KR100285686B1 - Link Controlling System of Router and Data Controlling Method thereof - Google Patents

Abstract

The present invention relates to a link management system and a data processing method of a router, and in particular, a wide area network link is configured as a multi-link to extend a bandwidth by the number of corresponding links. It relates to a link management system and a data processing method of a router.

The link management system of the router according to the present invention is characterized by including a load allocation application unit for mapping the physical link number and the logical link number between the WAN protocol and the WAN device driver. In addition, the data processing method in a link management system of a router according to the present invention comprises the steps of making a link designated as a primary link and a secondary link to a state capable of transmitting and receiving data; Transmitting data by changing a logical link value to an ID of a primary link or a secondary link; And converting each physical link value into an ID of the primary link to receive data.

Description

Link Control System of Router and Data Controlling Method

The present invention relates to a link management system and a data processing method in a router, and in particular, a router configured to expand a bandwidth by the number of links by configuring a WAN link as a multi-link. Relates to a link management system and a data processing method.

In general, a router is an internetworking device that relays data transmission between different networks, and interconnects a local area network (LAN) and a WAN to allow infinite network extension.

As the network expands on a large scale, the amount of data transmitted on the network increases in proportion. However, due to the limited bandwidth of the WAN, which is relatively slow compared to high-speed LAN, it causes a lot of data loss and consequently degrades the performance of the entire network. In addition, when one part of the network is disconnected, data transmission is impossible and the reliability of the network is greatly reduced.

In order to make up for these shortcomings, it is designed to have a bypass route in constructing a network.If there are multiple routes to one destination, RIP (Routing Information Protocol), OSPF (Open Shortest Path First Version 2, RFC 1583), Routing Protocols such as Interior Gateway Routing Protocol (Cisco Systems Product), Enhanced IGRP (Cisco Systems Product), etc., are used to select the optimal route.

In this case, when there are multiple paths to one destination, data may be concentrated on only one path which is considered optimal. In case of limited bandwidth, data loss may occur even when there are multiple paths between two routers. Done.

Of course, the routing protocols such as OSPF, IGRP, and EIGRP are multi-pathing protocols, and in the case of multiple paths, multipath routing (Bouting) is possible. A protocol for creating a data path to a destination based on the topology of the entire network.

To do this, all routers in the network exchange their own routing information, and with this information, they get information about the entire network. The router calculates an optimal route to a specific destination based on the obtained information and prepares a routing table.

In this case, if there are several routes to the destination, the router selects the route with the lowest cost among the several routes as the optimal route.When the data is transmitted, the router selects the optimal route by referring to the routing table. Will be sent.

If, as a result of the calculation of the optimal path selection, there are multiple routing paths such as OSPF, IGRP, and EIGRP to distribute the load when there are several paths having the same cost to the destination. In particular, the use of IGRP or EIGRP supports the ability to distribute the load even if the cost to the destination is different.

As such, when the routing protocol is used, the load is distributed based on a path through which data is transmitted in the entire network.

However, the OSPF is effective only when the cost of each path is the same, and the routing protocol such as IGRP or EIGRP is a protocol made by Cisco Systems and is not supported in most routers except Cisco.

In addition, a technique for improving bandwidth by configuring two links between two routers does not currently exist as a router applying related technologies, and in a case where there is more than one path from a source to a destination based on the entire network, Only protocols that set the optimal path are used.

In other words, IGRP or EIGRP is a product of the Cisco system, which is problematic to apply to all routers, and OSPF wastes all but the lowest cost paths when there are multiple paths with different costs. This results in a disadvantage in that a large number of IP addresses is required when the two routers are configured with multiple links, because separate IP addresses must be allocated to the multiple links.

In order to solve the problems described above, the present invention is to improve the bandwidth to the load sharing (Load Sharing) in the router, a plurality of physical in the WAN section is relatively slow compared to the high-speed LAN in the Internetworking environment It manages in-links as one logical link and allocates load to each physical link to expand bandwidth to improve the performance of the entire network and link reliability.

1 is a block diagram illustrating a link management system of a router according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a mapping relationship between multiple physical links and one logical link in FIG.

FIG. 3 is a block diagram illustrating the operation between two routers in FIG. 1; FIG.

4 is a flowchart illustrating a data processing method when the bandwidths of the physical links are the same in FIG.

FIG. 5 is a flowchart illustrating a data processing method when bandwidths of physical links are different in FIG. 1; FIG.

Explanation of symbols on the main parts of the drawings

10: IP (Internet Protocol)

20: Wide Area Network Protocol

30: load allocation application unit

40: WAN Device Driver

21: Point-to-Point Protocol (PPP)

22: Frame Relay

51, 52: Router

53-1 ~ 53-4: DSU / CSU (Digital Service Unit / Channel Service Unit)

The load allocation proposed in the present invention is intended to compensate for load balancing using a routing protocol. By configuring a low-speed WAN link between two routers as multiple links, the bandwidth is increased by the number of the corresponding multiple links to improve the performance of the entire network. In addition, when one link is broken, the reliability of the network is improved by providing a backup function that allows the remaining link to transmit data of the broken link.

In other words, the present invention complements one low-speed WAN link existing between two routers to increase the actual bandwidth by configuring multiple WAN links, so that multiple physical links can be managed as one local link. At this time, it does not need the help of routing protocols such as OSPF, IGRP, EGRP, etc., and although there are physically multiple links, only one link logically exists and only one IP address needs to be allocated.

In order to apply load allocation, instead of one physical link, a plurality of physical links are configured, one of the physical links is designated as a primary link, and the other links are secondary links. Links) to assign IP addresses only to the primary link.

For example, WAN protocols such as Point-to-Point Protocol (PPP) or Frame Relay communicate with the other router through one link number, so the WAN protocol is applied when the load allocation is applied. This link number to identify becomes the link set as the said primary link. However, since there are actually several physical links, the WAN device driver identifies each of the physical links as a separate link and transmits and receives data. As a result, in order to manage multiple physical links as a single local link, a part that maps the physical link number and the logical link number is required, and more bandwidth is provided through the part. Data transmission and reception with the other router can be performed to improve network performance.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

As a link management system for load allocation in a router in an internetworking environment according to an embodiment of the present invention, as shown in FIG. 1, an IP 10, a WAN protocol 20, and a load allocation application unit 30 are illustrated. And the WAN device driver 40. Here, the corresponding WAN protocol 20 uses a protocol such as PPP 21 or frame relay 22.

Since the IP 10, the WAN protocol 20, and the WAN device driver 40 are the same as in the conventional configuration, description thereof is omitted, and the load allocation application unit 30 is the WAN protocol 20 and the WAN. By mapping several physical link numbers and one logical link number between the device drivers 40, it manages several physical links as one local link. In this case, the bandwidths of the plurality of physical links may be different from each other, and the WAN protocol 20 recognizes a link designated as a primary link as its link identifier (ID) to communicate with a counterpart router.

Meanwhile, referring to FIG. 2, referring to the mapping relationship between several physical links and one logical link, one of the physical links is designated as the primary link and the remaining links are designated as the secondary links. The ID of the link designated as the head link becomes the ID used in the WAN protocol 20, that is, the number of the logical link. In other words, the ID value of the logical link is the number of the link designated as the primary link on the physical link.

Data is transmitted and received with the above-described mapping relationship. Looking at the load allocation application unit 30, the data transmission and reception process is as follows.

In the case of a router that transmits data, first, a link designated as a primary link and a secondary link is made available for data transmission and reception, and then the WAN device driver 40 transmits the transmission data applied from the PPP 21 or the frame relay 22. Change the logical link value to the link ID of the primary link or the secondary link before sending it. Accordingly, the WAN device driver 40 transmits the transmission data passing through the load allocation application unit 30 to the other party through each physical link. At this time, the corresponding WAN device driver 40 identifies each link through a physical link number.

