JP2001036558A - Bus switching method and priority control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology to solve a problem of an exceeded link capacity of a transmission line while keeping a high path accommodation efficiency of an ATM ring network. SOLUTION: On the occurrence of a fault in a VP contained in a ring network, in the case of restoring the fault by VP switching or loopback of an active VP, priority is placed on user cells passing through the active VP and no priority is placed on user cells passing through a standby VP. On the occurrence of an exceeded link capacity, the cells with no priority are sequentially aborted. The transmission of user cells is inhibited for a prescribed time until restoration processing of all paths at a transmission end of a standby VP of a switching path, where the VP selection system changes from the active VP into the standby VP, is finished. The transmission of user cells to a faulty link in a switching device placed at both ends of a faulty link is inhibited for a prescribed time after the restoration of the fault. Or switching processing with respect to the switching path is conducted after the switch back processing of all switch back paths is finished.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a virtual path switching method and a priority control method in an ATM (asynchronous transmission mode) ring network.

[0002]

2. Description of the Related Art The ATM communication system is a connection-type communication based on the transfer of a fixed-length cell (ATM cell) of 53 bytes consisting of a 5-byte cell header containing control information such as a destination and 48-byte user information. In this method, routing and multiplexing are performed based on a virtual path VP and a virtual channel VC.

In an ATM VP transfer network, an arbitrary VP capacity can be realized, so that a highly economical transfer network having excellent path accommodation efficiency can be constructed as compared with an STM (synchronous transmission mode). On the other hand, the network topology of the ring network is shorter in transmission line length than the network topologies such as the full mesh network and the star network, so that the cost of the optical fiber and the repeater is reduced, and the ATM communication system and the ring network are used. The ATM ring network combined with the above network topology is suitable for constructing a transport network having excellent economy.

[0004] The VP switching method at the time of failure in the ATM ring network having the highest path accommodation efficiency is bidirectional path switching /
Path failure detection ring (BPSR / PF) (Okada,
Uematsu, Ota, Tsuboi: 1999 IEICE General Conference B-
8-18 and Okada, Uematsu, Ota, Tsuboi: "A Study on Switching Method in ATM Ring Network" IEICE technical report IN99-7, CS9
9-7, MVE99-7 (see 1999-04). This BPSR / PF is
This is a backup VP switching method in which the path of a backup path is set in advance.
VP switching is performed at the node. Here, for a VP accommodated in a ring network, a device that inserts the VP into the ring network is a path transmitting end, and a device that separates the VP from the ring network is a path receiving end. For this reason, the length of the backup path becomes the shortest, and the path accommodation efficiency becomes the maximum.

FIG. 1 is a diagram showing a BPSR / PF switching method. FIG. 1A shows a normal state, and FIG. 1B shows a state at the time of failure recovery. In the figure, 1 is a working path, 2 is a protection path,
3 is a node and 4 is a fault point. The BPSR / PF is a bidirectional ring, performs switching in units of VP, performs VP switching at an Add / Drop node serving as a path transmission / reception end, switches back to the working path after recovery from a failure, When both systems have failed, select the working path, share the bandwidth between the backup VPs, and increase the path accommodation efficiency.
And so on.

In the BPSR / PF, each link has sufficient spare link capacity to remedy all single failures. FIG. 2 is a diagram showing an example of recovery from a single failure in a five-node ring network in which a duplex path is set to a full mesh. FIG. 2 (a) shows a normal state, and FIG. 2 (b) shows a failure recovery of link L2. Indicates the state of time. The working path indicated by the dotted line in FIG. The required link capacity of the backup path is maximized in the links L4 and L5, and the increase is equal to the link capacity of the working path accommodated in the failed link L2. That is, in the BPSR / PF, if the maximum link capacity of the working path is secured as the maximum link capacity of the backup path in each link in the steady state after the completion of the switching, the link capacity of the transmission line will be exceeded for a single failure. do not do.

Therefore, when this is expressed by an equation, the maximum required link capacity X of the backup path of the link j when the link i fails in the ring including N nodes is as follows: X = Max (i = 1 to N, i ≠ j) [link j after passing link i
Of the working paths that do not pass through the path] = Max (i = 1
NN, i ≠ j) [(link capacity of working path in failed link) − (total of path capacity of working paths passing through links i and j)].

