JPS61283963A - File retrieving device - Google Patents

Abstract

PURPOSE:To reduce the load of a host CPU and to perform the processing of a file retrieving device with no delay of the retrieving time despite the increase of the file capacity, by providing a sub-CPU at every unit and attaining the file retrieving actions independently of each other and in parallel with each other. CONSTITUTION:The sub-CPU 8a and 8b are provided for each retrieving unit consisting of a pair of data memories and retrieving memories. Then the retrieval key data are delivered simultaneously to both CPU 8a and 8b for the file retrieval of a host CPU 5. In other words, both CPU 8a and 8b collate the retrieval information recorded to the retrieval memories 2a and 2b with the retrieval key data and read the file data information out of the data memories 1a and 1b based on the result of said collation. This read file data information is reported to the CPU5. Thus the CPU5 can continue an optional other processing until it receives said report. This reduces the load of the CPU5. Furthermore the retrieving actions are carried out simultaneously and independently of each other. Thus the file retrieval processing is possible without increasing the retrieving time despite the increase of the file capacity.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a file search device for searching files such as a database.
The present invention relates to a file search device of a type that records file data information and index information on separate recording media.

In a conventional file search device, a data memory 1 consisting of a write-once memory for recording data information and index information corresponding to each data information are used. It has an index memory 2 consisting of a rewritable memory for recording data memory 1 and index memory 2.
are data memory control unit 3 and index memory control unit 4, respectively.
It is connected to the host CPU 5 via. Like this!
The search operation in the conventional file search device is as shown in the flowchart shown in FIG. That is, when search data is keyed into the host CPU 5 via an appropriate input device such as a keyboard (step 7a), the host CPU
5 reads frame data including only the index information from the memory 2 via the control unit 4 (step 7b). Next, in step 7C, it is determined whether the data read from the memory 2 matches the key-in data, and if it matches, the process proceeds to step 7d, where data is read from the data memory 1.
.

Read the corresponding file data. On the other hand, if no match is found in step 7c, proceed to step 7e, repeat the cycle including step 7b and step 7c until the end of the frame, then proceed to step 7f, for example. When displaying no data on a CRT or the like, a large amount of time is required to process the cycle consisting of steps 7b, 7c, and 7e, and therefore, as the search capacity increases, the search time becomes slower. There is.

In particular, if the data memory 1 and index memory 2 are added to increase the recording capacity, the host CPU 5 will be directly connected to the host CPU 5 through a corresponding control device.
The load on the PU 5 increases, and the search time becomes longer.

Therefore, it is an object of the present invention to provide a file search device that eliminates the drawbacks of the prior art as described above and does not cause delay in search time even when the search capacity is increased.

The file search device of the present invention has at least one data memory and at least one index memory, and the data information of the file is sent to the data memory and the index information of the file is sent to the index memory. In addition to the host CPU that controls the overall operation of the file search device, a sub CPU is provided for each pair of data memory and index memory, and the operation of the search unit consisting of each pair of data memory and index memory is performed by the sub CPU. This is a configuration that is performed via . In this case, in a preferred embodiment, the data memory is comprised of write-once memory, and the index memory is comprised of rewritable memory.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows one embodiment of the file search device of the present invention. In this case, for one host CPU 5,
Two pairs of file search units are provided. As shown in the figure, each file search unit connected to the host CPU 5 has a sub CPU 8, and the sub CPU 8 is connected to a pair of data memories 1 and index memories 2 via respective control units 3 and 4. ing. It should be noted that the number of file search units provided in the present invention is not limited to two pairs as shown in FIG. 1, but may be one pair or three or more pairs. .

What is important is that each file retrieval unit also has a second CPU 8 connected to its respective data memory 1 and index memory 2, and this sub-CPU 8 is connected to the host CPU 5. With this configuration, each file search unit can perform file searches independently of other units.
By performing parallel processing, it is possible to perform quick search processing even in the case of large-volume searches.

-Generally, in file searches, it is desirable that index information for referring to data can be changed depending on the purpose at the time and can be registered in an easy-to-use format. Therefore, it is desirable that the index memory 2 for recording index information is constructed of a rewritable memory. On the other hand, the data information that forms the database of the file itself is basically never changed, so the data memory 1
It is desirable that the system be constructed of a write-once type memory that can additionally record and store information. In each file search unit of the file search device shown in FIG. File management is performed with these as one set. still,
The control units 3 and 4 of each memory control the operation of the memory based on the signals from the sub CPU 8, and for example, control the rotation of the disk recording medium and the movement of the write/read head, as shown in FIG. Regarding the operation of the file search device,
This will be explained with reference to FIGS. 2 and 3.

