JPH0193219A - System switching system - Google Patents

Abstract

PURPOSE:To prevent the function of a system from being stopped at the time of switching the system to a normal system by setting both the system in use and a spare system at operating states simultaneously at the time of switching the system, and switching the system to the spare system at the time of completing the operation of the system in use and generating the vacant state of a process controller for the system in use. CONSTITUTION:When a process control part receives a system switching request by the request of the process controller, controller IOC0 and IOC1, external memory devices DK0 and DK1, and switching devices SW0 and SW1 are set at non-operating states. Next, it is checked whether the process controller is set at the vacant state, and when it is, the system switching request is issued to a switching circuit SEL. Thereby, the process controller of the spare system becomes the system in use. At this time, system switching interruption is generated in the process controller of the spare system, then, the switching devices SW0 and SW1, the external memory devices DK0 and DK1, and controller IOC0 and IOC1 are recovered to operating states, respectively. In such a way, the operation of the process controller is restarted, then, an ordinary operating state is set.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a system switching method for a duplex system, and in particular, a system switching system suitable for performing normal system switching [Prior art] Duplex system. The system switching method used in the system was originally designed to quickly switch to the standby system and restore normal operation when an abnormality occurs in the active system.

In general, system switching is performed by controlling each duplicated control device independently without going through a switching circuit to ensure continuity of processing, or by sacrificing continuity of processing and switching circuits. This is done by connecting to the control device via. In addition, as described in Japanese Patent Application Laid-Open No. 59-200531,
Control data is built into the control device to minimize system outages due to switching operations. However, when switching systems in the event of an abnormality, some degree of system suspension is unavoidable.

[Problem that the invention seeks to solve]

The idea behind system switching systems in recent years is to ensure the reliability of the backup system and to maintain reliability through preventive maintenance. system to ensure that both systems are used evenly. In other words, it is becoming more common to perform normal system switching instead of abnormal system switching. However, when performing normal system switching,
Since the conventional system switching method follows the abnormal system switching method, there is a problem that the system temporarily stops when switching the system. Abnormal system switching occurs when an abnormality occurs in the current system, and its occurrence frequency is low.

However, normal system switching is performed periodically;
If the system stops working each time, the impact will be significant.

An object of the present invention is to provide a system switching method that does not stop system functions during normal system switching, where system switching operations occur frequently.

[Means for solving problems]

The above object is to provide a system switching method for a working system and a standby system in a duplex system connected by a system switching circuit, in which the system switching circuit connects the active system to the backup system at the time of system switching.

This is achieved by putting the subordinate devices of both the backup system into an operating state, and switching to the backup system when the operation of the subordinate devices of the active system ends and the active processing control device becomes idle.

[For production]

The duplicated processing control devices are interconnected by a switching circuit. This switching circuit has a function of keeping each device in the standby system in an operating state, and each of the duplicated devices is brought into an operating state by the switching circuit when the system is switched.

In other words, each device under both systems is connected as an active system, and the active system processing control device transmits information through the switching circuit with respect to each device on the standby system. When the system is switched, the switching circuit is controlled to be switched and connected to the other system. In other words, the switch is always made to the standby system that is in operation as the active system, so the system switching time is Minimum. Note 1. While the active system is in operation, switching connections to other systems are performed, but if this occurs, continuity of operation is required in the backup system, which complicates the switching circuit and also complicates restart control in the backup system. However, in the present invention, the active system is searched for when it is in an empty state, and when it becomes empty, it is controlled to be switched and connected to another system, so even after switching to the standby system, the standby system remains in the idle state of operation. restart control, and the restart time due to system switching is minimized. In this way, there is no need for the backup system to continue the active state of the active system, and the backup system can reliably resume operation from an empty state. Moreover, software control can also be simplified.

〔Example〕

Hereinafter, one embodiment of the present invention will be described based on the drawings.

FIG. 1 is a system configuration diagram of an embodiment of the system switching system according to the present invention. Here, each device (memory device MMO, external storage device DKO, switch device SWO, input/output device 10CO, and system control device 5cO) subordinate to the processing control device CPO is controlled by the processing control device cpo.
Each device (memory device MM1. external storage device DK 1. switch device SWl, human output device 10CI, and system control device 5CI) that is an O-system device and subordinate to the processing control device CPI is controlled by the processing control device cpi. system device. SEL is the system control unit SCO, SC
This is a switching circuit that switches each device in the active system and each device in the standby system to be connected to either the processing control device cpo or crt. First, the switching circuit SEL sends information as to whether the control system is the active system to the system control device SCO or SC.
I, the active system processing control unit cpo receives information from both the active system and standby systems MMO, MMI, DKO, DKI, etc.
・Pull in.゛The processing control units CPO and CPI control the entire system, and the processing control unit CPI, which has become a backup system, is in a standby state, and the processing control unit cpo, which has become an active system, controls each device in its own system ( MMO, DKO, SW
Input/output data to/from (O, l0CO),
Each device of other system (MM 1.