Further, as a feature of the backup band design of the BPSR / PF, in the steady state after the completion of the switching, it is considered that the worst condition under a single failure, that is, the required link capacity of the backup path becomes the maximum under the single failure. it can. This will be described with reference to FIG. FIG. 3 is a diagram illustrating a comparison of path selection systems between a single failure and a multiple failure that occurs independently in time (double failure in the figure). FIG. 3 shows a case where only the working VP is accommodated in the clockwise transmission line. In the BPSR / PF, the principle that “either the VP of the working path or the protection VP of the redundant path is set to each link” and “the working path is selected when both systems fail” are time-dependent. For multiple faults that occur independently, as shown in FIG. 3, there is no new switching from the working VP to the backup VP except for a single fault, and the system selected at the time of the single fault is continuously selected. Or only a fallback from the backup VP to the working VP occurs. That is, at the time of multiple failures, only a part of the path for which the backup VP is selected at the time of a single failure is switched back to the working VP. Therefore, the link capacity used by the backup VP at the time of the multiple failure is smaller than that at the time of the single failure. Become.

[0009] Regarding multiple failures occurring at the same time, "the VP of either the working path or the protection VP of the redundant path is set to each link" and "the working path is selected when both systems fail. , The system to be selected is the same as in the case of multiple faults that occur independently in time. Therefore, in the spare band design of BPSR / PF, a single failure can be considered as the worst case, and the spare band can be designed in accordance with the above-mentioned X. For this reason, unlike other general network topologies, there is an advantage that even when the spare band is shared, it is not necessary to determine the free capacity of the transmission line when multiple failures occur.

The shared spare band design method for sharing the spare band has a smaller link capacity required for the spare VP than the dedicated spare band design method for exclusively allocating the spare VP band to the working VP. The storage efficiency increases. FIG. 4 is a diagram illustrating this fact. In a link network in which a duplex path having a path capacity of 1 Mb / s is connected in a full mesh, the maximum required backup link capacity of the backup VP between the dedicated backup bandwidth design method and the shared backup bandwidth design method is shown. It is a figure which shows the result of having compared.

As a BPSR / PF switching method, ITU-
It is possible to apply the T standard VP switching method. In the following, when applying the ITU-T standard VP switching method in which switching is performed by a one-phase protocol, a problematic BP
An instantaneous overflow of the link capacity of the transmission path in the SR / PF will be described.

Although the occurrence of multiple failures in the steady state of switching does not cause an overflow of the link capacity of the transmission line, the Wait-t described in ITU-T Recommendation I.630
In a protection state such as o-Restore (switching transient state within the failback protection time), if a new failure occurs on a link other than the first failure link, an unrelated link will momentarily link the transmission path. Overcapacity may occur.
In this case, a cell loss also occurs in a working VP of a duplicated path or a non-duplicated path that does not pass through a failed link and is irrelevant to a failure.

[0013] The Wait-to-Restore state is defined in ITU-T Recommendation I.630 as "a state in which a backup VP is selected until switching back to the working VP in the normal state of both systems after the cause of the failure is resolved." It is stipulated that the recommendation
It can be set at intervals of 1 minute for 30 minutes, and the standard setting is specified as 12 minutes. This Wait-to-Re
In the store state, a single failure of the working VP and a spare VP
When a request with a higher priority than Wait-to-Restore, such as a single failure, is received, Wait-to-Restore is released and switching is performed according to the request with higher priority. Over occurs.

FIG. 5 shows a failure pattern in which the link capacity of the transmission line actually exceeds and the working or backup V associated with the failure pattern.
The transition states of the selection system of P are classified into the following states according to the passage of time.
It is a figure distinguished and shown to (1)-state (4). State (1): Normal state State (2): First failure recovery and failback protection state State (3): Switching transient state in which switching of the failure detection station is completed due to second failure State (4): No. State where the failure of 2 has been recovered

The assumed ring network is one in which the number of ring nodes is 5, the transmission line link capacity is 2.4 Gb / s, and the 400 Mb / s duplex path is set to a full mesh. Regarding the failure pattern, as an example, the first failure is a link failure in the link L2 of the counterclockwise transmission line, and the second failure is the link L1 in the failback protection state after the first failure cause is resolved. It is assumed that a link failure occurs in the counterclockwise transmission line of the above.

