KR20120124641A - Internet Traffic Load Distribution Method and System using IGP Routing Protocol - Google Patents

Abstract

PURPOSE: A method and system for distributing internet traffic load using IGP routing protocol are provided to distribute correctly traffic load by setting a route of passing through adjacent routers for advertising a next hop of static declaration as a best route. CONSTITUTION: A router(D) selects a line passing through a router(A) as a best routing route on an IP prefix(a)(S10). The router(B,C) sets up next hop on the IP prefix by static declaration(S20). The router sets a line passing through the router(B,C) for advertising the fixed next hop in the best routing route(S40). The router distributes the traffic of the IP prefix through the router(A,B,C) to plural lines(S50). [Reference numerals] (AA,BB) BGP routing information linkage; (CC,DD) IGP redistribution; (S10) Selecting A as the best; (S30) Managing in a DB; (S40) Selecting all Next Hop routes according to the IGP redistribution; (S50) Distributing traffic through A,B,C

Description

Internet Traffic Load Distribution Method and System using IGP Routing Protocol

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and system for load balancing Internet traffic, and more particularly, to redistribute a message according to an Internal Gateway Protocol (IGP) routing protocol when routing a router to a next hop for a destination IP address. By including the route through neighboring routers advertizing next hops statically declared as the best route, accurate load balancing is easily achieved without changing the value of the Internal Border Gateway Protocol (BGP) attribute. The present invention relates to a method and system for load balancing Internet traffic, which is capable of multi-path routing through circuits connected to multiple routers, and enables accurate load balancing to the remaining circuits even in the event of a circuit failure.

In network (Internet) communication as shown in FIG. 1, for example, an AS-Autonomous System (AS) 100 belongs to an A-ISP (Internet Service Provider) and an AS 200 belongs to an B-ISP (Internet Service Provider). Let's take a look at the case where the router D of the A-ISP sends traffic to the destination IP address via a predetermined next hop of the B-ISP. In this case, as shown in FIG. 2, the router D uses the next hop for IP prefix (a) according to the BGP routing information received from the router A. 1.1.1.1. Next hops for IP prefix (a) based on BGP routing information received from Router B. 2.2.2.2. Next hop for IP prefix (a) according to the BGP routing information received from Router C. 3.3.3.3. Information can be managed in routing tables.

At this time, the router D according to the network policy according to the BGP (Internal Border Gateway Protocol) attribute value of the BGP routing information, for example, weight, AS-Path (path length), Local-Preference (self-affinity), IP (Internet Protocol) Assume that the circuit via router A is selected as the best path for prefix (a). That is, routers B and C also transmit BGP routing information including the respective next hop information for the IP prefix (a), but only the circuit passing through router A is selected as the best path, and router D is the IP prefix (a Traffic containing the destination IP address belonging to the "

Similarly, router D may select the circuit via router B as the best path for IP prefix (b) according to the BGP attribute value of BGP routing information, and select the circuit via router C for IP prefix (c). Can be selected as the best route. Accordingly, the router D may transmit traffic including the destination IP address belonging to the IP prefix (b) to the router B, and the traffic including the destination IP address belonging to the IP prefix (c) to the router C. As a result, the router D can distribute the traffic of each of the IP prefixes (a), (b), and (c) to circuits passing through routers A, B, and C, respectively.

However, in the conventional method of distributing the traffic transmission load to each router according to the BGP routing protocol, the traffic load may be different according to the IP prefix, so that it is difficult to accurately load balance and the traffic is biased to a specific line in case of a line failure. This happens. Furthermore, in case of line failure, by changing the BGP attribute value of BGP routing information of other routers except the router connected to the line, router D can receive the changed BGP attribute value, for example, There is inconvenience in operation that should be able to select the best path again according to Local-Preference.

Accordingly, an object of the present invention is to solve the above-described problem, and an object of the present invention is to establish an internal gateway protocol (IGP) routing protocol when routing from a router to a next hop for a certain destination IP address. Providing a route through adjacent routers advertising static hop declared next hop through the redistribution message according to the best route, to provide a method and system for Internet traffic load balancing capable of accurate load balancing There is.

