KR20040028331A - Method for selecting a new Label Switching Path on MPLS network - Google Patents

Abstract

The present invention is to reselect the path faster than the need to reselect the path due to congestion of a specific section in the MPLS network. In general, the MPLS backbone network consists of a core network composed of P routers and a PE router serving as an edge of the backbone network. When rerouting, source routing is performed to consider the entire resources of the network. It is always configured as a destination PE router via a backbone network consisting of P routers. According to the present invention, when the LSP path is reselected, the path change is usually performed in a backbone network composed of P routers. Referring to FIG. Use to determine whether to set the bypass route to re-select the route faster.

Description

Method for selecting a new Label Switching Path on MPLS network

The present invention relates to a multi-protocol label switching (MPLS) network, and more particularly, to reselect a new label switching path (LSP) due to congestion of one link in an MPLS network or to reselect a path using a network structure table when a change is made. It is about a method.

In general, a back bone network using MPLS is composed of a core network n1 and an edge n2 as shown in FIG. 1, and the core network n1 includes a plurality of LSRs (Label). Switching Router (hereinafter referred to as 'P Router') (P1, P2, P3, P4, etc.), and PE (Provider Edge) Router (PE11, P12, PE21, P22, PE31, etc.) at the edge of the backbone Are present.

MPLS is a technology that enhances the Internet function of the Asynchronous Transfer Mode (ATM) exchange, which is the core of the high-speed information network, and integrates Layer 3 and Layer 2 to support various network layer protocols by performing label assignment independently of the Layer 3 protocol. Not only can it be applied, but it can also be flexibly applied to lower layers.

In addition, MPLS reliably distributes the corresponding requirements of traffic to the network for the purpose of maximizing the utilization of network resources while providing reliable network operation.

This maximizes the utilization of network links and nodes and minimizes the effects of failure or congestion of one link or node by distributing network traffic properly across multiple links.

However, MPLS uses LSP selection and signaling protocols such as the Constraint-Based Label Distribution Protocol (CR-LDP) or Resource ReServation Protocol (RSVP), which collects the entire resource information from the source node (PE) to the destination. Source routing that finds routes that can be allocated by (other PE side), creates route list (hereinafter referred to as 'ER'), and sends them to intermediate nodes (ie core network) on the route to set route Select route by (source routing).

When congestion occurs in the route configured in this way, the P router of the node that detects the congestion sends a notify congestion message to the source node PE so that the new route can be set in the PE. In order to set, considering the bandwidth of the LSP to collect and bypass network resource information, the process performed at the first path selection is performed again.

However, as described above, the LSP reselection process in the conventional MPLS network has a problem in that the LSP reselection time becomes longer because the LSP reselection time becomes the same process as the initial LSP selection, and thus the path recovery is delayed.

That is, when a congestion occurs on a specific link, the P router that detects the congestion has informed the congestion of the congestion to the source node, and the path is re-selected as a new LSP path by performing the source routing operation. Recovery time has been delayed.

An object of the present invention is to speed up reselection due to congestion of traffic on a specific link or node in a core network of an MPLS backbone network.

1 is a structural diagram of a general MPLS network.

2 is a view showing that traffic congestion occurs on the path of one LSP in the MPLS network according to an embodiment of the present invention.

3 is a diagram illustrating a possible bypass path in the P router detecting the congestion of traffic according to an embodiment of the present invention.

4 is a flow chart illustrating a re-selection process in a P router detecting a congestion of traffic according to an embodiment of the present invention.

FIG. 5 is a flowchart illustrating a process of calculating a candidate bypass path in a P router detecting a congestion of traffic according to an embodiment of the present invention.

FIG. 6 is a table table showing a network table created in a P1 router according to an embodiment of the present invention.

7 is a diagram illustrating a network structure table in each P router according to an embodiment of the present invention.

According to the present invention for achieving the above technical problem, when traffic congestion occurs on a specific link or node in a core network of an MPLS backbone network and reselects an LSP, the congestion may be performed within the core network itself without informing the source node of the congestion. Re-routing allows a new LSP to be selected that bypasses the node on which the traffic congestion has occurred.

