JPS63206837A - System for accessing partitioned data set - Google Patents

Abstract

PURPOSE:To make access to the member of a partitioned data set by using plural retrieval keys expressed in recognizable marks by a user, by constituting an index member which represents correspondence relation between the retrieval key and a member name as one of the members of a data area in setting a partitioned data. CONSTITUTION:An accessing means 30 loads the index member 16 on an index table storing means 40 prior to all of the processings, and generates and stores a table which represents correspondence between the retrieval key and the member name on the means 40. The accessing means 30, when an access request by the retrieval key for the information of the member 11 of the partitioned data 10 being issued from a means 20, specifies a corresponding member name from the table on the means 40, and sends it to the set 10. The set 10 retrieves the member name area 14 of an address directory 13, and accesses to the information of the member 11 of the data area 12 based on the address information of an address area 15 making a pair with the said member name area. By adopting the index member 16, it is possible to access to the information by using the mark in which the index key and it value possible to be described in the index member are expressed as the recognizable marks by the user.

Description

[Detailed Description of the Invention] [Summary] The present invention provides a method for accessing a partitioned datacent based on a search key when a one-partition datacent is used as a data cache, in which a search is assigned to the member name of a member organizing a three-partition data set. In order to make the number of keys practically unlimited, and to solve the request for unlimited character string length for search keys, we have created a search key and member name as one of the members of the 9-part dataset. By constructing an index member that represents the correspondence between.

It is possible to access the members of a data center by using a plurality of search keys expressed in symbols that are easy for the user to understand.

[Industrial application field]

The present invention particularly relates to a partitioned data set access method that is divided and organized by member.

When using a partitioned data center as a database, members can be accessed using multiple search keys, and the search keys can be encoded in a way that the user can understand. Regarding the method.

[Conventional technology]

In a partitioned data set, a plurality of logical files are stored and accessed in the form of members in one physical auxiliary storage device, and is the most common data set format. The structure of this segmented data set is shown in FIG. In the figure, 10 is a segmented data set, and 11 is a member that is a set of related data. This segmented data set 10 consists of a data area 12 consisting of a plurality of members 11 and a directory area 13 for managing these members 11. In this directory area 13, a unique name given to each member 11 is provided. Member name area 14 that stores member names
and an address area 15 for storing address information in the data area 12 of the member 11 are configured as a pair. Partitioned dataset 1 configured like this
0, when reading data of a specific member 11, the member name stored in the member name area 14 of the directory area 13 is referred to, and the address stored in the address area 15 paired with the member name area 14 is read. Actions will be taken based on the information.

When using this segmented data set in a database, it is necessary to provide a search key for database operation, but as one of the conventional techniques for realizing this search key, There is a method of providing a search key by using the basic structure of the data set as it is and using member names as search keys, that is, a method of having one member name function as one search key. In addition, as another conventional technique, as shown in FIG. 5, member name area 1 used to represent member names
In some cases, it is possible to provide a plurality of search keys by dividing the four character areas as appropriate and assigning a search key to each of the divided areas.

[Problem that the invention seeks to solve]

In this way, when it becomes necessary to provide a search key for a one-part dataset, in the conventional technology that uses member names as search keys, the number of search keys corresponding to member 11 is limited to one. However, it has had the disadvantage of not being able to adequately respond to complex and highly functional search requests.

In addition, in the conventional technology in which a character area used to represent a member name is appropriately divided and a search key is assigned to each divided area to provide a search key, it is not possible to use multiple search keys. Although such a drawback does not occur, the number of digits in one character area is limited to, for example, eight digits.

There is a drawback that a sufficiently large number of search keys cannot be used, such as the inability to provide eight or more search keys. Moreover, with this conventional technology, if you try to increase the number of search keys even a little, you will not be able to increase the character string length of one search key itself, so the search key will be easier for the user to understand. There was also a situation where it became impossible to express the information using such symbols.

[Means for solving problems]

FIG. 1 is a diagram illustrating the principle structure of a partitioned data set access method according to the present invention.

In the figure, 10 is a partitioned data set, and 11 is a member.

12 is a data area, 13 is a directory area, 14 is a member name area, 15 is an address area, 16 is an index member, 20 is a data request means (for example, an application
program), 30 is an access means, and 40 is an index table storage means.

As shown in FIG. 1, in the partitioned data set access method of the present invention, a partitioned data set 10 is composed of a data area 12 and a directory area 13, and this data area 12 stores the data of each member 11 on a member-by-member basis. At the same time, information representing the correspondence between a search key and a member name is regarded as one member and stored as an index member 16. The directory area 13 also has a function of managing the member names of the members 11 and address information in the data area 12, and in order to realize this function, the member name area 1 is used to store the member names of the members 11.
4 and an address area 15 for storing address information in the data area 12 of the member 11 as a pair.

The data request means 20 is a means for requesting data of the member 11 of the partitioned data center 10 using a search key, and the access means 30 accesses the data of the member 11 of the partitioned data set 10 corresponding to this search key. It is a means to do so. Further, the index table storage means 40 is.

Index member 16 according to instructions from access means 30
This is a means for storing the index table created by loading.

Note that the classification data set 10 in FIG. Member 11, data area 12. Directory area 13゜Member Meijo 14
．． The address area 15 includes the segmented data set 10, member 11 . Data area 12. Direct area 13. Member Meijo 14. It is functionally identical to address area 15, and therefore the same reference numerals are used.

[Effect]

The access means 30 of the present invention loads the index member 16 into the index table storage means 40 prior to all access processing.

An index table representing the correspondence between search keys and member names is created and stored on the index table storage means 40. Thereafter, when the data requesting means 20 requests access to the data of the member 11 of the partitioned data set 10 using a search key, the access means 30 searches the index table on the index table storage means 40 and uses the search key. Identify the member name corresponding to .

Send this identified member name to the partitioned dataset.

The partitioned data set 10 that has received this specified member name has the specified member name in the directory area 1.
3. When the corresponding member name area 14 is found by the search, the address area 15 paired with the member name area 14 is searched.
Data in member 11 of data area 12 is accessed based on address information stored in .

By using the index member 16 in this way, there are no restrictions on the area, and the number of search keys that can be written in the member 16 can be reduced compared to the conventional case.
You can take as much as you want.

〔Example〕

An embodiment of the partitioned data center access method according to the present invention is shown in FIG. In the figure, the same symbols are used for the same components as in FIG. 4, which shows the prior art of a nine-section data set.

In FIG. 2, 1 is a data processing device, 2 is an auxiliary storage device, and 3 is a main storage device.The data processing device 1 loads the necessary data of the auxiliary storage device 2 into the main storage device 3, and 5 data. Execute processing.

And codes 10, 11.12. 13, 1, 4.15 degrees, 16.20 degrees, and 30.40 degrees correspond to those in FIG. 1, respectively, so their explanation will be omitted.

Next, the details of the structure of the index member 16, which is a feature of the present invention, will be explained.

In order to explain this structure, CAD data center of 1 section is used.
Assuming an example in which the system is applied to an IC design program library, each IC design program that is the content of the data in member 11 is:

For example, it is desirable to request access using multiple search keys such as IC code, function, technology, or designer name. Therefore, in order to respond to such access requests, the index member 16 of the present invention is constructed as shown in the embodiment of FIG.

The correspondence between these multiple search keys and the member names of the members 11 to which access is requested is defined as a table, and one of the members 11 in the data area 12 is defined as a table.
It is stored as one. This will be explained in the embodiment of FIG.

This means that the value of the IC code that is the search key is LSOO, the value of the function that is the search key is NAND, and the member name of member 11 of the technology that is the search key (direct is LSTTL) is Aooo. Therefore, if the index table created by this index member 16 is accessed with the LSQO of the IC code value, the member name AOOO will be specified, and if it is accessed with the NAND of the function value. In this case, member names A000 and A002 are specified.

Although this embodiment shows the index member 16 based on three types of search keys, 1. Since the present invention configures the index member 16 as one of the members 11 of the partitioned data set 10, the directory area 1
Unlike the conventional technology in which search keys are configured in the member name area 14 of 3, there is no area restriction, and therefore it is possible to define as many correspondences as possible between search keys and member names, and it is possible to It becomes possible to set the search key and its value in an expression format that is easy for the user to understand, such as NAND or INV of the value of .

Next, the operation of the present invention based on this configuration will be explained. Access program 30 first.

By loading the contents of the index member 16 previously stored in the data area 12 into the index table memory area 40 prior to all access processing, the correspondence between search keys and member names is stored on the main storage device 3. An index tape that represents.

process to create and store files. That is, through this process, the contents of the index member 16 as shown in FIG. 3 are stored as an index table in the index table memory area 40. The reason why the index table is created on the main storage device 3 in this way is mainly because the data processing device 1 can access this index table at high speed.

After this process, when there is an access request from the application program 20 to the data of the member 11 of the partitioned data set 10 using a search key, the access program 30 searches the index table in the index table memory area 40 and searches the index table using the search key. Identify the member name corresponding to . To specifically explain this process using the embodiment of the index member 16 shown in FIG. Suppose that there is
The access program 30 is.

Search the index table stored in the index table memory area 40, which is a table equivalent to that shown in FIG. 3, and specify the member name A002 indicating the member 11 in which the requested IC design program is stored. I will do it.

In this way, when the access program 30 specifies the member name of the member 11 in which the data requested by the application program 20 is stored, the access program 30 performs a process of sending the specified member name to the segmented data file 1-10. The access processing performed by the partitioned data center 10 from now on is basically the same as the access processing performed by the partitioned data center of the prior art. i.e. 1
The partitioned data set 10 searches for which member name area 14 in the system directory 13 the specified member name is stored, and when the corresponding member name area 14 is found through the search, the member name area 14 and the corresponding member name area 14 are stored. Based on the address information stored in the paired address areas 15, the data of the member 11 stored at the corresponding address in the data area 12 is accessed. The accessed data is then read out to the main storage device 3 and sent to the data requesting means 20.

In this way, the access program 30 can send to the application program 20 the data of the member 11 requested by the application program 20 specified by the search key.

〔Effect of the invention〕

According to the present invention, it is possible to access the contents of the members of one category of data using a virtually unlimited number of search keys. Furthermore, since there is no restriction on the length of the character string of the search key, it is possible to use symbols that are easily understood by the user as the search key and its value. For these reasons, when applying partitioned datasets to databases that require the use of multiple search keys, it is possible to flexibly and easily construct complex and sophisticated databases, including the management of relationships between members. It will become a reality.

Furthermore, in the present invention, members other than index members are
Since the members function in the same way as the members of conventional partitioned datasets, various conventional application software that is standardly equipped for single partitioned datasets can be used as is, making it possible to develop application software when constructing databases, etc. It will also be possible to save labor.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the basic configuration of the present invention in detail. FIG. 2 is a diagram illustrating an embodiment of the present invention, FIG. 3 is a diagram illustrating an embodiment of the structure of an index member of the present invention, and FIG. 4 is a diagram illustrating the configuration of a segmented data set. FIG. 5 is a diagram showing a conventional technique of search keys. In FIG. 1, 10 is a partitioned dataset, 11 is a member, 12 is a data area, 13 is a directory area, and 14
is the member name area, and 15 is the address area. 16 is an index member.

Claims (1)

[Claims] (1) A division having a data area (12) for storing data in units of members (11) and a directory area (13) for managing member names of these members (11) and address information in the data area (12). a data set (10), a data requesting means (20) for requesting data of a member (11) from this segmented data set (10) using a search key, and the segmented data according to the search key from this data requesting means (20). Access means (3) for accessing data of member (11) of set (10)
0), an index member (16) representing a correspondence relationship between one or more search keys and member names is configured as one of the members (11) of the data area (12). At the same time, this index member (
16), and the access means (30) stores the index table in the index table storage means (40). The search key from the data requesting means (20) is converted into a corresponding member name by referring to the index table provided in
10) is obtained by accessing the data of the member (11) stored in the data area (12) based on the address information of the directory area (13) according to the member name converted by the access means (30). A partitioned data set access method characterized by: