In the first method, which is a method of database generation management, when a database is updated, every time a series of update processing is completed and the database is closed, each previous generation is database, generate the entire updated database as a new generation, and specify the database of each generation in the update history table as shown in Figure 6 (a) to reach the required number of generations. Save and manage your database.

In addition, in the second method, for example, the sixth
As shown in Figure (b), the first contents of the changed palette are used as shared information, and, for example, the records of the updated portion are added as updated information each time an update is made.

When accessing this database, data of the required generation is obtained by sequentially stacking the updated information on the common information.

In the first method, the processing is relatively simple and easy to implement, and it is easy to access a certain generation, but the entire database is stored in multiple layers, so the storage space is inefficient. Although the second method has relatively good space efficiency, it has the problem of poor search performance because a specific generation cannot be obtained without tracing the update history.

An object of the present invention is to provide a database generation management processing method that can efficiently search for information for each generation and obtain difference information between generations, and also has good data storage efficiency.

[Means to solve the problem]

FIG. 1 is a block diagram showing the configuration of the present invention.

The figure shows the configuration of a database generation management processing method, in which a database 3 has one or more records 1 for storing data, and stores historical data managed by generations that are updated every time a predetermined addition, update, or deletion of the data is made. In a database system having a record L, a bit mask field 2 having a predetermined bit length and each bit corresponding to a predetermined generation number is provided in a record l, and a database management unit 4, which is a management means, stores the data. The addition of
In the case of updates and deletions, record 1 corresponding to the update and deletion is saved, and record 1 of the contents after the addition and update is saved.
and bitmask field 2 of each record 1
, each bit of a predetermined value indicating the validity of data held in the record 1 for each generation is stored.

[For production]

With the above processing method, the generation management processing required when adding, updating, and deleting records is limited to updating the bit mask field of the record.
Since only necessary portions of historical data are saved in record units, processing efficiency and space efficiency are good. Further, since records of a desired generation can be selected only by the value of the bit corresponding to the desired generation in the bit mask field, processing for selectively accessing the desired generation can be executed efficiently.

〔Example〕

As mentioned above, it is assumed that the database is managed by dividing records into palettes. -For example, set the maximum number of generations to 4
Then, each record consists of a 4-bit bitmask field, a data field of appropriate length, etc.
If there are three data palettes with identification names A, B, and C, which are made up of records that store data, and each data palette is made up of two codes, a database with the structure conceptually shown in Figure 2(a) and Become. The generation management palette and change management palette shown in the figure will be described later.

Each bitmask field is now numbered 1 starting from the left end.
- The bit corresponds to the 4th generation, and when the bit value is 1, it indicates that the data is valid in that generation, and when it is 0, it is invalid. The bit mask field of the newly added record is as follows: The bits of the generation set by the addition (designated generation) and subsequent generations (unused generations) are all set to 1, and the bits before that are set to O. Therefore, in the situation shown in Figure 2(a), all six records are in the first
This indicates that it was created (added) in the same generation.

In the case of record deletion, the record to be deleted is saved as is, all bits of the designated generation and subsequent unused generations in the bit mask field are set to O, and the other bits remain as they were.

In addition, in the case of updating a record, this record will remain as it was before the update, and a new record will be created with the data after the update.
The bitmask field of the previous record is set in the same way as for a deleted record, and the bitmask field of a new record is set in the same way as for an added record.

Figure 2 b) shows data A3 on pallet A in this way.
record is added, the record of data B2 of pallet B is updated to data B2', and the generation progresses to the second generation.
This shows the result of deleting the 2nd record and proceeding to the 3rd generation.

Note that in order to speed up processing, etc., a generation management palette and a change management palette, which have the same configuration as the data palette, are provided. The generation management palette sets generation 1 by setting the bitmask field of each record to 1 for the bit of the required generation.
The column indicates 0, and the data field is the base column.
The generation that is the basis for transition to the generation shown as generation 10 is shown as generation 10 using a similar display method. The data field may further store required additional information.

The generation management palette is used when a generation migration relationship is required when accessing a database. Normally, the generation is migrated in order from the first generation to the second generation to the third generation, but for example, the last generation of a bitmask field Beep) <7) When the generation progresses, the previous generation, for example, the 2nd generation, may be discarded and assigned to an unused generation. In such a case, the 1st generation is set as the base of the 3rd generation. Ru.

In addition, for the record of the change management palette, the bitmask field is a field that displays generation 10 similar to the above, and the data field is the identification name of the palette that contains the record that caused the transition to the generation indicated by 0. Do as shown. The change management palette is used to quickly detect records that have changed in a certain generation.

FIG. 3 is a process flowchart showing the change process by the database management unit 4 of FIG. 1. Note that the database management unit 4 stores the specified generation 10 in an appropriate storage area, uses the value to control the bit mask field, and when this process is performed, the specified generation 10 is stored in a suitable storage area. It is assumed that designated generation 10 is advanced to the next generation.

When the database management unit 4 is requested to add, update, or delete a record in the database 3, it identifies the process in a process step 10, and in the case of an addition, it executes a process to generate a new record in a process step 11. After that, in processing step 15, the bit mask field of the new record is set as described above based on the designated generation ID.