In this case, the present invention places a network table containing candidate bypass path list information in each P router of the core network to use the network table when reselecting a new LSP.

Accordingly, the present invention enables LSP reselection in the corresponding P router of the core network that is in charge of the node where the traffic congestion has occurred, so that the path recovery can be performed much faster than the conventional LSP reselection in the PE of the source node.

To this end, the present invention provides a method for reselecting a path in an MPLS network including a core network composed of a plurality of P routers and an edge where a plurality of PE routers connected to a source node are located, the LSP configured by one source node. A first step of detecting that a link from a first P router located in a network to a next hop is congested; and identifying the destination node of the LSP in the first network by the first P router, A second step of checking whether a bypass path exists by using a network table stored therein; and if the network path has a bypass path to the destination node, the first P router selects a destination path from the bypass path. Selecting a first bypass path that is the shortest path to the node, and wherein the first P router passes the first bypass path to intermediate nodes on the first bypass path; As far as the fourth step to determine whether to ask whether or not, when the first bypass path is enabled, and the first P router, a fifth step to establish a re-election of the first bypass path to the new LSP.

Preferably, the network table preferably stores a route information list for the remaining P routers in the same core network except for the P router in which the network table is stored.

Hereinafter, a path reselection method in an MPLS network according to an embodiment of the present invention will be described with reference to the accompanying drawings.

First, to explain the present invention, a process of selecting an initial LSP with reference to FIG. 2 and designating one of a specific link on the selected LSP as a point where traffic congestion occurs.

2 is a view showing that traffic congestion occurs on the path of one LSP in the MPLS network according to an embodiment of the present invention.

PE11, the source node located at the edge, finds the shortest path according to the source routing algorithm to establish the LSP (L10) to the destination node PE41 and makes a list of nodes on the path. At this time, the node list is PE11, P1, P4, PE41.

When the node list is created as described above, PE11 forwards the node list to each intermediate node (P1, P4) on the path, so that LSP (L10) started at PE11 traffics to the destination node PE41 through the corresponding node (P1, P4). By notifying that this will be delivered, the LSP L10 is set in advance at each intermediate node P1, P4.

When the LSP L10 is set as described above, the PE11 transmits data traffic along the LSP L10. However, when traffic is excessive on a specific link P1-P4 during data traffic transmission, congestion occurs.

Then, the P1 router detects the congestion of the links P1-P4.

Here, LSP L20 shown in FIG. 2 is an LSP reselected from the prior art.

Looking at the LSP L20 reselected in the prior art, the arrows are shown in both directions. This is because the data transmission path at the time of LSP reselection is transferred from the P1 router to PE11, and PE11 reselects LSP (L2) to transmit data to PE41.

However, the LSP reselection of the present invention is different from FIG. 2 as shown in FIG.

3 is a diagram illustrating a possible bypass path in the P router detecting the congestion of traffic according to an embodiment of the present invention.

As shown in FIG. 3, when the source node PE11 selects the initial LSP L3 and sends traffic to P1, the PE11 does not participate in the LSP reselection process for the initial LSP L3. LSP reselection is performed by a P1 router that detects a congestion of traffic on links P1-P4, which uses either a network table or a route (1), route (2), or route (3). Is reselected as a new LSP.

At this time, the LSP reselection criterion in the P1 router first selects a short path and selects a path without congestion.

In the network structure table, as shown in FIG. 7, route list (ER) information is recorded.

7 is a view showing the contents of a network structure table according to an embodiment of the present invention.

In FIG. 7, a is the content of the network table stored in the P1 router, b is the content of the network table stored in the P2 router, c is the content of the network table stored in the P3 table, and d is the network stored in the P4 table. The contents of the structure table.

Each network structure table is created under the assumption that the core network is composed of only P1, P2, P3, and P4 routers.

