JPH07281931A - Intermediate data storing method - Google Patents

Abstract

PURPOSE:To increase the total processing speed of inquiries by reducing the redundant disk input/output frequency and also reusing the key value needed for retrieval of the composite conditions. CONSTITUTION:A data control part 110 controls the storage and acquisition of the intermediate data for the intermediate data storage pages of a work pace area 102 of a main storage 10 and a temporary file 103 of a secondary storage 20. A data structure control part 120 controls the data structure including the type and size of the key value included in the intermediate data and offers the data structure to the part 110. A cursor part 130 holds the current information on an offset address set to the corresponding element entry as well as the intermediate data storing pages. A page control part 140 controls to show the specific parts where the intermediate data storing pages are included in the area 102 and the storage device 20 respectively and indicates these parts to the part 110.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of storing intermediate data of a query process for a database in an object-oriented database management system, and more particularly to efficiently manage an intermediate data set of intermediate results generated during the query process. In addition, the present invention relates to a method for storing intermediate data which effectively utilizes a main storage device and a secondary storage device to perform various inquiry processing including compound condition retrieval at high speed.

[0002]

2. Description of the Related Art A conventional intermediate data storage method uses an object identifier, which is a logical identifier capable of uniquely identifying a set of intermediate data obtained as an intermediate result by an individual database access function. In the case where the intermediate data is stored in a temporary file in the secondary storage device and used in subsequent database access functions and query processing, the logical identifiers stored in the temporary file are stored in order. The method of accessing, reading the data corresponding to the logical identifier again from the database, and proceeding with the inquiry processing is used.

FIG. 6 is a block diagram showing an example of a conventional intermediate data storage method. As shown in FIG. 6, the intermediate data generated by the database access function is stored in the temporary storage device 6 in the secondary storage device 620 in advance.
In 01, the object identifiers corresponding to these intermediate data are stored in the order in which they were obtained.

Then, when the inquiry processing thereafter progresses and it becomes necessary to access those intermediate data,
The object identifier 603 is sequentially fetched from the temporary file 601 onto the main storage device 610, and the object 604 corresponding to the object identifier 603 is read again from the database 602 and used.

As described above, the conventional intermediate data storage method can manage a set of intermediate data temporarily generated during query processing as an object identifier and reuse them.

[0006]

The above-mentioned conventional intermediate data storage method is for acquiring the objects corresponding to individual object identifiers when complex query processing such as complex condition search or generating a final result. However, it is necessary to access the database many times, and as a result, redundant disk input / output occurs and the processing speed of the entire query processing decreases.

The object of the present invention is to reduce the redundant disk input / output generated by such a conventional intermediate data storage method and to reuse the key value required for the compound condition search, thereby making the entire query processing. An object of the present invention is to provide an intermediate data storage method capable of speeding up.

[0008]

Therefore, the intermediate data storage method of the first invention is (A) individual search in the intermediate data storage method of the query processing including the compound condition search of the database of the object-oriented database management system. From the intermediate result generated during the execution of the database access function for processing, the object identifier of a logical identifier that can be uniquely identified in the database and the contents of some key values that are the attribute values of the object are used as intermediate data. And (B) forming an intermediate data storage page on the main storage device by a set of these intermediate data,
Each of the intermediate data storage pages is stored in a temporary file provided in a secondary storage device, and (C) in the subsequent database access function or inquiry processing, for each of the intermediate data storage pages stored in the temporary file. This includes retrieving the corresponding object identifier and key value, retrieving the intermediate data of the retrieval result, and utilizing it.

The intermediate data storage method of the second invention is the intermediate data storage method of the query processing including the compound condition search of the database of the object-oriented database management system, wherein (A) the data control unit operates the main storage device. The storage and acquisition of the intermediate data for the intermediate data storage page in the page area and the temporary file in the secondary storage device are controlled, and (B) the data structure management unit stores the type of the key value stored in the intermediate data and The data structure including the size is managed and provided to the data control unit, and (C) the cursor unit holds the current information of the offset address for the intermediate data storage page and the element entry in the intermediate data storage page. The data control unit is instructed where the element entry at the current position is, and the data control unit According to the instruction of the section, the cursor function is performed by moving to the element entry before and after, and (D) the page management section manages where each of the intermediate data storage pages exists in the working page area and the temporary file. This includes instructing the data control unit.

