JPS593916B2 - Resume processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention utilizes a simple inter-system communication method in an electronic switching system consisting of two systems, an active system and a standby system, to eliminate the need for a secondary file device for saving call information, and to improve call processing. This invention relates to a quick restart processing method in a standby system that reduces service degradation.

Conventionally, in this type of restart processing method, as shown in Figure 1, the core call information necessary for restoring the system state when a failure occurs is stored in a secondary file storage (FM) (magnetic drum, magnetic bubble, magnetic disk, floppy disk, etc.). (disk, IC memory, etc.), and when a failure occurs, the processing state is restored based on the information in the secondary file storage device (FM).

Conventionally, in this type of restart processing method, as shown in Figure 2, a information transfer unit (DTU) is installed between the systems, and both prefectures control the transfer equipment to transfer call information from the active system. method was used. The conventional system as shown in FIG. 2 not only requires an information transfer unit (DTU) but also requires complicated processing to control the information transfer unit (DTU). In addition, conventional restart processing equivalent to so-called phase 0.5 was applied to DIO type electronic exchanges and D20 type electronic exchanges.
However, this was done in my own family.

However, there was no other standby system that was performing phase 0.5 restart processing. An object of the present invention is to eliminate the above-mentioned drawbacks of the conventional system, and to save call information as in the conventional system in a redundant electronic switching system.
No next file storage device or inter-system information transfer device is required, and the phase is 0.5 or more in the standby system (For the phase restart processing program, please refer to 1. Program (Part 2)
)7. Restart process 1! The present invention provides a restart processing method in an electronic exchange that is capable of performing restart processing with serviceability (described in detail in 2007), so the description thereof will be omitted.

In order to achieve the above object, the present invention provides an electronic switching system consisting of two systems, an active system and a standby system, each equipped with a main storage device, by controlling the settings of a gate circuit during a write operation of the main storage device of the own system. It has an inter-system information transfer path that has the function of writing in parallel to the main memory of other systems, and each time a transfer request occurs, the central processing unit of the own system releases the transfer path by setting a gate circuit, and then By executing a write command to the device and transferring the transfer data to the same location on the other system's main memory as the own system's main memory, when a fault occurs in the active system, the standby system's central processing system can transfer the transferred information. The feature is that the call processing state is restored based on the .

Next, the details of the present invention will be explained by referring to examples.

FIG. 3 is a block diagram showing an example of crossing paths between the central control unit and the main memory. The gate circuit Glj opens and closes a path between the central control unit and the main memory, thereby controlling information transfer between systems. Gate circuit G
There are four states shown in FIG. 4 depending on the opening and closing of ij, and states 2 and 3 are duplex operation processing states. State 1 is a state in which both systems are independently performing different processes, and state 4 is a state in which both systems are inoperable. When information is transferred to the standby system while the system is operating, taking the system state 2 in Figure 4 as an example, open the crossing path and transfer the data area shown in Figure 5 to M.
It is transferred to the spare main memory by reading from MO (see 8) and writing to MMO and MMl (see 9), and after the transfer, the crossing path is closed.

FIG. 6 is a diagram showing the time flow of call processing operations. Using FIG. 6, data to be transferred and restart methods will be listed in three cases. Resuming method A (equivalent to phase 1) (1) Immediately before the transfer information processing state enters a stable state (see 1.6) Device status indicating what kind of device resources (trunk/communication path) are being used Transfer information and a flag indicating a stable state to the backup system.

Also, immediately before the processing state changes from the stable state to the processing state (see 2.7), a flag indicating the processing state is transferred to the standby system. (2) Resuming method When a failure occurs in the active system, the standby central processing system restores the call processing state using only the transferred equipment state information and initializes all other equipment groups, including the idle state.

Resuming method B (equivalent to phase 0.5) (1) Transfer information In addition to the information transferred using restart method A, immediately before issuing a command to the trunk/communication path (see 3), the edited command information group for the device is sent to the backup system. Transfer to main storage.

