JPH0266635A - Recovery controller - Google Patents

Abstract

PURPOSE:To secure the synchronization between main memories by reading only the data that is held between the start information and the end information out of another memory among those data stored in other memories with occurrence of a fault and restoring the read-out data to the main memory. CONSTITUTION:When a fault occurs, the contents recorded into a journal file 6 are once read and the process number of the journal data held between a start record 13 and an end record 14 having the same identifier 12 is picked up. When this process is over, the file 6 is rewound and the journal data on the picked-up process number is read out of the file 6. Then this journal data is restored into a data file 5. Therefore the data on the process number where only the record 13 is included with no existence of the record 14 is excluded out of the recovery subjects.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to a recovery control device used in an operating system such as a medical office system.

(Prior Art) In recent years, in the field of medical administration, journal data such as basic patient information, insurance information, medical examination information, etc. is written over time to a magnetic disk, which is the main storage means, and the data can be updated as needed. A medical office system that updates journal data and transmits it to a target terminal is widely used.

By the way, such a system is equipped with a magnetic tape device that records journal data recorded on a magnetic disk as evacuation data every time the contents are updated, and in the event of some kind of failure, the magnetic tape device can be saved. By transferring the recorded saved data to the magnetic disk device, the system is restored to the state before the failure occurred and processing is restarted. In this case, data transfer is performed by treating a plurality of mutually related records as one transaction data.

Hereinafter, this process will be referred to as recovery process.

However, with such conventional recovery processing, if a failure occurs while updating the data recorded on the magnetic disk, incomplete data is restored to the magnetic disk, causing the system to fail. There was a problem in that it was not possible to restore to the previous state.

In addition, multiple main storage means (
In systems that have a magnetic disk drive (magnetic disk drive), there may be discrepancies in the contents between the main storage means on the side that performed recovery processing and the main storage means on the side that did not perform recovery processing, and the synchronization between each main storage means will occur in the next processing operation. There was a problem that it could not be removed.

(Problems to be Solved by the Invention) The present invention is intended to solve the above-mentioned problems in the conventional recovery processing. By discarding the restoration of data in progress, it is possible to restore the contents of the main storage means to the state immediately before the failure occurred, and even in the case of having multiple main storage means, it is possible to maintain synchronization between each main storage means. The purpose is to provide a recovery control device that can perform

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above-mentioned object, the recovery control device of the present invention transfers data stored in the main storage means to other data every time the contents of the main storage means are updated. In a recovery control device that saves data stored in one storage means to the main storage means and restores the data stored in another storage means to the main storage means when a failure occurs, this data is saved before and after a set of data stored in the other storage means. A start/end information recording means for additionally recording information indicating the start/end of data, and only the data sandwiched between the start information and end information among the data stored in other storage means due to the occurrence of a failure. The recovery execution means reads data from the storage means and restores it to the main storage means.

(Function) In the recovery control device of the present invention, the recovery execution means
Among the data stored in other storage means due to the occurrence of a failure, only the data sandwiched between the start information and end information recorded by the start/end information recording means are read from the other storage means and stored in the main memory. Therefore, even if a failure occurs while updating data stored in the main memory, the restoration of the data being updated will be discarded and the contents of the main memory will be restored when the failure occurs. It is possible to restore the previous state, and even in the case of having a plurality of main storage means, it is possible to maintain synchronization between the main storage means.

(Example) Hereinafter, an example of the present invention will be described in detail based on the drawings.

FIG. 1 is a block diagram for explaining the configuration of a medical office system equipped with a recovery control device according to an embodiment of the present invention.

In the figure, 1 is an F that controls the system in an integrated manner.
E P (rront-end processor
), 2 is B E to reduce the load on FEPI.
P (back-end processor), 3 is a workstation for manual journaling of basic patient information, insurance information, medical examination information, etc. related to medical administrative processing, and 4 is used for business monitoring and data processing intervention for FEPI It's a console.

Further, 5 is a data file of a magnetic disk device, etc., which is a main storage means in which journal data input from each workstation 3 is recorded.
includes a normal file in which journal data is recorded, a start file in which a start record indicating the beginning of the data is recorded each time input of a related set of data is started from the workstation 3, and a workstation 3. Each time input of a group of data is completed, an end record indicating the end of the data is recorded.