FIG. 5 shows that in the switching transient state (3), the failure of the backup VP of the counter-clockwise transmission line is 2.8 Gb / s on the link L3, 3.2 Gb / s on the link L4, and 2.8 Gb / s on the link L5. It can be seen that instantaneous link capacity excess of the transmission line occurs in a link unrelated to the link. As described above, a cell loss occurs in the traffic of the working VP which is irrelevant to the failure that does not pass through the links L3, L4 and L5.

FIG. 6 is a diagram showing a transition state of the selection system regarding the working or backup VP in the switching device in the failure pattern of FIG. For a double link failure, the paths in the ring network are classified into the following four types. Path classification 1: working VP passes through L1 and L2: backup VP
None Classification 2: Working VP passes through L1: Backup VP passes through L2 Path classification 3: Working VP passes through L2: Backup VP passes through L1 Path classification 4: No working VP: Backup VPs L1 and L2
Go through

FIG. 6 shows that the fail-back protection state (2) after the failure cause of L2 is resolved, and the failure recovery state after the link failure of L1.
At the time of transition to (4), it is understood that the selection system of the working or backup VP in the switching device changes for the paths of the path classification 2 and the path classification 3. The reason that the link capacity of the transmission line is exceeded is that the switching between the failure detection station and the opposite station is performed in the state (3) where the failure detection station completes the switching for the second failure.
This is because the selection system of P momentarily becomes inconsistent with each of the paths of path classification 2 and path classification 3, and traffic concentrates on the backup VP.

However, normally, the frequency of occurrence of the above-mentioned failure patterns is very low, and the path accommodating efficiency is low in the backup band design which secures the link capacity of the backup VP corresponding to such failure patterns. , Resulting in reduced economic efficiency. FIG. 7 shows a dedicated spare band design method (a), a shared spare band design method (b) corresponding to an excess of the link capacity of the transmission line, and a link of the transmission line when the duplex path is set to a full mesh. FIG. 13 is a diagram showing a comparison of the maximum required link capacity in the spare band design by the shared spare band design method (c) that does not cope with excess capacity.
From FIG. 7, it can be seen that the maximum required link capacity of the backup VP greatly differs depending on the backup bandwidth design method.

[0020]

SUMMARY OF THE INVENTION The present invention provides a technique capable of solving the above-mentioned problem of excess transmission line link capacity while maintaining high path accommodation efficiency by a spare band design method in which a single failure is the worst condition. The purpose is to do.

[0021]

In order to achieve the above object, a path switching method according to the present invention provides a device for inserting a VP into a ring network and a device for separating the VP from the failure of a VP accommodated in the ring network. In the path switching method of the ATM ring network for restoring the failure by the VP switching in the above or by the return of the working VP in the equipment at both ends or the upstream end of the failed link, the user cell passing through the working VP is given priority and the backup VP is passed. In this case, the priority is given to non-priority cells and the priority is given to traffic when the link capacity is exceeded.

According to another path switching method of the present invention, in response to a failure of a VP accommodated in a ring network, VP switching is performed in a device for inserting the VP into the ring network and a device for separating the VP.
Alternatively, in the path switching method of the ATM ring network for recovering the failure by returning the working VP at the equipment at both ends or the upstream end of the failed link, the VP is switched from the working VP to the backup VP.
The transmission end of the backup VP of the switching path, in which the selection system changes, prohibits the transmission of the user cell for a certain time until the restoration processing of all the paths is completed.

Still another path switching method according to the present invention is such that, in response to a failure of a VP accommodated in a ring network, a VP is switched in a device for inserting the VP into the ring network and a device for separating the VP, or at both ends or upstream of the failed link. In the path switching method of the ATM ring network for restoring the failure by returning the working VP at the end device, transmission of the user cell to the failed link in the switching device located at both ends of the failed link is prohibited for a certain period after the restoration. It is characterized by doing.