Note that FIG. 2 is a flowchart showing the operation of the host CPU 5, and FIG. 3 is a flowchart showing the operation of each sub-CPU 8. When the host CPU 5 performs a file search, the search key data is sent to each sub-CPU.
8a and 8b simultaneously, and the host CPU 5 hands over the file search to each sub CPU 8a and 8b. Therefore, the host CPU 5 can perform any other processing until the search results are reported from each sub-CPU 8.
In this respect, the load on the host CPU 5 is significantly reduced.

After receiving the search key data from the host CPU 5, each sub CPU 8 collates it with index information recorded in the index memory 2, and reads file data information from the data memory 1 based on the collation result. , reports the search results obtained in this way to the host C:PU5. Such file search operations can be performed simultaneously and independently in each file search unit under the control of the respective sub CPUs 8. Therefore, a plurality of file search units are arranged and the search Even when there are a large number of files, the search process is basically completed in the search time required for a single file search unit, and the search process time is significantly shortened compared to the conventional technology. Furthermore, by appropriately selecting the interface between the host CPU 5 and the sub-CPU 8 (for example, GPIB interface), even when connecting multiple file search units to the host CPU 5, it is possible to avoid adding hardware on the host side. There is no need to do this, and in this respect it is highly extensible.

When writing a file, the host CPU 5 selects one of the file search units, that is, one of the sub CPUs 8, and transfers the index information and file data information to the sub CPU 8. The selected sub CPU 8 sends the index information to the index memory 2 via its control units 3 and 4.
On the other hand, file data information is written into the data memory 1, respectively.

When writing index information to index memory 2, naturally,
The control unit 4 additionally writes necessary information such as the address to be written into the memory 2, the date, and the error address.

In the embodiment described above, search key data is supplied from the host CPU 5 to each sub CPU 8 simultaneously;
It is also possible to configure the search key data to be supplied sequentially. Furthermore, it is necessary that the recording media used in the data memory 1 and the index memory 2 be a set of recording media in which there is a relationship between the information recorded in each. Therefore, when recording media are set in the respective memories 1 and 2, the sub C and PU 8 inspect whether or not the set combination of recording media is correct, and the inspection results are sent to each file search unit. It is desirable to have a configuration in which the information is displayed on a display device such as a CRT provided at each location. In this case, sub C
It may be displayed directly from the PU8 to the display device, but
It is also possible to have a configuration in which the test results are supplied to the host CPU 5. In such a configuration, the host CPU
It is possible to display the results of all file search units simultaneously on the same display device, such as a CRT, connected to U5.

As explained in detail above, according to the present invention, even when a plurality of file search units are connected to a single host CPU in order to increase the file capacity, the file search operation can be performed easily. Since processing is performed in parallel and independently for each unit, the load on the host C and PU is reduced, and even if multiple units are connected, the processing time is basically the same as for one unit. Therefore, even if the file capacity increases, the search processing time does not increase.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of implementation of the file search device of the present invention, FIG. 2 is a flowchart showing the operation of the host CPU in FIG. 1, and FIG. Flowchart diagram showing the operation of the sub CPU, FIG. 4 is a block diagram showing a conventional file search device, and FIG.
FIG. 2 is a flowchart showing the operation of the device shown in FIG. (Explanation of symbols) 1: Data memory 2: Index memory 5: Host CPU 8: Sub CPU Patent applicant Rico Co., Ltd. 17L-4 Let's enjoy the Lri (7b 1000~ishi〒-9 "8 Crest' 7゜Work 7L-4N

Claims (1)

[Scope of Claims] 1. A file search device for searching files consisting of data information and index information, which includes a data memory for recording data information of each file and an index memory for recording index information, and also includes a data memory for recording data information of each file and an index memory for recording index information. A file search device comprising: a memory control CPU that controls the operations of the file search device connected to a host CPU that controls the entire operation of the file search device. 2. The device according to claim 1, wherein the data memory is a write-once memory, and the index memory is a rewritable memory.