Further, reference numeral 6 denotes a journal file such as a magnetic tape device, which is another storage means, in which the journal data recorded in the data file 5 is recorded as save data every time the contents are updated.

The FEP 1 described above is equipped with a recovery control section (not shown) that performs recovery control specific to this embodiment system according to a predetermined recovery processing program.

Next, the data recording operation in this embodiment system will be explained using FIGS. 2 and 3.

Note that FIG. 2 is a diagram showing the format of data recorded in the journal file, and FIG. 3 is a flowchart for explaining the flow of the journal data writing operation.

As shown in FIG. 2, in the journal file 6 described above, a plurality of records 11 inputted from each workstation 3 are recorded in a temporal manner. Here, we divide the record group shown in this figure into two based on their relationship, and
Let a to lld be the record group of process A, and ll
Let the records from e to 11°g be the record group of process B.

Further, this process is identified by an identifier 12 attached to the head of each record 11.

Now, in this embodiment system, when data input of process A is started from workstation 3, process A is recorded in start file 5a in data file 5.
Data indicating the beginning of the record group is recorded (step 31 in FIG. 3).

Subsequently, the data recorded in the start file 5a is saved to the journal file 6, and recorded as a start record 13 on this journal file 6 with the identifier 13 of process A added thereto.

Next, normal writing processing of journal data input from the workstation 3 to the various data files 5 is performed (step 32 in FIG. 3).

After this, the data recorded in the data file 5 is saved to the journal file 6, and this journal file 6
It is recorded with the process A identifier 13 added thereto.

When writing of the journal data to the cheetah file 5 is completed, data indicating the end of the record group of process A is then recorded in the end file 5b (step 33 in FIG. 3).

Thereafter, an end record 14 is additionally recorded after the last record 11 recorded on the journal file 6.

With the above steps, saving data to the journal file 6 is completed.

Next, a recovery operation when some kind of failure occurs in the system will be explained with reference to FIG.

First, when a failure occurs, the contents recorded in the journal file 6 are read through (STEP 41) and the same identifier 1 is used.
The process number of the journal data sandwiched between the start record 13 and end record 14 having 2 is picked up (step knob 42).

When this process is completed, the journal file 6 is rewound (step 43), the journal data of the picked up process number is read from the journal file 6 (step 44), and this is restored to the data file 5 (step 45). ).

Therefore, data of a process number in which only a start record 13 exists without an end record 14 is excluded from the recovery target.

Thus, according to this embodiment, even if a failure occurs during the update of data recorded in the data file 5, the end record 14 is not additionally recorded on the journal file 6 for the data that is being updated. , are not subject to recovery, and as a result, the system can be restored to the state immediately before the failure occurred, and subsequent processing can be resumed promptly.

Furthermore, even if there are multiple data files 5 in which the same content is recorded, there will be no discrepancy in the content between the data file on the side where the recovery process is performed and the data file on the side where the recovery process is not performed. There is no need to worry about losing synchronization between files.

[Effects of the Invention] As explained above, according to the recovery control device of the present invention, even if a failure occurs during the update of data stored in the main storage means, the restoration is discarded for the data being updated. By doing so, the contents of the main storage means can be restored to the state immediately before the failure occurred, and even when a plurality of main storage means are provided, synchronization between the main storage means can be maintained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram for explaining the configuration of a medical office system equipped with a recovery control device according to an embodiment of the present invention;
Figure 2 is a diagram showing the data format of the journal file used in the medical office system shown in Figure 1, Figure 3 is a flowchart showing the flow of journal data writing processing in the medical office system shown in Figure 1, and Figure 4 is a diagram showing the data format of the journal file used in the medical office system shown in Figure 1. FIG. 2 is a block diagram showing the flow of the same recovery process. 1...REP, 2...BEP, 3...Workstation, 5...Data file, 6...Journal file, 11...Record, 12...Identifier, 1
3...Start record, 14...End record. Applicant: Toshiba Corporation

Claims (1)

[Claims] (1) The data stored in the main storage means is saved to another storage means every time the contents of the main storage means are updated, and when a failure occurs, the data stored in the other storage means is saved to the main storage means. In the recovery control device for restoring data to a computer, a start/end information recording means adds and records information indicating the start/end of the data before and after a set of data stored in the other storage means, and recovery execution means for reading only the data sandwiched between the start information and end information from the other storage means among the data stored in the other storage means and restoring this to the main storage means; A recovery control device characterized by: