Welcome to our Online Store!

Routing algorithms distinguish between the management distance and the maximum number of hops

2018-02-12 14:27:11

management distance is assigned a number, the artificially represented by the routing protocol of priority, digital yue xiaoyue preferred this routing protocol routing announcement.Static routes such as the default administrative distance is 0, rip is 120, if the routing announcement that reaches a segment from the two notices at the same time, will adopt the path of the static routing announcement.The

largest hop mainly for distance vector routing protocol, is that this routing protocol can send a routing announcement at most how many routers.The maximum number of jump, for example rip is 15, then has the rip protocol transmission notice a routing can only 15 routers (also is not a repeat by), if the 16th reaches a router, the router will think this sent routing is unreachable.

routing is divided into static routing and dynamic routing, its corresponding routing table referred to as a static route table and dynamic routing table.Static routing table in the system is installed by a network administrator according to the network configuration of the set in advance, after the network structure change by the network administrator manually modify the routing table.Dynamic routing changes over the performance of the network, the router based on the function of the routing protocols provide automatically calculates the best path of data transmission, the resulting dynamic routing table.

according to the Routing algorithm, dynamic Routing Protocol can be divided into the Distance Vector Routing Protocol (short Vector Routing Protocol) and link-state Routing Protocol (Link State Routing Protocol).Distance vector routing protocol based on Bellman - Ford algorithm, mainly are RIP and IGRP (IGRP for Cisco proprietary protocol);Link-state routing protocol based on graph theory is very famous in the Dijkstra algorithm, namely the Shortest preferred Path (Shortest Path First, SPF) algorithm, such as OSPF.In distance vector routing protocols, routers pass some or all of the routing table to adjacent routers;In link-state routing protocols, the router will link-state information to all routers in the same area.According to the position in the autonomous system (AS) a router, routing protocols can be divided into internal Gateway Protocol (Interior Gateway Protocol, IGP) and External Gateway Protocol (External Gateway Protocol, taking, also called inter-domain routing Protocol).Inter-domain routing protocol has two kinds: the external gateway protocol (taking) and the border gateway protocol (BGP).Taking is designed for a simple tree topology, in the treatment of the routing loops and set the routing strategy, has obvious drawbacks, has now been replaced by BGP.Eigrp is a Cisco proprietary protocol,

is a hybrid protocol, it has the characteristics of a distance vector routing protocols, and inherited the advantages of link-state routing protocol.All kinds of routing protocols have different features, suitable for different types of networks.The following are discussed respectively.

a, the static route

static routing table before they begin to choose routing by the network administrator to establish, and can only be changed by the network administrator, so only suitable for network transmission state simple environment.Static route has the following characteristics: there is no need to exchange routing

· static routing, so save the network bandwidth, CPU utilization, and router memory.

· the static routing has higher security.In the use of static routing in the network, all the router to connect to the Internet are required to set its corresponding routing on adjacent routers.Therefore, to some extent improved the security of the network.

, in some cases you must use the static routing, such as DDR, using NAT technology in the network environment.Static route has the following disadvantages:


, managers have to really understand the network topology and properly configured routing.The expansion of

, network performance is poor.If you want to add a network on the Internet, managers must add a routing on all routers.

· configuration cumbersome, especially when you need to cross a few router communication, the routing configuration is more complex.

2, dynamic routing, dynamic routing protocol is divided into

distance vector routing protocols and link-state routing protocols, two protocols have different features, the points mentioned below.

1. Distance vector (DV)

distance vector protocol refers to the protocol USES hop or vector to determine from one device to another device.The rate at which does not consider every hop link.

distance vector routing protocols do not use the normal neighbor relations, with two kinds of methods to learn the change of the topology and routing timeout:

, when a router can't directly receive routing updates from the connection of the router,

, when the router received an update from neighbors, somewhere informs the network topology changes.

in a small network (less than

router, or require less routing updates and computing environment), distance vector routing protocols running quite well.When small network extended to large network, the algorithm is provided by the extremely slow convergence speed, and in the process of its calculation, the network is in a transition state, circulation and cause temporary congestion may occur.Moreover, when the network technology is varied, the underlying link bandwidth each phase at the same time, the distance vector algorithm is a blind eye.

the characteristics of a distance vector routing protocol not only caused the delay of network convergence and the bandwidth consumed.With the increase of the routing table, need to consume more CPU resources, and consumes memory.

2. A link-state routing protocol (LS)

there is no limit to the hop count of link state routing protocols, using the theory of "graphics" or shortest path first algorithm.

link-state routing protocols have shorter convergence time, support VLSM (variable-length subnet mask) and CIDR.

link-state routing protocols between directly connected routing maintenance normal neighbor relations.This allows the routing faster convergence.Link-state routing protocols during the session by exchanging the Hello packets (also called link-state information) to create peer relationship, the relationship is to accelerate the routing convergence.

don't like distance vector routing protocols, update to send the entire routing table.Link-state routing protocol broadcast only update or change the network topology, which makes the smaller updates, save bandwidth and CPU utilization.In addition, if the network does not change, update only in specific period of time (usually a 30 min to 2 h).

3. The analysis of the commonly used dynamic routing protocol


rip (routing information protocol) is used between routers first open standards, is the most widely routing protocol, can get in on all IP routing platform.When using a RIP, a Cisco router can router connection with other vendors.RIP has two versions: RIPv1 and RIPv2, they are all based on the classical distance vector routing algorithm, maximum hop count is 15.

ripv1 is a race Routing (Classful Routing) agreement, not including mask information for Routing, so all the devices on the network must use the same subnet mask, and does not support VLSM.RIPv2 can send subnet mask information, is a species Routing (Classless Routing) protocol, support VLSM.

rip using UDP packet routing updates.Routers updated once every 30 s routing information, if not received adjacent routers in the 180 s response, argues that routing is not available to the router, the router may not arrive.If you do not receive the router in the 240 s after the reply, put on the router's routing information is removed from the routing table.RIP has the following characteristics:

, different vendors can RIP routers connected;

· configuration is simple;, suitable for small networks (less than 15 jump);

· RIPv1 does not support VLSM.Need

· wan bandwidth consumed;

need to consume CPU and memory resources.

RIP algorithm is simple, but the path is large, slow convergence speed in the bandwidth broadcast routing information resource is more, it is suitable for network topology structure is relatively simple and small network data link failure rate is extremely low, in large networks, generally don't use RIP.


Interior Gateway Routing Protocol (Interior Gateway Routing Protocol, IGRP) was developed by Cisco in the 1980 s, is a kind of dynamic, long span, the biggest jump can support 255) Routing Protocol, using metrics (vector) arrive to determine the best route, with a network composed of latency, bandwidth, reliability and load to calculate the optimal Routing, such as high span within it in the same autonomous system, suitable for complex network.Cisco IOS administrators allow routers to IGRP network bandwidth, delay, reliability, and the load of the weighting, to affect the measurement.

as RIP, IGRP USES UDP send routing table entries.Each router updated once every 90 s routing information, if not received within 270 s a router's response, argues that the router unreachable.If do not receive our reply within 630 s, IGRP process will be deleted from the routing table which consists of.

compared to the RIP and IGRP convergence time is longer, but reduce the bandwidth required of the transport routing information, in addition, IGRP no blank bytes of the packet format, so as to improve the efficiency of the IGRP message.But IGRP proprietary for Cisco company, limited to Cisco products.


along with the expanding of network size and the growth of user requirements, the original IGRP has ragged, as a result, Cisco and develop enhanced IGRP, namely, EIGRP.EIGRP USES the same as the IGRP routing algorithm, but it integrates link-state routing protocol and the advantages of distance vector routing protocol, at the same time to join the diffusing update algorithm (DUAL).Eigrp has the following characteristics:


, fast convergence.Fast convergence because of using the diffusing update algorithm, through the backup route in the routing table and implementation, is the minimum cost and time of arrival) network minimum cost (also called suitable successor and feasible successor) routing are stored in the routing table, when the minimum cost routing is not available, a quick switch to the on the minimum cost routing, so as to achieve the goal of fast convergence.

· reduce the bandwidth consumption.EIGRP does not like RIP and IGRP, every once in a while just to exchange routing information at a time, it is only when a certain purpose network routing state changes or routing metric changes, to send routing updates to the adjacent EIGRP router, as a result, the bandwidth required of the routing update - this way it is much less than RIP and EIGRP is triggered (triggered).

, increase the size of the network.For RIP, the network can only be 15 biggest jump (hop), and EIGRP can support 255 biggest jump (hop).

CPU utilization, reduce the router.Routing updates are sent only to the need to know the state change of adjacent routers, with the use of incremental updating, EIGRP than IGRP USES less CPU.

· support variable-length subnet mask.

· IGRP and EIGRP automatically transplantation.IGRP routing can be automatically redistributed to the EIGRP, EIGRP routing will also be automatically redistributed to IGRP.If you like, can also turn off routing redistribution.

· EIGRP supports three types of routing protocols (IP, IPX, AppleTalk).

Address: Room110,No.389 Jinwan Road,Shanghai,China

Email: daisy.dai@ccitel.com
service time: 7x24 hour