First, to summarize the features of the present invention, a traffic load balancing method for traffic transmission between Internet Service Providers (ISPs) interworking BGP routing information according to an aspect of the present invention for achieving the above object of the present invention (A) statically declaring a next hop for a predetermined IP prefix at the router and setting the static hop; And (B) as the router advertises the next hop statically declared according to the Internal Gateway Protocol (IGP) routing protocol through a redistribution message, the statically declared next to the IP prefix at the other router receiving the advertisement. Including in the routing path a circuit via the router advertising a hop.

In step (B), the other router includes a plurality of circuits through a router advertising the same next hop for the same IP prefix in a routing path, and distributes traffic of the corresponding IP prefix to the plurality of circuits for transmission. Can be. At this time, the traffic to be transmitted through the plurality of lines is equally distributed and transmitted.

When a faulty line occurs among the plurality of lines, traffic to be transmitted through the lines except for the corresponding line is equally distributed and transmitted.

In step (B), the other router receives BGP routing information including the next hop for the IP prefix from a plurality of routers, and includes a circuit via any one of the plurality of routers in the routing path. And a circuit via the router advertising the statically declared next hop may additionally be included in the routing path.

In addition, in a network system for traffic load balancing between Internet Service Providers (ISPs) interworking BGP routing information according to another aspect of the present invention, a next hop for a predetermined IP prefix is statically declared. A first router advertising the statically declared next hop according to an internal gateway protocol (IGP) routing protocol through a redistribution message; And a second router that receives the advertisement and includes a circuit in the routing path via the first router that advertises the static declared next hop for the IP prefix.

According to the Internet traffic load balancing method and system according to the present invention, a route through neighboring routers advertising static hop declared next hop through redistribution message according to IGP routing protocol is included as a best route. By doing so, accurate load balancing can be easily performed, and multi-path routing is possible through circuits connected to multiple routers, so that in the event of a circuit failure, accurate load balancing can be performed on the remaining circuits. In addition, it is possible to simply manage the routing according to the IGP routing protocol without changing the BGP attribute value, and to provide convenience in the operation of the routing equipment since it can be executed once with redistribution advertisement after static declaration.

1 is a view for explaining a conventional network system.
FIG. 2 is a diagram for describing routing table management and best path selection of router D in FIG. 1.
3 is a diagram illustrating a network system according to an embodiment of the present invention.
4 is a flowchart illustrating an operation of a network system according to an embodiment of the present invention.
FIG. 5 is an example for explaining BGP routing table management and best path selection of router D of FIG.
FIG. 6 is an example for explaining IGP routing table management and best path selection for the IP prefix (a) of router D of FIG. 3.
FIG. 7 is an example for explaining IGP routing table management and best path selection for the IP prefix (b) of router D of FIG. 3.
FIG. 8 is an example for explaining IGP routing table management and best path selection for the IP prefix c of the router D of FIG. 3.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout.

3 is a diagram illustrating a network system according to an embodiment of the present invention.

3, the network system according to an embodiment of the present invention, an AS (Autonomous System: autonomous system) that can interwork BGP routing information by exchanging IP information according to an external border gateway protocol (eBGP) routing protocol. It includes 100 A-ISP (Internet Service Provider) and AS 200 B-ISP (Internet Service Provider). For example, router D of B-ISP selects a route through routers A, B, and C as a routing path. To the destination 500 via the next hop (router) 1.1.1.1, or 2.2.2.2, or 3.3.3.3 of the A-ISP via other routers 300 or via other routers 300. A case of transmitting traffic will be described.

Referring to FIG. 4, routers A, B, and C of the A-ISP may manage BGP routing information by interworking with A-ISP according to the eBGP routing protocol, and routers A, B, C, and D of the A-ISP Composed of a full mesh form, BGP routing information (see FIG. 5) can be managed in a database by interworking BGP routing information according to an iBGP (Internal Border Gateway Protocol) routing protocol, and an IGP (Internal Gateway Protocol) Other IP information can be linked according to the routing protocol to manage the IGP routing table (see FIGS. 6, 7, and 8) in the database. For example, routers A, B, and C may advertise the redistribution message to router D according to the IGP routing protocol to refer to the best path selection. Routing type according to the IGP routing protocol may be RIP (Routing Information Protocol), Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-IS), etc. In addition to the information about the routing type, a metric value that is a reference for selecting the best path may be included.

First, according to the interworking of the above BGP routing information, Router D may manage the BGP routing table shown in FIG. 5 in the database. For example, Router D may use the Next Hop for IP Prefix (a) in accordance with the BGP routing information received from Router A. 1.1.1.1. Next hops for IP prefix (a) based on BGP routing information received from Router B. 2.2.2.2. Next hop for IP prefix (a) according to the BGP routing information received from Router C. 3.3.3.3. Information can be managed and based on the BGP attribute value of the BGP routing information in the network policy, e.g., weight, AS-Path (path length), Local-Preference (self-affinity), router for IP prefix (a). The circuit via A can be selected as the best routing path (see S10 in FIG. 4).

In a similar manner to the above, router D may select a circuit via router B as the best routing path for IP prefix (b) according to the BGP attribute value of BGP routing information, and select router C for IP prefix (c). The line passing through can be selected as the best route. Accordingly, the router D can distribute the traffic of each of the IP prefixes (a), (b), and (c) to circuits passing through routers A, B, and C, respectively.

Further, in the present invention, router D, for example, according to the BGP routing protocol, in addition to selecting the circuit via router A as the best routing path for IP prefix (a), as shown in Figs. In addition, by referring to the IGP routing table according to the IGP routing protocol, circuits via routers advertising statically declared next hops can be additionally included in the best routing path.

That is, in the present invention, it is easy to accurately load balance by including a route through adjacent routers advertising static declared next hops through a redistribution message according to the IGP routing protocol. In addition, multi-path routing is possible through circuits connected to multiple routers, so that in the event of a circuit failure, accurate load balancing is possible to the remaining circuits. In addition, it is possible to easily manage routing according to IGP routing protocol without changing BGP attribute value in routers, and to provide convenience in routing equipment operation because it can be executed by redistribution advertisement after static declaration. .

To this end, it will be described in more detail with reference to the flowchart of FIG. 4.

For example, routers B and C may statically declare the next hop for the IP prefix (a) by an operator or the like. For example, an operator can manually enter information such as next hops for IP prefixes to be redistributed in static declarations into each router, whereby routers B and C are internal gateway protocols (IGPs). According to the routing protocol, the next hop statically declared for the IP prefix (a) may be advertised to the router D through a redistribution message (S20).

Accordingly, the router D receives the corresponding advertisement advertised by the routers B and C through the redistribution message, and the next hop for the IP prefix (a) received from the routers B and C, respectively, as shown in Fig. 6. The information is managed in the IGP routing table of the database (S30), and the best routing of the circuit via the corresponding routers B and C, which advertises the next hops statically declared for the corresponding IP prefix (a) when the best routing path is selected. It may be included in the path (S40). That is, router D may include all the circuits through the router advertising the same next hop for the same IP prefix in the routing path, and distribute the traffic of the corresponding IP prefix (a) to the selected plurality of circuits and transmit them. Will be. As a result, Router D additionally adds the circuit via its routers B and C, which advertises the next hop statically declared for IP prefix (a), in addition to the circuit via Router A selected according to the BGP routing protocol. By including it, the traffic of the IP prefix (a) can be distributed and transmitted to a plurality of lines through routers A, B, and C (S50). If the IP prefix (a) is, for example, 10.10.10.0/24, traffic to be sent to a destination 500 IP address belonging to it, that is, 10.10.10.0 to 255 is transmitted through routers A, B, and C. Can be evenly distributed and transmitted. For example, when the traffic to be transmitted to the destination 500 is 300MB, the router D may distribute the 100MB evenly through the circuits of the routers A, B, and C and transmit the same.