Referring to each network structure table shown in FIG. 7, each router records route information to the routers other than itself. Here, if each row of the network table is called a point (i), the table records that the smaller the value of the point (i) (i.e., the higher the upper row of the table), the shorter the route to the destination. As (i) increases, record long paths sequentially.

Each router in the core network should be selected from the shortest path in case of reselection of the LSP.

Hereinafter, the LSP reselection operation in the P1 router that realizes FIG. 3 will be described with reference to FIG. 4.

4 is a flow chart illustrating a re-selection process in a P router detecting a congestion of traffic according to an embodiment of the present invention.

As shown in FIG. 4, when the P1 router detects the traffic congestion of the links P1-P4 (S410), it initializes the point i of the network table to be i = 1 (S411).

After performing the initialization process (S411), the P1 router determines whether a bypass path corresponding to i = 1 exists, that is, a bypass path to the destination node exists (S412).

If there is no bypass path to the destination node in the determination (S412), since the route reselection is impossible for the P1 router, the source node performs the route reselection as in the prior art. To this end, the P1 router transmits a congestion report message to the source node to allow the source node to re-renew the LSP path (S418).

However, if there is a detour path to the destination node as shown in FIG. 7 in step S412, i = 1 is selected as the first line from above, and it is determined whether the paths P1-P4 having i = 1 are valid. (S413).

To determine whether the validity is to check and determine whether the traffic of the path is congested or there is an error in the path.

In this determination, the router P1 increases i by one if the selected i-th path is invalid (S416), and checks whether there is a next path (S417).

Referring to FIG. 7, since the paths P1-P4 are not valid, the P1 router will select P1-P2-P4 which is the bypass path of the next point (i = 2).

However, if there is no next path in the check (S417), since the P1 router cannot reselect the LSP, the source node causes the LSP to reselect (S418).

However, if there is a next path in the check (S417), the P1 router again performs a process (S413) of determining whether the path (P1-P2-P3) corresponding to the point (i = 2) is valid, and if it is not valid The process (S416-417) is repeated, and if valid, reselects the currently selected path as a new LSP.

Here, in order to reselect a new LSP, the P1 router sends a message indicating whether new LSP configuration and bandwidth allocation are possible to all intermediate nodes P2 on the selected paths P1-P2-P3.

Then, when the P1 router receives the message that the route can be established at the intermediate node P2, the P1 router sends a new route list ER to the source node and the intermediate nodes P2 to be newly set up, thereby selecting P1-P2-. P4 is set to a new LSP (S415).

In the above, the P1 router repeats the processes 410-418 to find a valid path up to the last path of the candidate route list table, and sends a congestion notification message to the source node to configure a new LSP through source routing.

Hereinafter, a process of creating a network structure table will be described with reference to FIGS. 5 and 6.

FIG. 5 is a flowchart illustrating a process of creating a network table according to an embodiment of the present invention, and FIG. 6 is a table table illustrating making a network table in a P1 router according to an embodiment of the present invention.

All P routers (P1 routers, P2 routers, and P3 routers collectively) make respective destination node lists for all nodes except their own nodes (S510).

For example, referring to a of FIG. 7, it is as if the P1 router has a destination node list for the remaining nodes P2, P3, and P4 routers. P2, P3, and P4 routers will also create a list of four destination nodes, just like P1 routers.

Here, since the process of creating a network table in all P routers is the same, the process of creating a P1 router is described as an example, and the process of creating a route list for each destination node list in the P1 router is also the same. As an example, the process of making a recipe will be described.

After creating the target node, the P1 router displays a link, a node, and a state of each link for each target node (S511).

Referring to FIG. 2 and FIG. 6, the P1 router denotes its own node as the source node SN as SN in the node field, and is connected to its node as shown in FIG. 6. Record the next hop link (L1, L2, L3). The next hop link is the conclusion obtained from FIG. 2. The P1 router records the hops connected to each link in one hop space. At this time, since the next hop connected to the L3 link is P4, the DN indicated as the destination node is displayed instead of P4. Then, the status of each link is displayed. If the next hop of each link is the destination node, a candidate path list is created, otherwise it is marked as 'next hop'. In the first step of FIG. 6, the candidate path list is designated as the L3 link as the SN-DN. Here, when the next hop becomes the searched node, it records 'loop occurrence' in the status column.

