JPH0650501B2 - Preliminary bulk duplication method for multiple computer system - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to a spare bulk duplication method for an electronic computer system, particularly a multiple computer system for improving reliability in the multiple computer system.

[Technical background of the invention]

The multi-computer system is one of the methods for improving the reliability of the entire system, and the duplication of bulk is one of the methods for improving the reliability.

FIG. 7 is a system diagram in which a bulk is duplicated in a conventional general multi-system computer system. In FIG. 7, 1 is a central processing unit (hereinafter referred to as CPU), 3 is a master auxiliary storage device, and 4 is a slave auxiliary storage device (hereinafter, auxiliary storage device is referred to as bulk). A single computer system in which the bulk is duplicated with the individual bulks is configured. Also, 2 is a CPU, 5 is a master bulk, and 6 is a slave bulk, and these also constitute another computer system. Then, the above systems are combined to form a multi-system computer system. In short, each computer has two bulks.

[Problems of background technology]

Although the reliability of the bulk has increased due to recent technological improvements and the frequency of simultaneous failures of multiple bulks has decreased, the bulk of the computer systems that make up the conventional multi-system computer system is The system is completely duplicated and has a poor cost performance.

[Object of the Invention]

The present invention has been made to solve the above problems, and an object of the present invention is to provide a spare bulk duplication method for a multiple computer system with good cost performance without lowering the reliability of the entire system. .

[Outline of Invention]

In the present invention, at least one or more spare bulks are provided for a plurality of computer systems so that each central processing unit can access the spare bulks, and the spare bulks are connected to each other as needed to form a duplicated bulk configuration. It's about to get.

Example of Invention

Embodiments will be described below with reference to the drawings. FIG. 1 is a block diagram of an embodiment of a multiple computer system according to the present invention. In addition, in order to simplify the description, it is shown as a dual computer system. In FIG. 1, the same parts as those in FIG. Reference numeral 7 denotes a spare bulk, which can be accessed from each CPU 1 and 2.
Other configurations are the same as in FIG. 7.

The spare bulk 7 does not belong to either the computer system including the CPU 1 and the bulk 3 (hereinafter referred to as 1 system) or the computer system including the CPU 2 and the bulk 5 (hereinafter referred to as 2 system). That is, in this state, bulk duplication is not performed in both systems.

FIG. 2 (a) is a flow chart showing the processing for duplicating the bulk of one system, and FIG. 2 (b) is writing data to the bulk 3 from another process even during the duplication processing. Therefore, it is a flowchart showing the processing in this case.

A method for duplicating the bulk 3 and the spare bulk 7 will be described below.

First, in step 21, an instruction is made as to which bulk should be used for bulk duplication. This is to select an unused spare bulk from a plurality of spare bulks. That is, the configuration of FIG. 1 shows a double system as an example, but it is possible that a plurality of spare bulks exist in a multiple system of a double system or more. The designated spare bulk is the slave bulk, and the bulk already connected to the CPU and operating the system is the master. In step 22, the memory area of the CPU is cleared in preparation for storing the update information of the master bulk from the start of copying. That is, in step 24, copying is performed in order to match the bulk contents of both the master and slave, but this requires a time of several minutes to tens of minutes. However, the contents of the master bulk are updated during this period. Therefore, this content also needs to be reflected in the slave bulk after the copy is completed. In step 23, based on the initialization in step 22 described above, FIG.
Information for instructing to start saving the master update information is set in step 2C of the writing process to the bulk 3 in (b).
Then, in step 24, the entire contents of the master bulk are copied to the slave bulk. At this time, the master update information storage is instructed to step 2B on the writing processing side in FIG. 2 (b). In step 25, an instruction to update the slave bulk is issued. That is, since the contents of the master bulk and the slave bulk match in step 24 (except for the portion where the master bulk is updated after the copy is completed), step 2A
In, set the information to instruct the slave bulk contents to be updated when the master bulk contents are updated.
In step 26, the instruction to stop saving the master update information is set to step 2B as described above.

That is, in step 25, the slave bulk contents are updated at the same time when the master bulk is updated, so information indicating an instruction to stop saving the master update information is set. In step 27, the data is copied to the slave bulk based on the update information of the master bulk. That is, even during copying from the master bulk to the slave bulk in step 24, the master bulk contents are updated, and the update information is already saved in step 2C of FIG. 2 (b). Therefore, based on this update information,
Read the contents of the master bulk and write to the slave bulk. By this processing, the contents of the master bulk and the contents of the slave bulk completely match. Thus, FIG.
By performing steps 21 to 27 of (a), the bulk 3 and the spare bulk 7 are duplexed. FIG. 3 is a diagram showing a state in which bulk duplication is performed as a result of the above processing.

