A detailed description of routing protocols in computer networks

Source: Internet
Author: User

Before we talk about routing protocols, let's look at the structure of the routers:

A router is a dedicated computer with multiple input ports and multiple output ports whose task is to forward groupings.


The core of the routing part is the routing processor, which can construct a way out of the table according to the selected routing protocol, and Exchange routing information with regular or regular and neighboring routers to update and maintain its own routing table.

The packet forwarding section includes a set of input ports, a switching fabric, and a set of output ports. The switching structure, also known as the Exchange organization, is to forward the incoming IP datagram from the appropriate port on the router according to the forwarding post. The switching fabric can be seen as a network in the router.


Because the author mainly studies wmn, so first from the WMN routing protocol, but WMN routing protocol is a lot of reference to the cable network routing protocol, so the goal is to let the reader for the Network Layer routing protocol has a global understanding. Although there are some terminology in this article, it is basically explained in the context. If some of the content is missing or not explained in place, please ask the reader to search for information, please forgive me.


The wmn of dynamic and multi-hop makes the traditional routing protocol not be able to deal with the fast change of topological structure and link state in wireless mesh network.

Wmn by the development of ad hoc networks, ad hoc networks focus on mobile, while the WMN focus on wireless. At present, the main WMN routing protocol is developed from the Ad Hoc Network routing protocol. There are three types of routing protocols: Prior routing protocol, reactive routing protocol, and hybrid routing protocol.


In a priori routing protocol, each node uses periodic routing packet broadcasts to maintain a routing table containing routing information to other nodes, regardless of whether there is a communication requirement, and when the network topology changes, the node switches and updates the routing table in the network. Once the source node is sent to the packet, the route to the destination node can be obtained immediately.

This type of routing protocol is usually done by modifying the existing routing protocol to accommodate Ad Hoc wireless network requirements, such as by modifying the route Information Protocol (RIP) to obtain the purpose of the node sequence distance vector Protocol (DSDV). Therefore, the delay of the routing protocol is small, but the protocol requires a lot of routing control packet routing, the Protocol overhead is large. The common prior routing protocol has DSDV,HSR,GSR,WRP and so on.

DSDV protocol

DSDVthe basic principle of the protocol is that each node maintains a routing table to the other nodes, and the contents of the table are routed"Next Hop"node. DSDVThe innovation is to set a serial number for each route, the route with the large serial number is the preferred route, the sequence number is the same, the route with fewer hops is the preferred route. Normally, the sequence number of a node broadcast is a monotonically incrementing even, when the nodeBDiscovery to NodeDthe routing(The route sequence number iss)after the interrupt, the nodeBbroadcasts a routing message that informs the routingSerial Numberinto S+l, and sets the hop number to infinity so that any one of node A's routing tables that send information through B includes an infinity distance, which is until the a receives a valid route to D (the route sequence number is s+1-1).

Using a hierarchical routing method, the routing protocols of the Internet can be divided into:

Internal Gateway Protocol IGPI ( nterior Gateway Protocol ): There are a number of specific protocols, such as: RIP and the OSPF and so on.

External Gateway Protocol EGP ( External Gateway Protocol ): The protocol currently in use is BGP

Below we introduce separately:

Routing Information Protocol Rip (Routinginformation Protocol)

RIPusing distance vector algorithm, the biggest advantage is simple. is less applicable in practical use. RIPThe protocol allows a path to contain a maximum of only thea router. Therefore, the distance"equals -is equivalent to non-reach. VisibleRIPapplies only to small Internet. For larger networks, you should useOSPFagreement. RIPYou cannot use more than one route between two networks at the same time. RIPSelect a route with a minimum of routers (that is, the shortest path), even if there is another high-speed (low-latency) router with more routes. RIPprocess usesUDPof the520port to send and receiveRIPgrouping. RIPGroup each interval101sent in the form of a broadcast, in order to prevent the occurrence"Broadcast Storm", its subsequent groupings will be sent after a random delay. In theRIP, if a route is not brushed within 180s, the corresponding distance is set to infinity and the table entry is removed from the routing table. RIP groupings are divided into two types: request grouping and response grouping.


RIP the protocol is characterized by:

Exchange information only with neighboring routers

The Router exchange information is the current router knows all the information, namely its own routing table

Exchange routing information at regular intervals

RIP one problem with the protocol is that it takes a long time to transfer this information to all routers when the network fails.

Open Shortest Path First OSPF (openshortest Path first)

OSPF The principle is simple, the realization is complex, uses the Dijkstra the shortest path algorithm proposed SPF.

OSPF The main feature is the use of Distributed link state protocols rather than RIP such a distance vector protocol. the three points of OSPF and rip are different than RIP protocols .

1. send information to all routers in this autonomous system. The method used here is the flooding method (flooding)

Flooding method (flooding)

in the OSPF protocol, when the link state is changed, use the flooding to all Routers send a message.

For example, source node Want to send a piece of data to the target node. SOURCE node data transfer data transfer target node " Span style= "Color:rgb (51,51,51)" or data set Lifetime is 0 so far.

2. The information sent is the link state of all the routers that are adjacent to this router, but this is only part of the information that the router knows. " Link State " includes the router and which routers are adjacent, along with some metric information for the link. the RIP protocol sends the distance to all routers and the next hop router.

3. the router sends this information to all routers with flooding only when the link state is changed.

because of the frequent exchange of routing information between routers, all routers can eventually establish a link state database, which is actually a topology diagram of the whole network. each router in the RIP protocol knows the distance of all networks and the next hop router, but does not know the topology of the whole network (only to the next hop router, to know how the next hop should go).

OSPF the rapid convergence of the update process is an important advantage. ("Convergence" is the process of getting the correct routing information for all nodes in the autonomous system)

OSPF is sent directly with an IP datagram without UDP (the Protocol field value of the IP datagram header is 89)


OSPF has the following 5 types of grouping:

1, greeting (hello) group, used to detect and maintain the accessibility of the neighboring station.

2. Database Description Group, send the summary information of all link state items in the link state database to the neighboring station.

3. Link status requests (link state request) are grouped, like the other party requests to send details of certain link-state items.

4, Link status updates (link State update) group, with flooding method to update the link state of the whole network.

5. Link state acknowledgment (link acknowledgment) grouping, confirming the grouping of link updates.

To enable OSPF to be used in large networks, OSPF divides an autonomous system into smaller ranges called regions. Represents an autonomous system divided into 4 regions. Each region has a 32-bit area marker (in dotted decimal notation)

Of course, a region can not be too large, the router in a region should not be more than 200.


The advantage of dividing the region is that the scope of the exchange of link state information using flooding is confined to each region rather than the entire autonomous system, which reduces the amount of traffic on the entire network.


External Gateway Protocol BGP

BGP is the protocol for exchanging routing information between routers of different as (area System).

Border Gateway Protocol BGP is only trying to find a better route that can reach the destination network (not in circles), rather than looking for an optimal route. BGP uses a path vector routing protocol.

Generally speaking, two BGP speakers are connected through a shared network. A BGP spokesperson wants to exchange routing information with a BGP spokesperson for other autonomous systems by first establishing a TCP connection (port number 179), then exchanging BGP messages on this connection to establish a BGP session and exchanging routing information with the BGP session.


There are 4 types of messages BGP-4 in RFC4271:

1. Open (opened) message, used to establish a relationship with another BGP spokesperson, to initialize the communication.

2. Update message, which is used to advertise the information of a road, and to list the multiple routes to be revoked.

3, KEEPALIVE (KeepAlive message), used to periodically verify the connectivity of the neighboring station.

4, NOTIFICATION (notification) message, used to send the detected errors.

A router-request message was added to the RFC2918 to request the peer to re-advertise.



Reactive routing Protocol

The reactive routing protocol is also referred to as routing protocol or on-demand routing protocol. is dedicated to the mobile ad hoc network workgroup MANET (mobile ad-hoc NETworks ) environment proposed. In contrast to the table-driven routing protocol, this type of protocol does not generate a route beforehand, but only when the source node needs it. Therefore, the routing table topology part of the information.

nodes do not save the entire network of timely and accurate routing information. When the source node sends a message to the destination node, the source node initiates the route lookup process in the network and the corresponding route is found before the message is sent. To improve efficiency, nodes can save the found routes in the cache for subsequent send use. The characteristics of the on-demand routing of the reactive routing protocol can be better adapted to the wireless network environment where the nodes move more frequently, and only need to update the routing information of the relevant paths that need to send the data when the nodes move.  


Hybrid Routing protocol

Hybrid routing is a combination of the prior routing protocol and the reactive routing protocol, which uses a priori routing protocol in local scope, maintains accurate routing information, and can reduce the range of routing control message propagation. When the target node is far away, it uses the reactive routing protocol to find the discovery route. The ZRP protocol is the representative of a hybrid routing protocol.


The ZRP protocol is a tiled routing protocol designed for changing communication environments such as reconfigurable wireless networks, RwN. Define an area for each node that contains nodes where the distance (that is, the number of hops) is within a defined range. This distance is become the area radius rzone. Each node only needs to know the topology within its routing area, and its routing information is updated with the topology updates within the zone. This way, although the network is large, the update is done only in the local area. Because the distance is greater than 1, there is a large overlap in the area.


Resources:

1. Routing protocol Analysis of wireless mesh network . TechTarget[citation date 2016-6-27]

2, Shehiren. Computer Network (5th edition) [M]. Beijing: Electronic industry Press, China.

A detailed description of routing protocols in computer networks

Related Article

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.