JPS61181249A - Channel switching device - Google Patents

Abstract

PURPOSE:To decrease a talking intermission time at channel switching without provision of a crossing between channel systems by providing a channel control processor at each separate system of a duplicated system. CONSTITUTION:If a fault takes place in a talking system SW00 in the operating system SP0, a channel control processor TP0 uses the SW00 as the standby system and uses an SW01 as the operating system and brings the access mode as 'Both system write and operating system read'. Thus, the talking not related to a faulty part is not intermitted but the talking is kept. On the other hand, when a central control processor CP receives a fault notice from the TP0, the CP gives a command of the changeover of the SW and of the access mode by one system. When the talking system not in fault is changed over by SW11, SW12, the TP0 brings the access mode into the mode of one system by the SW01 according to the command from the CP.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a switching device for duplexed communication paths in an exchange, and particularly to a communication path switching device suitable for switching a faulty communication path.

[Prior art]

Dual systems include a heat reserve system and a cold reserve system.

The distinction is based on whether the standby system is in operation or not.
When the standby system is in operation, it is called the heat reserve system, and when it is stopped, it is called the cold reserve system.

In an electronic switching system with a duplex time-division communication path that employs a thermal reserve system, the writing and reading modes (hereinafter referred to as access modes) of the communication path memory, holding memory, etc. are as follows:
When both systems are normal, the memory contents of both systems are matched by "writing on both systems and reading on the active system (operational system)."

When a failure occurs in the active communication path, it is necessary to make the backup system the active system, disconnect the faulty system from the system, and at the same time change the access mode to "one-side write, one-side read" for only the disconnected system. be.

However, in a multiprocessor system, especially a load-balanced multiprocessor system, the communication path system is divided and controlled by Ijj number of processors. , selects an active system different from the active system of the communication path controlled by another processor, and the call is interrupted.

One possible way to deal with this problem is to provide interlacing in the shunter (the communication line that connects the switches) that connects each communication path system, thereby eliminating the need for each communication path operation system to be the same. Here, providing a confound means providing a communication path from the active system to the standby system and from the standby system to the active system.

However, junctor entanglement in the communication line system has the following problems compared to the case where there is no interlace.

1. Hardware Issue 2: ■Doubling the number of connectors complicates mounting work. (2) It is necessary to add a selection circuit (a function to display the active system for the opposite communication path system or a function to put the active system signal on both junctions), resulting in an increase in the number of packages and the number of connections.

2.Software/Software 2nd problem: Resource management becomes more complicated due to the increase in the number of junctors.

Regarding the switching of duplex communication channels, please refer to Japanese Unexamined Patent Publication No. 57-18
There is No. 5787. However, according to the description in this publication, no consideration is given to the simultaneous switching of each communication path in multiprocessor control.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide a communication path switching device that minimizes call interruption time due to rough 1-ware processing in order to eliminate junctor confounding between communication path systems that are divided and controlled by a plurality of processors. .

[Summary of the invention]

The present invention provides for a distributed and duplicated heat reserve dual communication line system, in which one of the duplexes is connected as an active system by a junction, and the other is connected as a standby system by a junction,
First, a communication path control block was provided in the dual distributed communication path system 47j to connect it to the central control processor. In this configuration there is no junctor confounding. When a failure is detected in the active system in the dual distributed communication path system supported by the communication path control processor, the currently operating active systems are set to standby = 3-, and the standby systems that are on standby are all set to standby. and make it operational.

In addition, after the processor that detected the failure performs system switching,
Although there is a possibility that the call will be interrupted until the other processors perform system switching, this can be greatly reduced by transitioning the access mode to the one-system mode after all processors are switched.

[Embodiments of the invention]

FIG. 1 shows an embodiment of the invention. The entire communication path system is a duplicated distributed communication path system swooo, 5WOI; 5W.
iO, 5WII; Consists of 5W20, 5W21. That is, the number of distributed units is 3, 5WOO, 5WOI, 5WIO
and 5WII, 5W20 and 5W21 each constitute a double system. Furthermore, 5WOO, 5WIO, and 5W20 are used as active systems, and 5WO1, 5WII, and 5W21 are used as standby systems.
The entire active system is collectively handled as the active channel system SPO, and the entire standby system is collectively handled as the standby channel system SPI.

Communication path control processors TPO and T support distributed communication path systems.
Pl, TP2 were provided. These three processors constitute a so-called distributed multiprocessor.

Three communication path control Procerasago PO, TP1. .

A central control processor CP is provided to have the role of controlling and managing TP2.

In the operational communication path system spo, swooo and 5WIO.

5WIO and 5W20, and 5W20 and swooo were connected by a junctor (line) so that they could communicate with each other. Similarly, in the standby channel system S P 1, SWo
1- and 5WII, SWl, 1 and 5W21゜SW21 and S
A mutual connection was made with WO1 using a junctor (■).