On the other hand, in the case of a router that receives data, first, a link designated as a primary link and a secondary link is made available for data transmission and reception, and then the data received by the WAN device driver 40 through the primary link or the secondary link. Before transmitting to the upper PPP 21 or the frame relay 22, the link ID of the received data with each physical link value is replaced with the link ID of the primary link. Accordingly, the received data is transmitted to the IP 10 through the PPP 21 or the frame relay 22.

On the other hand, when the load allocation is applied in the load allocation application unit 30 and the primary link and the secondary link is established, even if there are actually multiple physical links, the PPP 21 or the frame relay 22 has only one link. It is perceived to exist. In this case, by using the mutual mapping relationship between the physical link number and the logical link number, it is possible to manage several physical links as one logical link. As a result, the load allocation application unit 30 may achieve the purpose of load allocation by converting the link number between the WAN protocol 20 and the WAN device driver 40.

Then, the data transmission and reception process is divided into cases where the bandwidth of the physical link is the same and different, respectively.

First, when the bandwidth of the physical link is the same, the data processing sequence will be described with reference to FIG. 4. First, a link number of data transmitted by the WAN protocol 20 at the time of data transmission is round-robin. In this manner, each physical link number is changed and transmitted to the WAN device driver 40.

In other words, it is checked whether load allocation is enabled at the time of data transmission (step S1). If the load allocation is enabled, the link number of the corresponding transmission data is round-robin in primary link or secondary. Change to the link number of the link (step S2) and transmit it to the WAN device driver 40, and conversely, if the load allocation is not enabled, the transmission data is transmitted to the WAN device driver 40 as it is ( Step S3).

Accordingly, the WAN device driver 40 transmits the transmission data to the other router through the link. Then, it is checked whether there is data to be transmitted again (step S4), and if the transmission data exists, the above-described process is repeated.

On the other hand, when data is received, the link number of the data received by the WAN device driver 40 from each physical link is changed to the primary link number and transmitted to the WAN protocol 20.

In other words, before transmitting the data received to the WAN device driver 40 to the upper WAN protocol 20 through each physical link, it is checked whether the load allocation is enabled (step S5). If the allocation is enabled, change the link number of the corresponding received data to the primary link number (step S6) and send it to the PPP 21 or the frame relay 22, and if the load allocation is not enabled, The received data is sent to the PPP 21 or the frame relay 22 as it is (step S7).

Then, it is again checked whether there is data received (step S8), and if the corresponding data is received, the above-described process is repeated.

Second, when the bandwidths of the physical links are different from each other, the data processing sequence is described with reference to the flowchart of FIG. 5. First, the link number of data transmitted by the WAN protocol 20 at the time of data transmission is used to determine the bandwidth of each link. Each link number is changed according to the ratio and transmitted to the WAN device driver 40.

In other words, it is checked whether the load allocation is enabled at the time of data transmission (step S9). If the load allocation is enabled, the link number of the corresponding transmission data is adjusted according to the bandwidth ratio (step S10) and the primary link is assigned. Change to the link number of the secondary link (step S11) and transmit it to the WAN device driver 40. On the contrary, if the load assignment is not enabled, the transmission data is transmitted to the WAN device driver 40 as it is. (Step S12).

Accordingly, the WAN device driver 40 transmits the transmission data to the other router through the link. Then, it is checked whether there is data to be transmitted again (step S13), and if the corresponding transmission data exists, the above-described process is repeated.

On the other hand, when receiving data, the link number of the data received by the WAN device driver 40 from each physical link is converted into a primary link number and transmitted to the WAN protocol 20.

In other words, before transmitting the data received to the WAN device driver 40 to the upper WAN protocol 20 through each physical link, it is checked whether the load allocation is enabled (step S14). If the allocation is enabled, the link number of the corresponding received data is changed to the primary link number (step S15) and sent to the PPP 21 or the frame relay 22, and if the load allocation is not enabled, The received data is sent to the PPP 21 or the frame relay 22 as it is (step S16).