Still another path switching method according to the present invention is such that, in response to a failure of a VP accommodated in a ring network, a VP is switched in a device for inserting the VP into the ring network and a device for separating the VP, or at both ends or upstream of the failed link. In the path switching method of the ATM ring network for restoring the failure by returning the working VP in the terminal device, the switching process for the switching path is performed after the switching process of all the switching paths is completed.

In the priority control method according to the present invention, in a network topology sharing a protection band, a user cell passing through the working VP is given priority and a user cell passing through the protection VP is given no priority. When the link capacity of the active VP exceeds the link capacity unrelated to the failure, the cells are sequentially discarded from the non-priority cells, and priority control of the traffic is performed.

According to another priority control method of the present invention, when a working path of a high-reliability path and a link capacity of a backup VP are exceeded in a network topology sharing a protection band, the priority cells are sequentially discarded from non-priority cells to prioritize traffic. The control is performed.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, first, priority control of traffic is performed (feature 1), and second, a VP selection system is switched at a path transmitting / receiving end of a backup VP which changes from a working VP to a backup VP. Third, the user cell is not transmitted to the failed link for a certain period of time after the restoration of the failure in the switching devices located at both ends of the failed link (Feature 3). Fourth, the failure recovery is performed in the order of switching back and forth (characteristic 4). Examples of each feature will be described below.

[Embodiment 1] An embodiment to which the method of performing the traffic priority control of the feature 1 is applied will be described. The overflow of the link capacity of the transmission line is caused by the traffic being concentrated on the backup VP in the switching transient state and exceeding the link capacity. Therefore, in the present invention, priority control is performed such that the user cell of the working VP is prioritized and the user cell of the backup VP is not prioritized, so that the user cell of the backup VP has an influence on the user cell of the working VP. Do not affect.

When priority control of traffic is performed by the method of the present invention, even if the link capacity of the transmission line exceeds, the user cells of the backup VP are sequentially discarded, so that the traffic of the working VP is not affected. . Any priority control may be applied as long as the priority control is such that the priority cell does not affect the priority cell at all, or the priority control performs weighting within a range where the priority cell does not affect the priority cell. . The simplest priority processing generally known is strict priority scheduling which outputs non-priority traffic input cells only when there is no priority traffic input cell in the buffer.

As a method for identifying a priority VP and a non-priority VP in each switching device in the ring, there are a tagging method and a priority setting method. The tagging method is a method of performing control based on a CLP field indicating priority or non-priority of a cell in a cell header. For example, at the Add node serving as the transmission end of the path, “0” indicating priority is set in the CLP field of the cell header of the user cell passing through the working VP, and the CLP of the cell header of the user cell passing through the backup VP.
"1" indicating non-priority is set in the field. At the time of congestion in which an overflow of the ring capacity of the transmission line occurs, each switching device refers to the contents of the CLP field and sequentially discards the non-priority cells, thereby guaranteeing the quality of the user cells passing through the working VP.

The priority setting method is based on the V through which an ATM cell passes.
This is a method of assigning priority or non-priority to PI. In this method, priority or non-priority is assigned to the passing VPI, and when congestion occurs when the transmission line link capacity is exceeded, each switching device refers to the contents of the VPI field of the cell header and sequentially starts from cells passing through the non-priority VPI. Perform disposal.

When priority control of the traffic of the feature 1 is performed for the problematic failure pattern (state (3)) shown in FIG. 5, the links L3 and L4 in which the link capacity of the transmission line is momentarily exceeded. Priority control is performed by the node N3 on cells passing through the L2 and L5 to prevent the traffic passing through the working VP from being affected.

FIG. 8 is a diagram showing the transition of the used link capacity in the transmission line in this embodiment to which the feature 1 is applied for the link L4 of the counterclockwise transmission line in the state (3) of FIG. FIG. 8A shows a state before the method of this embodiment is applied, and FIG. 8B shows a state after the method of this embodiment is applied. In the traffic of the backup VP passing through the link L4, since the cells of the backup VP are preferentially discarded by the priority control, a cell loss occurs in the backup VP, but the traffic passing through the working VP is not affected. . However, in this case, there is a possibility that a cell loss may occur in a backup VP between the nodes N1 and N3, which is not directly related to the link failure in the second L1, that is, a path classified into the path classification 1. However, this is not a problem when the traffic passing through the backup VP is small.