DKI, 5WI) only outputs data,
Keep each device in operation at all times. For example, writing to a memory device is performed through the switching circuit SEL between the own memory device MMO and the other memory device MMI. Reading from the memory device is performed only from the own memory device MMO. The processing control units CPO and CPI have process control units PCO and Pct, respectively, which control the start processing, termination processing, and queuing processing of the processing process in response to requests from the processing control units CPO and CPI, respectively, as follows. I'm going to

FIG. 2 is a process control flow diagram and a process management table of the process control section. Here, the processing control device is
When an interrupt from each device, a clock interrupt, or a command request occurs, a request is made to the process control unit to start the process.

A standby request for the process, a stop request for the process, or a system switching request is issued.

The process control unit PC accepts this request and controls the process to start, stop, and standby.

This process control unit PC performs control management using a process state management table consisting of two tables: a process startup table and a process WAIT table. When a process activation request is accepted, the process is registered in the process activation table and the process is activated. When the process is on standby, when registering the process in the process activation table, the process is deleted from the process WAIT table and WAIT is released to activate the process. When a standby request for a process is accepted, since the process is being activated, the process is moved from the process activation table to the process WATT table and the process is brought to a halted state. When a request to stop a process is accepted, if the process is running,
The process is deleted from the process startup table and the operation of the process is stopped. Further, when the process is on standby, the process is terminated by deleting the process from the process WAIT table.

FIG. 3 is a process control flow diagram of an embodiment of the system switching method according to the present invention. The processing control device receives interrupts from each device and clock interrupts.

When a request by a command occurs, it issues a process start request, a process stop request, a process standby request, or a system switching request to the process control unit. As a result, the process control unit accepts this request and starts, stops, and puts the process into a standby state using the process control table.

When the process control unit accepts a system switching request in response to a request from the process control device, first, each device, that is, 10 control devices 10
The CI, 10C2, external storage devices DKO, DKI, and switch devices SWO, SWI are set to a usage prohibited state, respectively. As a result, the process that performs input/output for each device from now on will be
Since the device becomes unusable, it is in a process standby state until it becomes usable. Next, it is checked whether the processing control device is in an empty state. If it is empty, the switching circuit SEL
System switching is performed by issuing a system switching request to the system, and the standby system processing control unit becomes the active system, and not only each device in its own system but also each device that has become the standby system is connected through the switching circuit SEL. Put into working condition.

When this system switching request occurs, a system switching interrupt is generated in the standby system processing control device, and by extension of this interrupt, each device, that is, switch devices Hswo, sWl, external storage devices DKO, DKI, and To control device. Return 10CI and 10C2 to their usable states. At this time, since each device is in a disabled state, a request to start up the waiting processes is issued to the process control unit, which accepts the request and starts up the processes in sequence.

As a result, the processing control device restarts its operation and enters a normal operating state.

FIG. 4 is a flowchart for checking whether the processing control device is in an empty state. Upon receiving a process activation request from the processing control device, the process control unit activates the process and simultaneously registers the process in the process activation table. This means that the process is running and the processing control device is also in use. When a process uses each device, for example, when using the external storage device DKO, it can use the device DKO only when the external storage device DKO is available.

However, when the device to be used is disabled, for example,
When the IO control device 10GO is disabled, the process issues a standby request to the process control unit to enter the standby state, and simultaneously moves the process from the process execution table to the process WAIT table. Also, when a process ends, the process control accepts a process stop request from the process control device, stops the process, and at the same time deletes the process from the process startup table.

In this way, the process control unit accepts requests for starting, stopping, and waiting processes from the processing control unit, and starts and stops processes based on the process starting table. The process startup table allows management of whether the processing control device is running or in an empty state. If a process is registered in the process startup table, it means that the processing control device is running, and if there are no processes in the process startup table, it means that the processing control device is in an empty state.

As explained above, in electronic exchanges, the usage rate of the processing control device is generally around 60%, and the processing control device is often in an empty state, so system switching is performed when this processing device is in an empty state. It is effective to do this. Furthermore, since the system is switched due to an empty state, it is no longer necessary to maintain continuity of processing, and the system switching time is greatly shortened. Software control is also simplified.

〔Effect of the invention〕

According to the present invention, since system switching is performed when the processing control device is in an empty state, there is no need to continue processing on the standby system, and software control can be greatly simplified.

In addition, system switching can be performed without considering the system status.
Furthermore, since there is no system switching time, there is no interruption of functions due to system switching.

[Brief explanation of the drawing]

FIG. 1 is a system configuration diagram of an embodiment of the system switching method according to the present invention, FIG. 2 is a process control flow diagram of an embodiment of the system switching method according to the present invention, and FIG. FIG. 4 is a process control flow diagram of an embodiment of the system switching method according to the present invention. FIG. CPO, CPI...processing control device, PCO, PCI・
...Process control unit, sco, sci...system control device, swo, swt...switch device, DKO. DKI...external storage device, MMO, MMI...storage device, l0CO, l0CI...10 control device, TTY
...Typewriter, SEL...switching circuit. Agent Patent Attorney Tadashi Akimoto Actual Figure 1 Figure 3 Figure 3 - 1 - Child 4

Claims (1)

[Claims] 1. In a system switching system between the active system and the standby system of a duplex system connected by a system switching circuit, at the time of system switching, the system switching circuit puts the subordinate devices of both the active system and the standby system into the operating state, and the active system A system switching method characterized by switching to a standby system when a subordinate device completes its operation and an active system processing control device becomes idle.