KR20120119783A - Protection switching method in packet ring network - Google Patents

Abstract

PURPOSE: A protection changeover method of a node in a network is provided to improve protection changeover function by optimizing a flush procedure which consumes a lot of time on protection changeover in a network. CONSTITUTION: The generation of a local event is sensed(S200). The block setup or block release of ports is executed according to the local event(S210). A protection changeover message is transmitted(S220). A flush command of a filtering database is generated for a blocked port according to the local event(S230). The flush command of the filtering database is generated for the other ports(S250). [Reference numerals] (AA) Start; (BB) End; (S200) Local event generation sensing; (S210) The block of a port is released or set according to a local event; (S220) A protection switchover message is transmitted; (S230) FDB flush command generation for the blocked port; (S240) Traffic flooding; (S250) FDB flush command generation for ports which exclude the blocked port

Description

Protection switching method in packet ring network

The present invention relates to a method of protection switching of a node in a network, and more particularly, to a method of improving switching performance by optimizing a flush process.

Referring to the ring network among the various networks to which the present invention is applicable, the ring network prevents the formation of a ring by deactivating one link or port included in the ring, in order to prevent an infinite loop of traffic, and then general Ethernet MAC. (Media Access Control) Prevents infinite loop propagation by writing FDB of each node using address learning method. Since the block link is logically set up but the physical link remains intact, traffic can be delivered immediately by removing the logical block.

If a failure occurs in the ring network where the logical block link exists, the failed link provides the physical or logical block, so that after the block of the logical block link is removed, the switching is performed so that traffic can pass through the link where the block is removed. Happens.

When a physical block link due to a failure and a block link for preventing a loop are newly selected, a forwarding table, a filtering database (FDB), is no longer valid, and all nodes must newly create an FDB. In other words, FDB is initialized to initialize each port of FDB and new FDB is created by learning new address.

An object of the present invention is to provide a method of improving the protection switching performance by optimizing the flushing process that takes a lot of time during the protection switching in the network.

In order to achieve the above technical problem, an example of a method of protection switching of a node in a network according to the present invention is a method of switching a node adjacent to a link in which a local event causing a switching in a network occurs according to the local event. Generating a flush command of the filtering database for the blocked port; And generating a flush command of the filtering database for ports other than the blocked port.

Another example of a method of protection switching of a node in a network according to the present invention for achieving the above technical problem is a protection switching message in a method of switching protection when a node including a block port in a network is received. When the block of the block port is released due to the reception of a signal, generating a flush command of the filtering database for a port in a direction opposite to the port where the block is released; And generating a flush command of the filtering database for the port from which the block is released.

In order to achieve the above technical problem, another example of a method of protection switching of a node in a network according to the present invention is a switching method in a node that receives a protection switching message in a network, wherein the port receiving the protection switching message is received. Generating a flush command for the filtering database; And generating a flush command of the filtering database for ports other than the port for receiving the protection switching message.

According to the present invention, it is possible to quickly recover traffic by first flushing a port necessary for traffic transmission during protection switching.

1 is a view showing the structure of an example of a ring network to which the present invention is applied,
2 is a flowchart illustrating an example of a method of switching at a node adjacent to a link where a local event has occurred, according to the present invention;
3 is a flowchart illustrating an example of a method of switching at a node having a block link among switching methods at a node receiving a protection switching message according to the present invention;
4 is a flowchart illustrating an example of a switching method in a general node having no block link among switching methods at a node receiving a protection switching message according to the present invention.

Hereinafter, a method of protection switching of a node in a network according to the present invention will be described in detail with reference to the accompanying drawings.

Prior to examining the embodiments, the present invention is applicable to the case where new address learning is required for the changed topology during protection switching and restoration of the network, and the ring network is only one embodiment to which the present invention is applied. It doesn't happen. However, for convenience of description, the case of the ring network will be described below.

1 is a diagram illustrating an example of a structure of a ring network to which the present invention is applied.

Referring to FIG. 1, a ring network has a structure in which each node is connected in a ring shape, and includes a block link between node 1 100 and node 4 130 to prevent an infinite loop. Here, Node 1 (100) is a Ring Protection Link (RPL) Owner, and Node 4 (130) is an RLP Neighbor. In some embodiments, the RLP neighbor may be omitted. In addition, the ring network to which the present invention is applied may have various forms such as a single ring structure as shown in FIG. 1, an interconnected multiple ring, and an overlapping ring structure using a plurality of protection switching schemes. This is possible.