[Embodiment 2] The VP selection system of the feature 2 is an active V
An embodiment will be described in which a method is used in which a user cell is not transmitted for a fixed time until switching is completed at a path transmitting / receiving end of a backup VP that changes from P to a backup VP. As shown in FIG. 6, the excess of the link capacity of the transmission line is caused by the mismatch between the active and standby VP selection systems between the failure detecting station and the opposite station in the switching transient state (state (3)).
Since the link capacity over occurs only in the backup VP,
In the switching path from the working VP to the backup VP excluding the switchback path, by not transmitting user cells until the selected system matches, it is possible to prevent the link capacity of the transmission path from being exceeded. The corresponding path is a path classified into the path classification 2 in FIG. In this case, the path types that prohibit the transmission of the user cells may be a switching path (path classification 2) and a reverting path (path classification 3).

The trigger for prohibiting the transmission of the user cell is:
For example, a failure alarm signal such as a path AIS or the like, and a trigger for resuming transmission of a user cell may include, for example, a switching transition state until the selection system between the failure detection station and the opposite station changes from non-coincidence to coincidence. Set the waiting time,
A method can be considered in which the time when the waiting time expires is used as a trigger.

The prohibition of the transmission of the user cell in the device is performed, for example, at the low-speed input interface in the switching device of the Add node serving as the path transmitting end, or at the high-speed input in the switching device of the Add node serving as the path transmitting end. In the interface can be done.

FIG. 9 shows a change in traffic when the method of prohibiting the transmission of user cells for a certain period of feature 2 is applied to the problematic failure pattern (state (3)) shown in FIG. FIG. In FIG. 9, the state (3 ′) indicates the path classification 2
When the transmission end of the switching path of the user has set the prohibition of the transmission of the user cell, the state after the completion of the switching at the link L1 failure detecting station is referred to as a state (3 ").
This shows a state in which the user cell transmission prohibition of P is set. In each figure, a black circle indicates that transmission of user cells is prohibited. In any state, it can be understood that the link capacity of the backup VP is avoided in the links L3, L4, and L5 where the link capacity of the transmission line has momentarily exceeded, and the link capacity of the transmission path does not occur.

FIG. 10 is a diagram showing the transition of the used link capacity in the transmission line in the embodiment to which the feature 2 is applied for the link L4 of the counterclockwise transmission line in FIG.
(a) is a state before applying the method of this embodiment, FIG.
Indicates a state after the method of this embodiment is applied. In order to prohibit the traffic of the backup VP causing the overflow from being transmitted for a time sufficiently including the switching transient state, the working VP and the backup VP irrespective of the failure of the link L4.
It can be seen that the link capacity is not exceeded in each case. However, even if the switching is completed, traffic cannot be transmitted to the switched path until the timer expires, so that communication on the switched path is not established during that time.

[Embodiment 3] An embodiment will be described in which a switching device located at both ends of a failed link according to feature 3 employs a method in which a user cell is not transmitted to a failed link for a certain period of time after restoration from a failure. If the two transmission lines are divided at two different links of the ring network, the ring network is completely divided into two segments. In this case, one segment is not affected by the traffic of the other segment. In addition, since the link capacity over occurs only in the link capacity of the backup path, if the transmission of the user cell of the backup VP of the switching path in the failed link is prohibited, the link capacity excess can be avoided. FIG. 11 is a diagram showing a state in which the transmission of the user cell of the backup VP to the failed link is prohibited. FIG.
11 (b) shows the state at the time of failure. It can be seen that the transmission prohibition of the user cells other than the working VP is prohibited only by the backup VP, so that the excess of the link capacity of the transmission path is avoided.

Regarding the time during which the transmission of the user cell on the backup path is prohibited, the BPSR /
In the PF, since the link capacity of the transmission line does not overflow, there is no problem as long as the time sufficiently includes the transient state of switching.

The simplest transmission prohibition setting is to set all backup Vs on the failed link for a sufficient time from the detection of the failure until the failback protection state expires and the completion of the failback processing.
This is a method of prohibiting the transmission of the P user cell. In this case, the setting of the transmission prohibition can be performed only by the section (the switching devices located at both ends of the failed link). However,
In this case, since the transmission is prohibited until the timer of the failback protection state of the first failed link L1 expires, the path of the path classification 2 in which the switching is caused by the second failure is a timer of the failback protection state. No two-way communication is established until the expiration.

FIG. 12 shows a case where the method of not transmitting a user cell to a failed link for a certain period of time after the failure recovery of the feature 3 is applied to the failure pattern (state (3)) which causes the problem shown in FIG. It is a figure which shows the change of the traffic of FIG. In FIG.
State (3 ') is a state in which transmission of user cells to the failed link is prohibited after the switching of the L1 failure detecting station is completed.
(3 ") indicates a state in which transmission of user cells to the failed link after L1 failure recovery is prohibited. In each figure, a black circle indicates that transmission of user cells is prohibited. It can be seen that in the links L3, L4 and L5 in which the link capacity of the transmission line is excessively exceeded, the link capacity of the backup VP is avoided and the link capacity of the transmission path is not exceeded.

FIG. 13 is a diagram showing the transition of the used link capacity in the transmission line in the embodiment to which the feature 3 is applied for the link L4 of the counterclockwise transmission line in FIG.
(a) is a state before applying the method of this embodiment, FIG.
Indicates a state after the method of this embodiment is applied. Since connection of only the backup VP is prohibited, it can be seen that there is no influence of instantaneous interruption, cell loss, etc. on the traffic of the working VP which is unrelated to the failure. However, Wait-to-Restor
e In the case of prohibiting the pass-through connection of the backup VP until the state expires, the path communicating with the backup VP passing through the link for which the passage connection is prohibited is not established until the Wait-to-Restore state expires.

The method of not transmitting a user cell to a failed link for a certain period of time after recovery from a failure can be applied not only to a backup VP but also to a user cell of an active VP. However, in this case, bidirectional communication will not be established even for the failback path of the path classification 3 until the timer of the failback protection state expires.

[Embodiment 4] An embodiment will be described in which a method of recovering from a failure is applied in the order of switching after switching back to feature 4. The link capacity over of the transmission line occurs due to the link capacity over of the backup VP in the switching transient state. Here, considering the used link capacity for switching and switching back, for switching, the used link capacity of the working VP is reduced and the used link capacity of the backup VP is increased. For switching back, the current VP
, The link capacity of the backup VP is reduced. On the other hand, in BPSR / PF, a spare band is designed so that the link capacity of the transmission line does not exceed in the steady state of switching.

Therefore, as for the order of the recovery processing, switchback (failure recovery processing due to the failure of the backup VP) is performed immediately after failure detection (order 1), and switching (failure recovery processing due to the failure of the working VP) is switched back. The order is to provide a certain waiting time that can sufficiently cover until completion (order 2). As a result, the switching can be performed in a state where the used link capacity of the backup VP is reduced after the completion of the failback, and the link capacity of the backup VP does not exceed.

FIG. 14 shows the change in traffic when the method of performing the fault recovery in the order of switching back and forth of the feature 4 is applied to the fault pattern (state (3)) which causes the problem shown in FIG. FIG. By performing the switchback process first, the link capacity of the backup VP is reduced in the links L3, L4, and L5 where the link capacity of the transmission line has momentarily exceeded. Next, by performing the switching process, the recovery process is performed without causing the link capacity of the backup VP to be exceeded, and it can be seen that the link capacity of the transmission path is not exceeded.

FIG. 15 is a diagram showing the transition of the used link capacity in the transmission line in the embodiment to which the feature 4 is applied for the link L4 of the counterclockwise transmission line in FIG.
(a) is a state before applying the method of this embodiment, FIG.
Indicates a state after the method of this embodiment is applied. Since the switching is performed after the failback, it can be seen that the link capacity is not exceeded for the traffic of the working VP and the backup VP, which is unrelated to the failure. However, in this case, since the process of switching back and the process of switching are divided, the instantaneous interruption to the switching path is longer than in the case where the normal recovery process is performed.

