JPS62231346A - Data storage system - Google Patents

Abstract

PURPOSE:To avoid the input to the main memory device of a data page without an empty area and to speed up a data storing processing by inputting an inventory page to the main memory device prior to the storing processing of data. CONSTITUTION:At the time of a data storing request, the inventory page 9 of a data base 3A is inputted to the main memory device 1, and until the data page with the empty area out of respective data pages 4-8 is found, one next entry on the page 9 is successively taken-out. Next, a data page 6 with the empty area is checked, the corresponding data page 6 is read into the main memory devic1 1 and the data are stored into the data page 6. Thus, the input to the main memory device of the data page without the empty area can be avoided. Consequently, the data storing processing can be made highly speedy.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a data storage method for avoiding wasteful input/output operations associated with data storage in a database of an electronic computer.

[Conventional technology]

Figure 3 is a block diagram showing a data storage system in a conventional electronic computer. CoK, disk with database (3), (Hiro)
~(Ir) are multiple data pages provided in this database (3), among which (lI), (r)
and (ri) are data pages without free space, and (6) and (tr) are data pages with air space.

The conventional data storage method is configured as described above, and its operation will be explained below.

Database r on disk (a), 7) When storing K data, first input the first data page (4) into main storage (t), check the free space, and if there is no free space, move to the next The data page (ri) is stored in the main memory (O).
In this way, the page is inputted one after another until a page with free space is reached.

Therefore, in the state shown in Figure 3, the third data page (6
) is input to the main storage device (1), - Data was stored in the free space of this data page (6).

[Problem that the invention seeks to solve]

Conventional data storage methods are structured as described above, so when storing data in a database on disk,
Regardless of whether there is free space or not, all data pages must be input to the main memory in sequence, and access operations for inputting data pages that do not have free space are completely wasteful, and this speeds up processing operations. There was a problem in that it caused an unfavorable situation.

This invention was made to solve the above problems, and it prevents unnecessary input of data pages with no free space into the main memory, and also allows data to be stored at high speed. The purpose is to obtain a data storage method that allows

[Means for solving problems]

The data storage method according to the present invention manages a database on a disk by dividing it into a plurality of data pages, and stores data in each data page, which is created corresponding to each data page. An inventory page consisting of several entries is provided, and the inventory page is entered into the main memory during data processing to avoid processing operations on data pages with no free space when storing data. It is something.

[Effect]

In the data storage method according to the present invention, prior to data storage processing, the presence or absence of free space for each data page is determined in advance by inputting pages into the main memory, and the presence or absence of free space for each data page is determined in advance. Data page input is avoided and data pages with free space are input directly to the main memory, thereby speeding up data storage processing.

〔Example〕

FIG. 1 is a block diagram showing the data storage system according to the present invention. In the figure, (/A) is the main storage device, (-A) is the main storage device (/A), and data is received and received. and a disk having a database (3A);
(ll) to (g) are multiple data pages provided in this database (3A), among which (s, (g)
(7) are data pages with no free space, (6) and (pr) are data pages with free space, and (TE) are the presence or absence of free space in each data page (lI) to (1) mentioned above. In order to manage each data page (lI) ~ (1
) is a special page consisting of several entries (not shown) created in correspondence with the inventory page.

The above-mentioned inventory page (9), for example, is generally configured as shown in Figure 1, with the size of data page n being 20'1 GB and occupying main memory (
/A) Predetermined ratio (10
) and 0%, as shown in Figure 3, the remainder of each data page n is 70 hirobytes (: 10g x 0%).
%) or less, the remaining size will be included, and until then, [OJ will be included.

In FIG. 3, the remaining size of data page n is iooθ bytes, and the remaining size of page n+/ is more than 102'l bytes (in this case, it remains rOJ),
The remaining size of data page n-1-J is /byte, indicating that the remaining size of data page n + j kt is more than /θλ gbytes, and as described above,
All data pages (ri) to (g) are managed by inventory page (9)K.

The data storage system according to the present invention is configured as described above, and its operation will be explained below.

Regarding the data storage flow, as shown in the flowchart in Figure 1, when a data storage request (step l) is made, the database (3A) inventory)
Page (q) is input into the main memory (1) (step 2), and the next data page on the inventory page is sequentially entered until a data page with free space is found among each data page (su) to (g). (Step 3)
. Next, the data page (6) with free space is checked (stepped), and if πS, the corresponding data page (6) is read (stepped) into the main memory (1), and this data Data is stored in page (6) (step 6).

Therefore, in the above-mentioned steps 3 and d, if there is no free space in the data page, the next entry is taken out one after another, so the button
Even data pages with no free space are not input into the main memory (1) once, and the processing speed of data storage can be greatly improved.

〔Effect of the invention〕

As described above, according to the present invention, the presence or absence of free space for each data page is managed using the inventory page, and this inventory page is input into the main storage device first. A data storage method that avoids inputting empty data pages to the main memory and only inputs data pages with free space, which reduces the number of times data pages are input to the main memory and enables high-speed processing. There is an effect that can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a data storage system according to an embodiment of the present invention, and FIGS.
・An explanatory diagram showing a page, FIG. 7 is a flowchart showing data storage processing, and FIG. 3 is a block diagram showing a conventional data storage method. (lA) is main storage, (-A) is disk, (3A)
is a database, (4t) to (g) are data pages, (
9) is the inventory page. In addition, in the figures, the same reference numerals indicate the same or equivalent parts. Procedural amendment (voluntary) Showa 4t16' layer, -ri91

Claims (1)

[Claims] In a data storage method that manages a database on disk by dividing it into multiple data pages and stores data in each data page, an inventory page that consists of several entries created corresponding to each data page. A data storage method, characterized in that the inventory page is input to a main storage device during data processing, thereby avoiding processing operations for data pages with no free space when storing data.