Each node can physically contain more than one port, and the ports are divided into two types, the West port and the East port. For example, the East port and the West port of each node may each be one physical port, or the East port or the West port may include two or more ports according to embodiments. Hereinafter, for convenience of description, it is assumed that the east port and the west port of each node are one physical port.

Each node includes a filtering database (FDB) for traffic transmission by port. For example, node 1 (100) has an FDB for the East port because the West port is blocked, node 4 (130) has an FDB for the West port because the East port is blocked, and nodes 2, 3 (110, 120). ) Has separate FDBs for the East and West ports.

In a ring network, a switchover occurs when a local event such as a link failure, failover, or forced switchover by a user occurs. For convenience of explanation, it is assumed that a local event occurs between node 1 and node 2 and the protection switching method thereof will be described in detail.

2 is a flowchart illustrating an example of a switching method in a node adjacent to a link where a local event occurs.

Referring to FIG. 2, a node adjacent to a link having a local event detects a local event (eg, a link failure, forced switchover, link recovery, etc.) that causes a switchover (S200). Detection of local events can be detected through Continuous Check Message (CCM) or link detection, including protection recovery.

The node detecting the local event performs block setting or block release of ports according to the generated local event (S210). For example, in the ring network of FIG. 1, between a node 1 (100) and a node 2 (110), a signal failure (SF), manual switching (MS), or forced switching (FS) is performed. If a local event of RLP Owner occurs, Node 1 (100), the RLP Owner, detects the occurrence of the local event, unblocks the logical block port (that is, the West port), and the port of the link where the local event occurred (that is, the East port). Set the block of. On the other hand, since node 2 110 does not have a separate logical block port, only node 2 110 sets a block of a port where a local event occurs. As another example, when a local event of failover between node 1 100 and node 2 110 occurs, node 1 100 establishes a block of ports (ie, West ports) between node 4 130. And unblock the port (ie, East port) between node 1 (100). In this way, the node adjacent to the link where the local event occurs sets or releases a block of a specific port according to the type of local event and its state.

When the block is set up or released, the node adjacent to the link where the local event occurred multicasts a protection transfer message (eg, R-APS (SF / MS / FS)) to all other nodes in the ring (S220). ).

Next, the node generates an FDB flush command for a port blocked due to the occurrence of a local event before the FDB flush command of another port (S230). When the FDB flush command for the blocked port is generated, the node performs flooding outputting through all ports for traffic output to the blocked port (S240). The flooding is not performed after the FDB flush on the blocked port is completed, but at the moment when the FDB flush is started on the blocked port, that is, at the moment the FDB flush command occurs. When the FDB flush for the blocked port is completed, the node generates an FDB flush command for ports other than the blocked port (S250).

For example, in FIG. 1, when a local event such as a failure between node 1 100 and node 2 110 occurs, the traffic output from the node 1 to the link where the local event occurs is flooded when the East port is flushed. If the first port flush of the node 1 (100) is performed first, the traffic recovery time is delayed because the east port is flushed after a delay according to the flush of the west port. However, in the present embodiment, the first port of the node 1 (100) is delayed. Because the East port is flushed first, traffic can be recovered immediately without delays caused by the West port flushing.

3 is a flowchart illustrating an example of a method of switching at a node having a block link among switching methods at a node receiving a protection switching message according to the present invention.

Referring to FIG. 3, when a node receives a protection transfer message (S300), the node releases a block of a previously port (S410). The node having a block port may be an RPL neighbor such as node 4 130 of FIG. 1, or may be a general node. In addition, when a node receives a protection transfer message, the node does not have to release the block of the existing block port, and the block may not be released depending on the type of the protection transfer message received or the situation of the node, but the block is not released. If the case is outside the scope of the present invention, a detailed description thereof will be omitted.

The node generates an FDB flush command for the port in the opposite direction of the port in which the block is released (S320), and floods the traffic (S330). When the FDB flush is completed for the port in the opposite direction to the port in which the block is released, the node generates an FDB flush command for the port in which the block is released (S440).

For example, if a node with a blocked port is an RPL neighbor such as node 4 130 of FIG. 1, when the node receives the first protection transfer message, it discards it without sending it to the next node, and then the same protection transfer. If a message is received back on the same port, it is sent to the next node.