[0049]

As described above, the method according to the present invention comprises:
There are various effects as described below.

(1) If it is a VP switching system that satisfies the six characteristics of BPSR / PF, it can be applied to all VP switching systems. For example, an ITU standard VP switching method using the one-phase switching protocol described above,
Compared to the VP switching method using a switching protocol of two phases, three phases, etc., the method of the present invention
Since only the switching order in the switching device of the failure detecting station and the opposite station is different, the present invention can be applied essentially without distinction.

(2) The method of the present invention is characterized by the above features 1 to 4. These features can be applied to ATM ring network switching systems other than BPSR / PF that perform switching at the failed link end. FIG. 16 shows a BPSR / LF (Line failure detection) that performs a return at both ends of a failed link and a BPSR / HF (Hybrid failured) that performs a return at an upstream end.
FIG. 7 is a diagram showing the result of the comparison with BPSR / PF. / LF is the VP that performs loopback at both ends of the failed link
Switching, and / HF is VP switching for performing loopback at the upstream end of the failed link. In / LF and / HF, the failure detection is performed in units of sections, that is, only at nodes located at both ends of the failure link. For this reason, the feature 3 that the processing can be performed in section units: “Do not send user cells to the failed link for a certain period of time after the failure recovery at the switching devices located at both ends of the failed link” is applied to these switching methods. It is easier to implement than BPSR / PF.

(3) In a general network topology, it is difficult to design a shared spare band for guaranteeing the quality of traffic passing through a working VP at the time of multiple failures, a spare VP for highly reliable service, and the like due to its complexity. . In such a case,
The priority control of the traffic in the feature 1 is extremely effective as a means for simplifying the design of the backup band.

FIG. 17 shows that the active VP is used on the transmission line in which the link capacity is exceeded at the time of multiple failures by applying the feature 1.
FIG. 17 (a) is a diagram showing a method for guaranteeing the traffic quality of the conventional method compared with the conventional method, and FIG. 17 (b) shows the conventional method.
Shows after applying the method of the present invention. By the method using the priority control of the traffic of the feature 1, since the traffic of the working path is prioritized, the quality of the traffic of the working VP is guaranteed without being affected by cell loss or the like even if the link capacity of the protection VP traffic is exceeded. Is done.

FIG. 18 is a diagram showing a method for guaranteeing the quality of backup VP traffic of a highly reliable service class at the time of multiple failures to which the feature 1 is applied, in comparison with a conventional method. a) shows the conventional method, and FIG. 18 (b) shows the state after applying the method of the present invention. In FIG. 18, it can be seen that by setting the priority of the high-reliability backup VP traffic higher than the low-reliability backup VP, the quality of traffic passing through the high-reliability service class backup VP is guaranteed. However, (a) high reliability working VP, (b) low reliability working VP,
(c) High-reliability backup VP, (d) Regarding the combination of priorities of low-reliability backup VPs, (a) = (b) = (c) → (d), (a)
→ (b) → (c) → (d), etc., can be set freely according to the judgment of the operator.

[Brief description of the drawings]

FIG. 1 is a diagram showing a BPSR / PF switching method.

FIG. 2 is a diagram illustrating an example of recovery from a single failure.

FIG. 3 is a diagram showing a comparison of a path selection system between a single failure and multiple failures.

FIG. 4 is a diagram showing a comparison of the maximum required spare link capacity of the spare VP between the dedicated spare band design scheme and the shared spare band design scheme.

FIG. 5 is a diagram showing a failure pattern in which a link capacity of a transmission line is actually exceeded and a transition state of a VP selection system accompanying the failure pattern.

FIG. 6 shows V in the switching device in the failure pattern of FIG.
It is a figure which shows the transition state of the selection system of P.

FIG. 7 is a diagram showing a comparison of the maximum required link capacity of the backup VP according to the backup bandwidth design method of BPSR / PF.