(2) Resume method When the active system fails, the standby central processing system restores the corresponding call processing state from stable equipment state information, initializes only the equipment group that is in an unstable state or in a command execution state, and performs other operations. The empty equipment group will be left alone.

(3) In effect restart processing method 1, all empty trunks and communication paths are initialized using commands for trunks and communication paths that take time to control one device.

In this method, empty trunks and communication paths are left unattended, resulting in a significant reduction in processing time. Resuming method C (equivalent to phase 0.1) (1) Transfer information In addition to the information transferred in restart method 2, display information during command sending (see 3) before sending a command to the device, and display information during command execution (
(Refer to 4) Transfers a communication line transaction pointer indicating which command is being executed, and upon completion of the command (Refer to 5) Transfers command completion display information to the spare main storage device.

(2) Resume method When the active system fails, the standby central processing system restores the corresponding call processing state from the device information in the stable state, and if input processing or internal processing is in progress, based on the transferred information. Processing is performed to restore the state, and if a command is being sent to the device, only one command that is currently being executed is re-executed.

(3) Effects Since the execution time of an instruction for the tranter/talk path is significantly longer than the running time of the program, the time required for restart processing can be approximated by the number of instructions issued to the tranter/talk path.

With this method, only one command is issued to the trunk/communication path, so the restart processing time is minimized. As explained above, according to the present invention, in a duplex electronic switching system, an intersection is provided in the path between the central control unit and the main storage device,
When the active system central controller writes call information to the main memory, it temporarily opens the crossing path and writes the same information to the backup main memory at the same time, making the stored contents of both systems match, and then restarting the active system. By restoring the call processing state based on the information of the corresponding zeta area on the standby system in the event of a failure, it becomes possible to economically restart the electronic switching system to maintain processing continuity, and the effect is extremely large. It is.

[Brief explanation of the drawing]

FIG. 1 shows a call information saving method in a conventional restart processing method, FIG. 2 shows an inter-system information transfer method in a conventional restart processing method, and FIG. 3 shows an inter-system information transfer method used as a means in the present invention. FIG. 4 is a system state diagram according to the present invention, FIG. 5 is a diagram showing the inter-system notification method according to the present invention, and FIG. 6 is a diagram showing at what point in the call processing operation information is transferred. CC is a central control unit, MM is a main memory, DTU is an intersystem information transfer device, FM is a file storage device, DCH is a data channel device, G is a gate circuit, and arrows indicate data flow.

Claims (1)

[Claims] 1. In an electronic switching system consisting of two systems, a working system and a standby system, each equipped with a main storage device, the main storage system of the other system is It has an inter-system information transfer path with a function that allows writing to the storage device in parallel, and each time a transfer request occurs, the central processing unit of its own system opens the transfer path by setting a gate circuit, and then transfers information to the main storage device. By executing the write command and transferring the transfer data to the same location on the main memory of the other system as the main memory of the own system, when a failure occurs in the active system, the central processing system on the standby system can use the transfer information based on the transfer information. A restart processing method characterized by restoring a call processing state. 2. In claim 1, the transferred data includes device status information and stable state display information before the processing state of the active central processing unit enters the stable state, and stable state end display information after the stable state ends. A restart processing method that transfers the data to the spare main storage device. 3 In claim 1, the transferred data includes device status information and stable state display information before the active central processing system enters the stable state, stable state end display information after the stable state ends, and stable state end display information. A restart processing method that transfers the edited command information group for the device to the standby main storage device before sending the command to the device. 4. In claim 1, the transferred data includes device status information and stable state display information before the processing state enters the stable state, and stable state end display information after the stable state ends. , Before sending a command to the device, the edited command information group for the device and the display information during command sending, which command is being executed when the command is executed, and the command completion display information when the command is completed are transferred to the spare main storage device. restart processing method.