Furthermore, the intermediate data storage method of the third invention is the intermediate data storage method of the first invention and the second invention, wherein the intermediate data storage page of the first invention and the second invention is (A ) An entry area, which is a set of element entries in which object identifiers and fixed-length key values for individual intermediate data and offset addresses to variable-length key values are arranged, and (B) variable-length keys for individual intermediate data Each of the values is separately stored in a data area in which each value is arranged as data, and the storage order and attribute information of these intermediate data are managed.

[0011]

According to the intermediate data storage method of the present invention, an object can be specified by an object identifier forming the intermediate data for complicated query processing such as complex condition retrieval, and a key value which is an attribute value of the object. Since it is also stored together, the intermediate data can be reused without accessing the database again by using the cursor function of the device.

The intermediate data storage method of the present invention is
As a method of managing intermediate data, by dividing the storage page format into an entry area and a data area, it is possible to handle variable-length data having various lengths as key values.

[0013]

Embodiments of the present invention will now be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the intermediate data storage method of the present invention. As shown in FIG. 1, the intermediate data storage unit 101 in the main storage device 10 includes a data control unit 110 that controls the storage and acquisition of data, and a data structure that includes the type and size of the stored key value. A data structure management unit 120 for managing data, a cursor unit 130 having current information of offset addresses for pages and element entries, and a page management unit 14 for managing where each page in which data is stored exists.
It is composed of 0 and 0.

Further, the intermediate data storage means 101 is provided with a work page area 1 in the main memory 10 for working the intermediate data.
02, the temporary file 103 secured in the secondary storage device 20 is used so that a large amount of intermediate data can be dealt with. These work page area 102 and temporary file 103 are instructed by the page management unit 140 as intermediate data storage pages and are controlled by the data control unit 110.

FIG. 2 is an information structure diagram showing an example of a storage form regarding the intermediate data storage page 200. Also, FIG.
FIG. 3 is an information structure diagram showing an example of the contents of the element entry 300 in the entry area 201. As shown in FIG. 2, the intermediate data storage page 200 has a page footer 204 for holding a page number, the number of elements in the page, etc., and an entry area 201 for storing an object identifier and a key value and data. It is divided into a data area 203 for storing.

The entry area 201 is secured in the front area of the intermediate data storage page 200 in order from the head, and the data area 203 is secured in the rear area of the page footer 204 from the back. The remaining area is a free area 202. A free space 2 for storing the entire element in the intermediate data storage page 200
When 02 is insufficient, a new intermediate data storage page is created.

On the other hand, as shown in FIG.
The element entry 300 stored in 01 includes an object identifier 304, and includes a basic type key value 301, a fixed length key value 302, an offset address 303 to a variable length key value stored in the data area 203, and the like. It is configured to have.

FIG. 4 is a flow chart showing an example of a procedure for storing intermediate data. Below, this intermediate data storage page 2
00, the procedure for storing the object identifier and the key value of the intermediate data will be described with reference to FIGS.

First, when a request to store intermediate data is issued to the data control unit 110, the data control unit 110 uses the current information of the cursor unit 130 to move the position of the element entry 300 next to the current offset of the current page. Then, the intermediate data is stored using the storage information of the data storage structure management unit 120.

Then, in step 401, it is judged whether or not there is a remaining key value. If there is a remainder, step 402 determines if the key value is a variable length key value. If it is a variable length key value, step 4
After storing the variable length key value in the data area 203 in 03, in step 404, the offset address 303 to the variable length key value is stored in the element entry 300 of the entry area 201, and the flow returns to step 401 to move to the next key. It has moved to value processing.

If the key value is the fixed length key value in step 402, the entry area 201 is determined in step 405.
The fixed length key value 302 is stored in the element entry 300, and the process returns to step 401 and shifts to the processing of the next key value. On the other hand, when there is no remaining key value in step 401, the object identifier 304 is stored in the element entry 300 of the entry area 201 in step 406, and the process is terminated.

When the intermediate data storage page 200 which is the current page overflows in the above process,
A new intermediate data storage page is created
The current information of 30 has been updated.

FIG. 5 is a flow chart showing an example of the procedure for acquiring the stored intermediate data. Hereinafter, a procedure for acquiring the intermediate data stored in the intermediate data storage page 200 will be described with reference to FIGS. 1, 3 and 5.

First, when the data control unit 110 is requested to acquire the object identifier 304 in the element entry 300, which is the next intermediate data, the data control unit 110 causes the current information of the cursor unit 130 to change the current in step 501. Intermediate data storage page 20 which is a page
The current offset of 0 is advanced to the position of the next element entry 300.