A case where Router D transmits traffic of IP prefix (a) has been described above as an example. Router D manages the IGP routing table in a database as shown in FIGS. 7 and 8, while traffic D of IP prefixes (b) and (c) The routers A, B, and C can also be distributed over a plurality of lines for transmission.

For example, Router D receives the corresponding advertisement advertised by the routers A and C through the redistribution message and next hops to IP prefixes (b) received from the routers A and C, respectively, as shown in FIG. Next hops are managed in the IGP routing table of the database, and the corresponding hops advertised by routers A and B through redistribution messages are received by routers A and B as shown in FIG. . Information can be managed in the database's IGP routing table.

Accordingly, router D not only passes the circuit via router B selected according to the BGP routing protocol for IP prefix (b), but also via the corresponding routers A and C that advertise next hops statically declared for IP prefix (b). By additionally including the circuit to the best routing path, it is possible to distribute the traffic of the IP prefix (b) to a plurality of circuits through routers A, B, and C to transmit.

In addition to the circuit via router C selected according to the BGP routing protocol for the IP prefix (c), the router D also passes through the corresponding routers A and B that advertise next hops statically declared for the IP prefix (c). By additionally including the circuit in the best routing path, the traffic of the IP prefix (c) can be distributed and transmitted to the plurality of circuits through routers A, B, and C.

In turn, Router D distributes traffic that must be sent to the destination 500 of all IP prefixes (e.g., IP prefixes (a), (b), (c)) managed through all of its adjacently connected routers. Can be sent. As a result, accurate load balancing is possible, and multi-path routing is possible through a line connected to a plurality of routers. Therefore, when a faulty line occurs among a plurality of lines, traffic to be transmitted through the lines except for the corresponding line is equally distributed and transmitted. Can be. For example, router D may equally distribute the traffic of IP prefix (a) through routers B and C even when the circuit via router A has a failure in the above structure.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

Internal Border Gateway Protocol (iBGP)
eBGP (external Border Gateway Protocol)
AS (Autonomous System)
Internal Gateway Protocol (IGP)
Routing Information Protocol (RIP)
Open Shortest Path First,
IS-IS (Intermediate System to Intermediate System)

Claims (6)

In the traffic load balancing method for traffic transmission between Internet Service Providers (ISPs) interworking BGP routing information,
(A) statically declaring a next hop for a predetermined IP prefix in the router and setting the static hop; And
(B) As the router advertises next hops statically declared in accordance with the Internal Gateway Protocol (IGP) routing protocol through redistribution messages, the next hops statically declared for that IP prefix at other routers receiving the advertisements. Including in the routing path a circuit via the router advertising the
Traffic load balancing method comprising a. The method of claim 1,
In step (B), the other router includes a plurality of circuits through the router advertising the same next hop for the same IP prefix in a routing path, and distributes traffic of the corresponding IP prefix to the plurality of circuits and transmits the same. Traffic load balancing method, characterized in that. The method of claim 2,
And distributing the traffic to be transmitted through the plurality of lines by equally distributing the traffic load balancing method. The method of claim 2,
When a faulty line is generated among the plurality of lines, the traffic load balancing method, characterized by equally distributing the traffic to be transmitted through the lines except for the corresponding line. The method of claim 1,
In step (B), the other router receives BGP routing information including the next hop for the IP prefix from a plurality of routers, and includes a circuit via any one of the plurality of routers in the routing path. And additionally including a circuit in the routing path via the router advertising the statically declared next hop. In the network system for traffic load balancing between Internet Service Providers (ISPs) interworking BGP routing information,
A next router configured to statically declare a next hop for a predetermined IP prefix, the first router advertising the statically declared next hop through a redistribution message according to an Internal Gateway Protocol (IGP) routing protocol; And
A second router that receives the advertisement and includes a circuit in the routing path via the first router that advertises the statically declared next hop for the IP prefix;
Network system comprising a.

Images