Such interconnections are not confounding. Confounding is swooo
This refers to the provision of a junctor between the active system and the standby system, such as 5WII, 5WOO, and 5W21. In the above embodiment, the connection between the active systems.

This is a mutual connection between standby systems and does not mean confounding.

The operation of the configuration shown in FIG. 1 will be explained using FIGS. 2, 3, and 4.

Under normal conditions with no failures, 5W00,5W
iO, 5W20 works as an operational system, 5WOI, 5WI
I, 5W21 works as a standby system.

This instruction is TPO, 5 for swooo and 5WO1.
WIO and SWl, ]- for TPI, 5W20 and S
TP2 performs this for W21. In this case, the contents of the access are such that writing is performed on both the active system and the standby system, and reading is performed only on the active system.

Now, as shown in Figure 2, the communication path system s in the operational SPO
Suppose that a failure occurs in woo. Processor T
PO detects. FIG. 3 shows the processing of a TPO in which a failure has been detected. When a failure is detected (Fl), the TPO switches to the active system swoo
as the standby system, and switch the standby system 5WO1 to the active system (
F2). Next, an indication (flag) indicating that a failure has been detected is set. (F3).

Next, the TPO issues a notification to the CP that it has detected a failure (F4). At this time, the access mode in TPO is
Leave it as "Write on both systems, read on the active system". Therefore, the same data is written to both 5WOO and 5WO1,
Calls not related to the faulty part will not be interrupted and will be maintained.

The CP receives the notification from the TPO (as shown in Figure 4).
F5), it instructs all TPs to switch the SW and change the access mode to one system (F6).

FIG. 5 shows the processing at the TP upon receiving the switching instruction (Fl) from the CP. TPI and TP2 are inspected with F8, and since the TP is not detecting a fault, immediately upon receiving a switching instruction, the communication path system is changed to SW1, 0-+5W11.5W20→SW2.
Switch to 1- (F9). That is, 5WiO, 5W20
standby system, 5W11. , SW2] is the active system. Next, change the access mode to 5W11. , SW2]- (Fl, 0).

On the other hand, the TPO that has received the instruction from the CP knows that it is the TP at the time of failure detection by looking at the failure detection display flag. If this flag is set, it is known that switching has already been completed, and system switching will not be performed. Regarding the access mode, only one-side switching of swo] is performed.

The difference between TPO and TPI (and TP2) is the difference in communication path system switching processing time (10-odd nanoseconds), the difference in the time required for communication from CP (several milliseconds or less), and the execution time due to the degree of convergence of the processor. Although there is a possibility that there is a slight deviation such as a difference in the number of seconds (several milliseconds or less), access modes 1 to 1 are switched to SPI single system designation all at once. At this point, the faulty channel system swoo
This means that the SPO containing the SPO has been disconnected from the system.

According to this implementation, even if there is no system confounding in the communication path system, by switching the access mode of the communication path system almost simultaneously in all TPs, it is possible to minimize the call interruption time when switching the communication path system. can.

In this embodiment, there is a higher-level processor and communication is used via an inter-processor bus; however, there are other methods (1) to provide the fault detection processor with the functions of a third-level processor; (2) There is also a method of using a common memory instead of inter-processor bus communication. Furthermore, the present invention is not limited to the case where the number of distributed devices is 3.1゜[Effects of the Invention] 18= According to the present invention, communication path system switching can be performed to reduce call interruption time without creating confounding between communication systems. Since this can be implemented, problems such as an increase in the number of selection circuits due to system confounding, changes in the display function of the operating system, and complication of software resource management can be eliminated, and there are effects such as improvement in maintainability and reduction in costs.

[Brief explanation of drawings]

FIG. 1 is an embodiment diagram of the present invention, FIG. 2 is an embodiment diagram including a part of the operation flow, and FIGS. 3 to 5 are processing flow diagrams. cp...Central control processor, TPO-TP2...
Processor for communication path control, SPO... operational communication path system,
SPI: Standby communication path system, swooo~SW22: Communication path system.

Claims (1)

[Claims] 1. A plurality of distributed time-division communication channel systems are configured as a dual system with heat reserve, and one of the dual systems is an active system and the other is a standby system, and all active systems are interconnected by a junctor. , in a communication line system in which all standby systems are interconnected by a junctor, detecting means for detecting the occurrence of a failure in the above-mentioned active system; and upon detection, switching the active system in which the failure has occurred to the standby system; A first switching means for switching a standby system forming a dual system into an active system, and a second switching means for switching a normally operating system of the remaining dual system to a standby system and a standby system to an active system after the switching. and a third switching means for switching the access mode after switching over the entire duplex system.