After performing the process (S511), the P1 router checks the status column of each link to check whether each link has a status of 'next hop' (S512), and if the status is 'next hop', increases the number of hops (S513). After checking what the next hop state is, it performs the same operation as the process (S511) (S514).

After the process (S514), if there is a link having a 'next hop' in the status column, the process (S512-514) is performed on the link.

In the process (S512), if there is a link that is not recorded as a next hop or loop occurrence in the status field, the link is a link to which the source node and the destination node are connected, as shown in Fig. 6, so that the P1 router is a source node. And the path of the link to which the destination node is connected are recorded as candidate path lists in the network table according to the hop number, router order, or ID order (S515).

The results thus recorded are shown in FIG. 7.

The process S515 is repeatedly performed until the number of target nodes is recorded (S516-S517).

After the repetition process, the P1 router completes the preparation of the candidate path list for the P4 target node (S518).

The technical spirit of the present invention has been described above with reference to the accompanying drawings, but this is by way of example only and not intended to limit the present invention. In addition, it is obvious that any person skilled in the art can make various modifications and imitations without departing from the scope of the technical idea of the present invention.

The present invention has an effect of enabling a fast reselection using a network structure table in which candidate path list information for bypassing a link in a core network is recorded when a link or node congestion occurs in a core network in an MPLS backbone network.

Claims (6)

In the re-selection method in the MPLS network including a core network consisting of a plurality of P routers, and an edge where a plurality of PE routers connected to the source node is located, A first step of detecting that a link to a next hop is a traffic congestion in a first P router located on an LSP set by one source node; A second step of the first P router identifying a destination node of the LSP existing in the core network and confirming whether a bypass path to the destination node exists by using a network table stored therein; A third step of the first P router selecting a first bypass path, which is the shortest path to the destination node, from the bypass path if there is a bypass path to the destination node in the network table; A fourth step of the first P router asking the intermediate nodes on the first bypass path whether or not the first bypass path is valid; And And if the first detour path is valid, the first P router re-selects the first detour path to a new LSP. The method according to claim 1, The network structure table rerouting method in the MPLS network, characterized in that the route information list for the remaining P routers in the same core network except the P router is stored as the destination node. The method according to claim 2, The fifth step, Receiving a message that is valid from intermediate nodes on the first bypass path, transmitting information of a new LSP to the intermediate nodes on the source node and the new LSP, and converting the first bypass path to a new LSP Reselection method comprising the step of re-selection in the MPLS network. The method according to claim 1 or 3, After the second step, And if the bypass path does not exist in the network table, the first P router notifies the source node of the congestion of traffic of the corresponding link and causes the source node to reselect a new LSP. Reselection method. The method according to claim 4, After the fourth step, If the first bypass path is invalid, the first router examines its network table to determine if another bypass path exists; and if the other bypass path does not exist in the check, a new LSP at the source node. Selecting a bypass path, which is the shortest path to the destination node, from among the bypass paths existing as the second bypass path, and performing the fourth step, if another bypass path exists in the inspection. Reselection method in the MPLS network, characterized in that. On your computer, A first step of detecting that a link to a next hop is a traffic congestion in a first P router located on an LSP set by one source node; A second step of the first P router identifying a destination node of the LSP existing in the core network and confirming whether a bypass path to the destination node exists by using a network table stored therein; A third step of the first P router selecting a first bypass path, which is the shortest path to the destination node, from the bypass path if there is a bypass path to the destination node in the network table; A fourth step of the first P router asking the intermediate nodes on the first bypass path whether or not the first bypass path is valid; And And if the first bypass path is valid, the first P router may execute a fifth step of reselecting the first bypass path to a new LSP.