4 is a flowchart illustrating an example of a switching method in a general node having no block link among switching methods at a node receiving a protection switching message according to the present invention.

Referring to FIG. 4, when a node receives a protection transfer message (S400), the node first issues an FDB flush command for the port that received the message (S410), and then flushes an FDB of a port in the opposite direction in which the protection transfer message is received. Generate a command (S420). Therefore, the traffic output to the port that received the protection beacon message is flooded first to recover the traffic. The node then transmits the received protection transfer message to the next node.

For example, in FIG. 1, when node 3 120 receives a protection transfer message transmitted from node 2 110 through the west port, first flush the FDB on the west port and then flush the FDB on the east port. If the protection transfer message transmitted from the node 1 (100) is received through the East port, the FDB of the East port is first flushed, and then the FDB of the East port is flushed. Therefore, traffic for the first flushed port is recovered first.

As another example, if a local event occurs between node 1 (100) and node 2 (110) when there is traffic between node 1 (100) and node 3 (120), traffic directed to node 1 (100). Must be switched to transmit via node 4 (130). In this case, Node 3 (120) must flush the West port before the East port, so that traffic can be quickly recovered by transmitting traffic to Node 1 (100) directly through Node 4 (130) without the delay caused by the East port flush. It is possible. However, in some cases, node 3 (120) may receive the protection transfer message transmitted from node 1 (100) prior to the protection transfer message transmitted from node 2 (110). Since the East port is flushed first, Node 3 has to wait for the West port to flush after the East port flush, resulting in traffic recovery delays.

This can be solved by placing an RLP neighbor in the ring network. When the node 4 (130), which is an RPL neighbor, receives the protection switching message from the node 1 (100), performs the protection switching according to the flow shown in FIG. 4, and then discards the protection switching message without transmitting to the next node. According to an embodiment, nodes 1 (100) and 2 (110) adjacent to a link having a local event may send the same protection transfer message several times (for example, three times) at a predetermined interval (for example, 3.3 ms). In this case, Node 4 (130), which is an RPL neighbor, discards the first protection transfer message and then sends the protection transfer message to the next node. Therefore, node 3 (120) transmits the protection transfer message transmitted from node 2 (110). You can receive it first. When the node 3 120 receives the same protection transfer message through the same port, the node 3 120 discards the same protection transfer message and does not transmit it to the next node.

The present invention can also be embodied as computer-readable codes on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

So far I looked at the center of the preferred embodiment for the present invention. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

Claims (11)

In a method of switching a node adjacent to a link having a local event causing a switchover in a network,
Generating a flush command of a filtering database for a port blocked according to the local event; And
Generating a flush command of the filtering database for ports other than the blocked port. The method of claim 1,
Detecting the occurrence of the local event;
Performing block setting or block unblocking of ports according to the local event; And
Transmitting a protection switching message; before the first step. The method of claim 1,
And flooding the traffic output to the blocked port when the flush command of the first step occurs. The method of claim 1,
If a node adjacent to the link where the local event occurred includes a logical block port for preventing the formation of a loop, upon detection of a local event of a failure, the block of the logical block port is released and a block of the failed port is detected. Step of switching node in the network comprising a. In the method of switching when a protection transfer message is received from a node including a block port in a network,
When the block of the block port is released due to the reception of a protection transfer message, generating a flush command of the filtering database for the port in a direction opposite to the port where the block is released; And
And generating a flush command of the filtering database for the port from which the block is released. 6. The method of claim 5,
Receiving a protection transfer message; And
Releasing the block of the block port; before the first step. 6. The method of claim 5,
And flooding the traffic output to the port in a direction opposite to the port from which the block is released when the flush command of the first step occurs. 6. The method of claim 5,
In case that the node is a ring protection link neighbor (RPL neighbor), if the same protection transfer message is received at least two times through the same port, the first received protection transfer message is discarded without being transmitted to the next node. Transmitting the protection switching message received from the node to the next; The method of switching nodes in a network comprising a. In the method of switching at the node receiving the protection switching message in the network,
Generating a flush command of the filtering database for the port that received the protection transfer message; And
And generating a flush command of the filtering database for ports other than the port for receiving the protection switching message. The method of claim 9,
Discarding the repeatedly received protection switching message when the same protection switching message is repeatedly received through the same port. A computer-readable recording medium having recorded thereon a program for performing the method according to any one of claims 1 to 10.

Images