FIG. 8 is a diagram illustrating a transition of a used link capacity in a transmission line according to the first embodiment.

9 is a diagram illustrating a change in traffic when the method of the second embodiment is applied to the problematic failure pattern illustrated in FIG. 5;

FIG. 10 is a diagram illustrating a transition of a used link capacity in a transmission line according to the second embodiment.

FIG. 11 is a diagram illustrating a state in which transmission of a user cell of a backup VP to a failed link is prohibited.

12 is a diagram illustrating a change in traffic when the method according to the third embodiment is applied to the problematic failure pattern illustrated in FIG. 5;

FIG. 13 is a diagram illustrating a transition of a used link capacity in a transmission line according to a third embodiment.

FIG. 14 is a diagram showing a change in traffic when the method of the fourth embodiment is applied to the problematic failure pattern shown in FIG. 5;

FIG. 15 is a diagram illustrating a transition of a used link capacity in a transmission line according to a fourth embodiment.

FIG. 16 shows BPSR / LF and BPSR / HF as B
It is a figure shown in comparison with PSR / PF.

FIG. 17 is a diagram illustrating an example of application of the method according to the first embodiment.

FIG. 18 is a diagram illustrating another example of application of the method of the first embodiment.

[Explanation of symbols]

 1 working path 2 backup path 3 node 4 fault point L1, L2, L3, L4, L5 link N1, N2, N3, N4, N5 node

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Hiroshi Ota 2-3-1 Otemachi, Chiyoda-ku, Tokyo Inside Nippon Telegraph and Telephone Corporation (72) Inventor Toshinori Tsuboi 2-3-3 Otemachi, Chiyoda-ku, Tokyo No. 1 Nippon Telegraph and Telephone Corporation F term (reference) 5K030 GA12 HA10 HB17 HC15 LA03 LB08 LC18 LE05 MD02 5K031 AA08 CB01 EB05 5K042 AA01 CA05 CA13 DA35 FA01 9A001 BB04 CC07 DD10 LL02

Claims (6)

[Claims] 1. In response to a failure of a VP accommodated in a ring network, a VP is switched in a device that inserts the VP into the ring network and a device that separates the VP, or by turning back a working VP in a device at both ends or an upstream end of the failed link. In a path switching method of an ATM ring network for restoring a failure,
Priority is given to user cells passing through the working VP, and backup V
A path switching method, wherein a user cell passing through P is given non-priority, and when link capacity is exceeded, the cells are sequentially discarded from the non-priority cell and traffic priority control is performed. 2. In response to a failure of a VP housed in a ring network, a VP is switched in a device for inserting the VP into the ring network and a device for separating the VP, or by folding back a working VP in a device at both ends or an upstream end of the failed link. In a path switching method of an ATM ring network for restoring a failure,
A path switching method, wherein transmission of a user cell is prohibited for a certain period of time at a transmission end of a backup VP of a switching path in which a VP selection system changes from an active VP to a backup VP until all paths are restored. 3. In response to a failure of a VP housed in a ring network, a VP is switched in a device that inserts the VP into the ring network and a device that separates the VP, or by folding back a working VP in a device at both ends or an upstream end of the failed link. In a path switching method of an ATM ring network for restoring a failure,
A path switching method, wherein transmission of a user cell to a failed link in a switching device located at both ends of the failed link is prohibited for a certain period of time after restoration of the failure. 4. In response to a failure of a VP accommodated in a ring network, a VP is switched in a device that inserts the VP into the ring network and a device that separates the VP, or by folding back a working VP in a device at both ends or an upstream end of the failed link. In a path switching method of an ATM ring network for restoring a failure,
A path switching method, wherein a switching process is performed on a switching path after the switching process of all the switching paths is completed. 5. A system in which a user cell passing through a working VP in a network topology sharing a protection band is given priority and a user cell passing through a protection VP is given no priority. A priority control method characterized by sequentially discarding non-priority cells when a link capacity over occurs, and performing traffic priority control. 6. When a working path of a reliable path and a link capacity of a protection VP exceed a link capacity in a network topology sharing a protection band, non-priority cells are sequentially discarded, and priority control of traffic is performed. Priority control method.