Then, it is checked again whether there is data received (step S17), and if the corresponding data is received, the above-described process is repeated.

For example, if two routers performing the above-described operation are configured and described as shown in FIG. 3, a separate link is added to one low-speed WAN link connecting the two routers, thereby providing a plurality of physical connections between the two routers. Allow inlink to exist. In this case, it is assumed that the bandwidths of the plurality of physical links are the same.

Here, because there are two links, there are two link numbers managed by the WAN device driver. One load is set as the primary link and the other is set as the secondary link by applying load allocation to the corresponding multilink. . At this time, an IP address is assigned only to the corresponding primary link.

First, after enabling load allocation to each of the two routers 51 and 52, the first link s0 of the first router 51 is designated as the primary link, and the second link of the first router 51 ( s1) is designated as the secondary link, and the first link s0 of the second router 52 is designated as the primary link and the second link s1 of the second router 52 is designated as the secondary link.

At this time, an IP address is assigned only to the first link s0 of each of the routers 51 and 52 designated as primary links of the two routers 51 and 52. It should be noted that the primary links of the two routers 51 and 52 must be identically designated.

Then, if the first router 51 transmits data to the second router 52, the first router 51 has a primary link number that is a logical link number that it recognizes, and the second router 51 The data is transmitted to 52. At this time, the corresponding data is transmitted to the second router 52 along the physical link through the WAN device driver through the WAN protocol. If load allocation is applied, the link number of the data obtained from the WAN protocol through the load allocation application unit is applied. Is converted into link values of the first link (s0) and the second link (s1), which are physical links, in turn, is passed to the WAN device driver, and the transmission is performed in half and half of the physical links (s0, s1).

On the other hand, the data received by the first router 51 through the first link (s0) and the second link (s1) in the second router 52 goes up to the WAN protocol through the WAN device driver. In this case, the link number of the first link s0 and the second link s1 is converted into the link number of the first link s0 which is the primary link and goes up to the WAN protocol while the load allocation application unit is applied.

In addition, the second router 52 performs the same function as the first router 51.

In the end, there are two physical links between the two routers, but in practice each router operates as if there is only one logical link, resulting in the use of two physical links as one, increasing bandwidth. It will also improve the performance of the entire network.

Moreover, even if one of the two physical links is disconnected, data transmission is continued through the remaining links, which is a great help to improve the reliability of the network.

As described above, in the case of allocating load through multiple links according to the present invention, by using multiple physical links as one logical link, the performance of the entire network can be improved, and even if one link is broken, another link is Backing up can greatly help improve network reliability.

Claims (6)

A link management system of a router, comprising a load allocation application unit for mapping a physical link number and a logical link number between a WAN protocol and a WAN device driver. The method of claim 1, Load allocation is applied by configuring the physical link as a plurality, and one of the plurality of physical links is designated as the primary link and the remaining links are designated as the secondary link. Router's link management system. Making a link designated as a primary link and a secondary link to a state in which data can be transmitted and received; Transmitting data by changing a logical link value to an ID of a primary link or a secondary link; And receiving data by replacing each physical link value with an ID of the primary link. The method of claim 3, The data transmission process includes checking whether the bandwidth of the physical link is the same and load allocation is enabled; Changing the link number of transmission data to the link number of the primary link or the secondary link in a round-robin manner and transmitting the link number; And determining whether there is data to be retransmitted. The method of claim 3, The data transmission process includes checking whether the bandwidths of the physical links are different and load allocation is enabled; Changing the link number of transmission data to link numbers of the primary link and the secondary link according to the bandwidth ratio of each link and transmitting the link number; And determining whether there is data to be retransmitted. The method of claim 3, The data receiving process includes checking whether load allocation is enabled; Changing the link number of data received from each physical link to the primary link number and receiving the link number; And re-checking whether there is data received again.