In the case of updating, in processing step 13, for example, a specified change is made to the data of a new record that is a copy of the contents of the original record, and a new record with the updated contents is generated, and then in processing step 14, the data of the old record is changed. Set the bitmask field as in the deletion record above and process step 15.
to set the bitmask field of the new record.

Further, in the case of deletion, in processing step 12, the bit mask field of the record designated for deletion is set as described above. Thereafter, the generation management palette is updated in processing step 16, the change management palette is updated in processing step 17, and the processing ends.

Figure 4 is a flowchart of the process of accessing records of a certain generation in a database configured as described above. For example, when an access request is made by specifying a generation and a palette,
The following processing is executed by sequentially reading the records of the designated palette until a predetermined record end symbol is detected in processing step 20 of the figure. In processing step 21, for the extracted record, it is determined whether the bit corresponding to the specified generation in the bit mask field is 1 or not. If it is 0, the process immediately returns to processing step 20 to process the next record, but if the bit is 1, If so, the record is stored as a valid record in process step 22, and then the process returns to process step 20.

In this way, when all records of the specified palette are processed, processing branches from processing step 20 to processing step 23,
The record stored as a valid record is passed to the request source and the process ends.

For example, if access is requested to the second generation of palette B (generation l0 = 0100) in the database in the state shown in Figure 2 (C), this generation 10 and the bit mask field of each record of palette B The record in which a bit with a bit value of 1 remains in the logical product is the required record, so B1 and B2° are passed to the request source.

FIG. 5 is an example of processing to search for difference information between a certain generation and the previous generation, and the following processing is changed until the identification and record end symbol is detected in processing step 30 of FIG. 5(a). Read and execute records in the management palette sequentially. First, for the code taken out from the change management palette in processing step 31, it is determined whether the bit corresponding to the specified generation in the bit mask field is 1, and if it is 0, the process immediately returns to processing step 30 to process the next record. , if the bit is 1, the palette identification name held in the record is stored in processing step 32, and then processing returns to processing step 30.

When all records in the change management palette are processed in this way, processing branches from processing step 30 to processing step 33.

In processing step 33, the generation management palette is accessed, and the base generation IO is obtained from the specified generation unit.
The following process is executed by sequentially reading records for one of the data palettes in which the palette a alias is stored until the record end symbol is identified and detected in process step 34.

That is, when comparing the bit mask field of the record read from the data palette in processing step 35 with the specified generation 10 and the base generation 10, the states shown in the table can be identified by the corresponding 2-bit combinations, so 00 In the case of 01 and 11, the process immediately returns to processing step 34 to process the next record, but in the case of 01 and 10, the data of that record and the addition/update or deletion status information are stored as difference information in processing step 36. After that, the process returns to step 34.

In this way, when all records of one palette in which identification names are stored are processed, processing branches from processing step 34 to processing step 37, and if there is an unprocessed palette in which identification names are stored, processing step 34 is performed. Return to process the next pallet. If all palettes have been processed, the process proceeds to step 38, where the difference information stored in the above process is passed to the request source, and the process ends.

For example, if a change management palette is searched for in the database in the state shown in FIG. 2(C) by specifying the third generation (generation ID=OO10), palette C is detected in processing step 31. Specify the record of pare ytoC generation l0 = 0010
By searching with base generation ID=0100,
In processing step 35, since the record of data C2 corresponds to the state of IO, that is, deletion, C2 is passed to the request source as difference information.

Table (Effects of the Invention) As is clear from the above description, according to the present invention, in generation management of a database that stores update history, database change processing, information search processing for each generation, and difference information between generations are performed. It has remarkable industrial effects in that acquisition processing and the like can be executed efficiently, and data storage is space-efficient.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of the present invention, Fig. 2 is an explanatory diagram of an example of a database of the present invention, Fig. 3 is a flow chart of the processing of the present invention, Fig. 4 is a flow chart of generation search processing, Fig. 5 is a flowchart of differential information search processing, and FIG. 6 is an explanatory diagram of an example of a conventional generation management method. In the figure, l is a record, 2 is a bit mask field, 3 is a database, 4 is a database management unit, Fig. 3 is a flowchart of the processing of the present invention. Fig. 3 is a block diagram showing the configuration of the present invention. Fig. 1 is an explanation of an example of a database of the present invention. Figure 2 Figure 2 Flowchart of generation search processing Figure 4 Illustration of an example of conventional generation management method Figure 6

Claims (1)

[Scope of Claim] A database system that stores historical data that has one or more records (1) that store data, and that stores historical data that is managed by generations that are updated every time a predetermined addition, update, or deletion of the data, comprising: (1) is provided with a bit mask field (2) which has a predetermined bit length and each bit corresponds to a generation number of a predetermined generation, and the management means (4) controls the addition, updating, and processing of the data. In the case of deletion, save the record (1) corresponding to the update and deletion, add a record with the content after the addition and update, and store it in the bitmask field (2) of each record. 1. A database generation management processing method, characterized in that the database generation management processing method is configured to store each bit of a predetermined value indicating the validity of data for each generation.