Then, in step 502, the page management unit 140 causes the intermediate data storage page 2 of the current page to be stored.
00 is checked on the main storage device 10. If it does not exist, in step 506, an empty area of the work page area 102 is searched. Further, if there is no free area, in step 508, the work page area 1
Any page in 02 is saved in the temporary file 103.

Then, in step 507, the intermediate data storage page 2 of the current page is read from the temporary file 103.
00 is being read. Therefore, in step 503, the element entry 300 that is the current offset in the data storage page 200 is acquired. And step 5
At 04, it is checked whether the element entry 300 exists. If it exists, in step 505, the object identifier 3 in the element entry 300 at the current position
04 has been acquired and processing has ended.

On the other hand, in step 504, the element entry 3
If 00 does not exist, it is checked in step 509 whether the next intermediate data storage page exists.
If it exists, in step 510, the current page is advanced, the current offset is cleared, and then the process returns to step 502 to move to the operation of acquiring the element entry from the next intermediate data storage page. Incidentally, in step 509, if the next intermediate data storage page does not exist, the processing is ended as it is.

[0028]

As described above, according to the intermediate data storage method of the present invention, the key value stored as the intermediate data can be reused in the complicated query processing including the compound condition search and the like. There is no need to read it again, and as a result, only temporary files that store intermediate data are accessed, and the overhead for disk input / output is small, which has the effect of speeding up the overall query processing. is doing.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of an intermediate data storage method of the present invention.

FIG. 2 is an information configuration diagram showing an example of a storage form regarding an intermediate data storage page 200.

FIG. 3 is an element entry 300 in an entry area 201.
It is an information block diagram which shows an example of the content of.

FIG. 4 is a flowchart showing an example of a procedure for storing intermediate data.

FIG. 5 is a flowchart showing an example of a procedure for acquiring stored intermediate data.

FIG. 6 is a block diagram showing an example of a conventional intermediate data storage method.

[Explanation of symbols]

 10 Main Storage Device 20 Secondary Storage Device 101 Intermediate Data Storage Unit 102 Work Page Area 103 Temporary File 110 Data Control Section 120 Data Structure Management Section 130 Cursor Section 140 Page Management Section 200 Intermediate Data Storage Page 201 Entry Area 202 Free Area 203 Data Area 204 Page footer 300 Element entry 301 Basic type key value 302 Fixed length key value 303 Offset address to variable length key value 304 Object identifier 610 Main storage 620 Secondary storage 601 Temporary file 602 Database 603 Object identifier 604 Object

Claims (3)

[Claims] 1. An intermediate data storage method for a query process including a compound condition search for a database of an object-oriented database management system, comprising: (A) an intermediate result generated during execution of a database access function for performing an individual search process. , The contents of some key values that are attribute values of the object are acquired as intermediate data together with the object identifier of the logical identifier that can be uniquely identified in the database, and (B) on the main storage device by the collection of these intermediate data Form the intermediate data storage page in
Each of the intermediate data storage pages is stored in a temporary file provided in a secondary storage device, and (C) in the subsequent database access function or inquiry processing, for each of the intermediate data storage pages stored in the temporary file. A method for storing intermediate data, characterized in that the corresponding object identifier and key value are searched, and the intermediate data of the search result is retrieved and used. 2. An intermediate data storage method for a query process including a compound condition search for a database of an object-oriented database management system, comprising: (A) a work page area of a main storage device in a data control unit and a temporary file in a secondary storage device. The storage and acquisition of the intermediate data in the intermediate data storage page in (1) are controlled, and (B) the data structure management unit manages the data structure including the type and size of the key value stored in the intermediate data to manage the data. And (C) holding the current information of the offset address for the intermediate data storage page and the element entry in the intermediate data storage page with the cursor portion, and thereby the location of the element entry at the current position is stored. Instruct the data control unit to move to the previous / next element entry by the instruction of the data control unit. And performs a cursor function, and (D) the page management unit instructs the data control unit by managing where each of the intermediate data storage pages exists in the working page area and the temporary file. An intermediate data storage method characterized by the above. 3. The intermediate data storage page according to claim 1 or 2, wherein (A) each object identifier and fixed length key value for each intermediate data and offset address to a variable length key value are arranged. The storage area is divided into an entry area that is a set of element entries and (B) a data area in which each variable-length key value for each intermediate data is arranged as each data, and the storage order of these intermediate data and 3. The intermediate data storage method according to claim 1, wherein the attribute information is managed.