FIG. 4 is a flow chart for explaining the process of canceling the bulk duplication and the process of exchanging the master bulk and the slave bulk.

In FIG. 4, steps 41 to 45 are bulk 2
This is a process for stopping the duplication, and step 46 describes the process for exchanging the master bulk and the slave bulk. First, in step 41, it is determined whether or not the bulk is being duplicated. If it is, the process proceeds to step 42. If not duplicated, the process proceeds to step 47 and an alarm is issued. That is, the alarm is issued to notify the operator when the operation is performed when the bulk is not duplicated. In step 42, designation is made as to whether the duplication of the bulk is stopped or the master bulk and the slave bulk are exchanged. If the duplication is stopped, the process proceeds to step 43, and if the bulk is replaced, the process proceeds to step 46. In step 43, it is selected which one of the master bulk and the slave bulk is separated to stop the duplication. Then, in the case of master bulk separation, the process proceeds to step 44, and in the case of slave bulk separation, the process passes step 44 and proceeds to step 45. In step 44, when the master bulk is separated, the master bulk and the slave bulk are temporarily exchanged. This is because the bulk finally connected to the CPU 1 is the master bulk. In step 45, the slave bulk is separated from the CPU 1, and further information for stopping the updating of the slave bulk contents is set in step 29 of FIG. 2 (b). By carrying out Step 41 to Step 45 in this way, the duplication stop of the auxiliary storage can be realized. If the master bulk and slave bulk are to be replaced, step 4
Carry out 6. FIG. 5 is a view showing a state in which the bulk 3 is separated and the bulk duplication is stopped. FIG. 6 is a diagram showing a state in which the master bulk 5 of the two systems and the spare bulk 7 are duplicated. In this case also, the processing method is the same as that of FIG. 5 described above.

As is clear from the above description, since the system reliability is sufficient for bulk reliability, a dual computer system is sufficient, but one system may be overhauled, regularly inspected, or maintained by software. The system may be stopped, and in this case, the reliability of the system can be secured by setting the other system as a duplicated bulk by using a spare bulk. Furthermore, even if the bulk fails, it can be merged into the system when the repair is completed, and the system can be stopped as before, and the contents of both bulks can be matched and then initialized. There is no need to start up. In any case, 2 bulks for each system
There is an advantage that it is possible to share a spare bulk without having to duplicate.

Further, as an embodiment, the spare bulk is a pack exchange type, and when the bulk duplication is not performed, the software can be updated by inserting the pack storing the system program source and the database source. Becomes It is also possible to put another pack into a double bulk structure once and then separate it to obtain the latest backup bulk. That is, by using the pack type, the disc can be effectively used.

In the above embodiment, the case of the dual computer system has been described in which each system shares one spare bulk. However, basically, from the above-mentioned improvement in the reliability of the bulk, it is clear that even if the bulk is composed of two or more N heavy systems, at least one spare bulk is sufficient.

〔The invention's effect〕

As described above, according to the present invention, it is possible to achieve the same effect as that of duplicating all the auxiliary storage devices with at least one spare bulk in the multiplex system, and the cost performance is improved. It is possible to provide a preliminary bulk duplication method for a good multi-computer system.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of a multiple computer system according to the present invention, and FIG. 2 (a) is a flow chart showing a process for duplicating a bulk of one system in the case of a dual system, FIG. 2 (b) ) Is a flowchart showing the process of writing data to the bulk,
The figure is a state diagram in which the bulk of one system is duplicated. Fig. 4 is a flowchart showing the process of canceling the bulk duplication and the process of exchanging the master bulk and the slave bulk. Fig. 5 shows the master bulk being separated. FIG. 6 is a state diagram in which the duplication is stopped, FIG. 6 is a state diagram in which the second system side is duplicated, and FIG. 7 is a conventional duplicated bulk configuration diagram. 1, 2 ... Central processing unit 3, 4, 5, 6 ... Auxiliary storage device 7 ... Spare auxiliary storage device

Claims (1)

[Claims] 1. A plurality of central processing units and an auxiliary storage device provided corresponding to the central processing units (hereinafter referred to as bulk).
In an electronic computer system in which the system composed of at least two series or more forms a multiple system, at least one spare bulk is provided for each system, and each bulk is processed by each central processing unit. Accessible to the device, the central processing unit determines which spare bulk to use in duplicating the bulk, and copies all the contents in the main bulk to the selected spare bulk, The update information to the main bulk is saved from the start of this copy, and after the copy to the spare bulk is completed, the saving of the update information to the main bulk is stopped, and thereafter, the main bulk is updated based on the update information. Is stored in the spare bulk and is stored in the spare bulk.