JPH0325636A - Managing method for data record - Google Patents

Abstract

PURPOSE:To generate an index record and to accelerate the access to an arbitrary data record by adding the volume identification record of a volume, where the data record exists, and position information from the leading volume. CONSTITUTION:When a data file 3 composed of plural data records 5 exists over plural volumes 14, 15 and 16, the volume identification information of the volumes 14, 15 and 16, where the data record 5 exists, and the position information from the leading volume are added and the index record is generated. When retrieval is executed, the volumes 14, 15 and 16 are selected based on the volume identification information of the index record. Thus, when the data file 3 spreads over the plural volumes the access to the arbitrary data record can be accelerated.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for managing data records in write-once optical disc devices and the like. In particular, it relates to a method of managing data records that can be searched randomly using keys.

[Prior Art] The configuration of a conventional file system will be explained with reference to Figs. 2 and 3. FIG. 2 is an explanatory diagram showing an index file and a data file used in a conventional data record management method, and FIG. 3 is an explanatory diagram showing an index record used in a conventional data record management method.

In FIG. 2, an index file (1) is created from a large number of index records (2) and created on a magnetic disk.

Furthermore, data files <3> and (4) are composed of a large number of data records (5), and data records (
5) consists of a key part (6) and a data part (7). Note that the data files <3> and <4) are created on different volumes (optical disks).

In FIG. 3, the index record (2) is the key after performing forward and backward compression processing on the key part (6) of the corresponding data record (5) based on the relationship between the keys of the preceding and succeeding records.
8), the number of bytes for forward compression (9), the key length after compression processing (10), and the relative address from the beginning of the data file of the volume (11>). , this index record (2) corresponds to each data record (5).Next, the operation of the above-mentioned conventional example will be explained.

First, for example, when searching for an arbitrary data record (5) in a data file (3), a key value representing the data record (5) is specified as a parameter. As a result, the file system searches the index file (1) on the magnetic disk and selects the index record (2) having the corresponding key. Based on the relative address <11> in this index record (2), the physical address of the target data record (5) from the beginning of the data file (3) is calculated, and the deniter record ( 5) is read from the corresponding volume into the main storage and handed over to the program. By the way, if the data file (3) spans multiple volumes, the index record (2) does not have information indicating in which volume the data record (5) exists, so all Any data record (5) cannot be accessed unless the volumes are mounted, the relative address on each volume is checked from the label, and this is managed in the main memory. [Problem to be solved by the invention] In the conventional data record management method as described above,
When a data file spans multiple volumes, there is a problem in that it takes time to access any data record. This invention was made to solve the above-mentioned problems, and provides a data record management method that can speed up access to any data record when a data file spans multiple volumes. The purpose is to [Means for Solving the Problems] The data record management method according to the present invention consists of the steps described below. (i) When a data file consisting of multiple data records exists across multiple volumes, the index is added by adding the volume identification information of the volume where the data record exists and the position information from the first volume. Step to generate records. (ii). When searching for the data record, selecting the volume based on the volume identification information of the index record, calculating a physical address based on the position information of the index record, and reading the data record. [Operation] In the present invention, in the first step, when a data file consisting of multiple data records exists across multiple volumes, the volume identification information of the volume in which the data record exists and from the first volume are The location information is added to generate an index record. Further, in the next step, when searching the data record, the volume is selected based on the volume identification information of the index record;
A physical address is calculated based on the position information of the index record, and the data record is read. [Embodiment] The configuration of a file system using an embodiment of the present invention will be described with reference to FIG.

FIG. 1 is an explanatory diagram showing the concept of an embodiment of the present invention, and the structure of the data record (5) is exactly the same as that of the conventional example described above. In FIG. 1, the index record (2^) used in one embodiment of the present invention is compressed forward and backward with respect to the key part <6) of the corresponding data record (5) due to the relationship between the keys of the records before and after it. Key <8> after processing and forward compression. The number of bytes (9), the key length after compression processing (10), the relative volume number (12) representing the relative displacement from the first volume (14), and the number of bytes of the first volume (14) from the beginning of the data file. A relative address representing the relative displacement of (
13) and has been sidelined. Note that this index record (2^) corresponds to each data record (5).

An index file (l^), not shown, is composed of a large number of index records (2^) and is created on the magnetic disk 2L. In addition, the data file (3) has three volumes (1
4), (15) and (16), that is, it spans three optical disks (14), (15) and (16). Next, the operation of the above-mentioned embodiment will be explained with reference to FIG. First, the index record (2
When generating 8〉, the relative displacement “3” from the first volume (14) of the volume (16) in which the key part (6) is “01010” (5〉) exists is expressed as the relative volume number (12). The relative displacement of the second first volume (14) from the beginning of the data file is added as a relative address <13) to the index record <2^>. Then, when searching for data record (5) whose key part (6) is “01010”, the program searches for “01010”.
The key ``10'' is set as a parameter. This causes the file system to search the index file (1^) created on the magnetic disk and obtain the index record (2^) having this key.

After that, based on the relative volume number (12) "3" of the obtained index record (2^), select the volume "1B" in which the corresponding data record (5) exists, and select the drive device (not shown). Next, the relative address (
13>, the relative displacement from the beginning of the data file in the first volume (14) is obtained, and the corresponding data record (5
), and calculate the physical address of the corresponding volume (optical disk>(16) to the corresponding data record (5)
Perform reading.

One embodiment of the present invention, as described above, is capable of measuring the volume where the target data record exists and the relative displacement of the target data record without mounting all the volumes and checking the relative displacement of the data record. You can directly search for the desired data record without performing a record search. Therefore, it is possible to equalize the access time for each record of a data file spanning multiple volumemes. Note that the above example shows the case of a write-once optical disk, so the index file is recorded on a rewritable magnetic disk. However, when recording data records on a rewritable recording medium, the index file is Similar behavior can be expected even if the file is recorded on the same recording medium as the data record. [Effects of the Invention] As explained above, when a data file consisting of a plurality of data records exists across a plurality of volumes, the present invention provides volume identification information of the volume in which the data record exists and the volume identification information of the first volume. a step of generating an index record by adding location information from the index record; when searching the data record, selecting the volume based on the volume identification information of the index record; Since this method consists of the steps of calculating a physical address and reading the data record, it is possible to speed up access to any data record when the data file spans multiple volumes.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram showing the concept of one embodiment of the present invention, and FIG.
The figure is an explanatory diagram showing an index file and a data file used in the conventional data record management method, and FIG. 3 is an explanatory diagram showing the index record used in the conventional data record management method. In the figure, (2^> ... index record, (3) ... data file, (5) ... data record, (6) ... key part, (7) ... data section, (12) ... Relative volume number, (13)
... Relative address, (14), (15), (16) ... Optical disk (volume). In each figure, the same reference numerals indicate the same or corresponding parts Otsu 1 (7
)Ln's

Claims (1)

[Claims] When a data file consisting of multiple data records exists across multiple volumes, an index record is generated by adding the volume identification information of the volume where the data record exists and the position information from the first volume. and when searching the data record, selecting the volume based on the volume identification information of the index record, calculating a physical address based on the position information of the index record, and reading the data record. A data record management method characterized by comprising: