samedi 12 septembre 2009

The Basics of Routing Protocols

The goal of routing protocols is to learn which routes are available on the company network, to build routing tables and make routing decisions. Some routing protocols are the most common RIP, IGRP, EIGRP, OSPF, IS-IS and BGP. There are two main types of routing protocols although many different routing protocols defined with these two types. Link state protocols and distance vector includes the primary types. Protocols distance vector advertise their routing table to all neighbors directly connected to regular and frequent intervals with bandwidth and are slow to converge. When a route becomes unavailable, all router tables must be updated with new information. The problem is with each router having to announce that new information to its neighbors, it is time for all routers have an accurate and current network. Protocols distance vector using masks subnet fixed length which are not scalable. Routing Protocols Link State Advertising for updates only when they occur, using bandwidth more efficiently. Routers do not advertise the routing table which makes the convergence faster. The routing protocol to flood the network with listings of state links to all routers capita by region to try to converge the network a new information highway. The incremental change is all that is advertised to all routers as an update multicast LSA. They use masks of variable length subnet, which are scalable and use to address more effectively.

Interior Gateway Routing Protocol (IGRP)

Interior Gateway Routing Protocol is a routing protocol distance vector developed by Cisco Systems for routing multiple protocols through the small and medium-sized Cisco networks. It is property that requires you to use Cisco routers. This contrasts with the IP RIP and IPX RIP, designed for multi-vendor networks. IGRP will route IP, IPX, AppleTalk and DECnet which makes it very versatile for many clients running different protocols. It is a bit more scalable than RIP because it supports a number of jumps of 100, said that every 90 seconds and uses a composite of five different indicators to choose the best path destination. Note that since the advertising IGRP less often, it uses less bandwidth than RIP but converges much slower because it is 90 seconds before IGRP routers are aware of changes in network topology. IGRP does not recognize the assignment of different autonomous systems and summarizes at the network boundaries of class. In addition, there is the possibility of balancing the traffic across equal cost paths or uneven metric.

Features

· Distance Vector

· Roads IP, IPX, DECnet, Appletalk

· Routing table announcements every 90 seconds

· Metric: bandwidth, delay, reliability, load, MTU Size

· Hop Count: 100

Fixed Length Subnet Masks

· Summarization of Network Address class

· Load Balancing of 6 equal or unequal Cost Paths (IOS 11.0)

· Metric Calculation = destination path minimum BW * Delay (USEC)

· Split Horizon

· Timers: Invalid timer (270 sec), Flush timer (630 sec), holddown timer (280 sec)

Enhanced Interior Gateway Routing Protocol (EIGRP)

Enhanced Interior Gateway Routing Protocol is a hybrid routing protocol developed by Cisco Systems for many routing protocols on a corporate network, Cisco. It has characteristics of both distance vector routing protocols and link state routing protocols. It is property that requires you to use Cisco routers. EIGRP will forward the same protocols as roads IGRP (IP, IPX, DECnet and AppleTalk) and use the same measures as composite IGRP to select the best path destination. In addition, there is the possibility of balancing the traffic across equal cost paths or uneven metric. A summary is automatic network address to a class but can be configured to summarize the limits of sub-network as well. Redistribution between IGRP and EIGRP is automatic. There is support for a number of sections 255 and masks of variable length subnet.

Convergence

Convergence with EIGRP is faster because it uses an algorithm called algorithm updated to double or double, which is initiated when a router that detects a particular route is not available. The router queries its neighbors in search of a possible successor. Which is defined as a neighbor with a least cost route to a particular destination that does not cause routing loops. EIGRP will update its routing table with the new route and the associated metric. The route changes are announced only to affected routers when changes occur. Who uses more bandwidth efficient than routing protocols distance vector.

Autonomous Systems

EIGRP does not recognize the transfer of various autonomous systems which are processes that run in the same administrative domain routing. Assigning different autonomous system numbers is not the definition of a spine as with OSPF. With IGRP, and EIGRP is used to change the route redistribution, filtering and summarization points.

Features

· Advanced Distance Vector

· Roads IP, IPX, DECnet, Appletalk

Classifieds · Routing: Route partial when changes

· Metrics: Bandwidth, delay, reliability, load, MTU Size

· Hop Count: 255

· Variable Length Subnet Masks

· Summarization of Network Address class or sub-network boundary

· Load Balancing of 6 equal or unequal Cost Paths (IOS 11.0)

· Timers: Active Time (180 sec)

· Metric Calculation = destination path minimum BW * Delay (ms) * 256

· Split Horizon

· LSA Multicast Address: 224.0.0.10

Open Shortest Path First (OSPF)

Open Shortest Path First protocol is a true link state developed as an open standard for IP routing through several major provider networks. A protocol link state advertisements send link state to all the people connected to the same region to provide route information. Each OSPF router running, when it starts, will send hello packets to all directly connected OSPF routers. Hello packets contain information such as timers router, router ID and subnet mask network. If the routers agree on the information they become OSPF neighbors. Once the routers become neighbors they establish adjacencies by exchanging databases link state. Routers on point-to-point and point to multipoint (as specified with the parameter type interface OSPF) automatically establishes adjacencies. With interfaces OSPF routers configured as a broadcast (Ethernet) and NBMA (Frame Relay) will use a router that sets the designated neighborhoods.

Domains

OSPF uses a hierarchy with the affected areas that connect to a backbone core router. Each zone is defined by one or more routers that have established neighborhoods. OSPF has defined the backbone area 0, buffer zones, the not-so-short areas and areas completely cut off. Zone 0 is constructed with a group of routers connected to a designated office or WAN links between multiple offices. It is better to have all area 0 routers connected to a full mesh using an Ethernet segment to a basic office. This provides for high performance and prevents partitioning of space if a router connection fails. Zone 0 is a transit area for trafficking in enclosed areas. All traffic must cross-domain Route through area 0 first. Stub areas use a default route to route traffic destined to an external network such as EIGRP router from the border zone does not send or receive external connections. Inter-area and intra-area routing is as usual. Areas totally Stubby Cisco is a specification that uses a default route for inter-area and external destinations. The ABR does not send or receive external or inter-area's LSA. The Not-So-Stubby Area ABR will advertise external routes with type 7 LSA. External routes are not received this type of area. Inter-area and intra-area routing is as usual. Defines internal OSPF routers, routers, backbone routers, area border (ABR) and autonomous system boundary routers (ASBR). Routers internal to a specific area. Routers, area border have interfaces that are assigned to more than one area as Zone 0 and Zone 10. A border router autonomous system has interfaces to OSPF and assigned a different routing protocol like EIGRP or BGP. A virtual link is used when a region has no direct connection to Zone 0. A virtual link is established between a border router of area to an area that is not connected to area 0, router and a border zone in an area that is connected to area 0. Zone design is to consider the location of offices and traffic flows across the enterprise. It is important to be able to summarize addresses for many offices by area and minimize broadcast traffic.

Convergence

Rapid convergence is achieved with the SPF (Dijkstra) algorithm determines a shortest path from source to destination. The routing table is constructed from running SPF determines that all roads in the neighboring routers. Given that each OSPF router has a copy of the basic topology and routing table to its particular area, route changes are detected faster than protocols distance vector and other routes are determined.

Designated Router

Broadcast Networks such as Ethernet and Non-Broadcast Multi-access networks such as Frame Relay have a designated router (DR) and backup designated router (BDR) are elected. Designated routers establish adjacencies with all routers on a network segment. This is to reduce emissions from all routers periodically send hello packets to its neighbors. The DR sends multicast packets to all routers that it has established with the neighborhoods. If the DR fails, BDR is sending the multicast router specific. Each router is assigned an identifier router, which is the highest IP address assigned to a working interface. OSPF uses the router ID (RID) for all routing processes.

Features

· Link State

· IP Routes

Classifieds · Routing: Route partial when changes

· Metric: Cost composition of each router to your destination (100000000 / interface speed)

· Hop Count: None (limited by the network)

· Variable Length Subnet Masks

· Summarization of Network Address class or sub-network boundary

· Load Balancing Across 4 equal cost paths

Router * Types: Internal, Backbone, ABR, ASBR

* Types of surface: Backbone, Stubby, Not-So-Stubby, Totally Stubby

LSA * Types: intra-area (1,2) inter-area (3.4), external (5.7)

· Timers: Hello Interval and Dead Interval (different for different types of network)

· LSA Multicast Address: 224.0.0.5 and 224.0.0.6 (DR / BDR) Do not filter!

* Types of interface: Point to Point, Broadcast, Non-Broadcast, Point to Multipoint, Loopback

Integrated IS-IS

Integrated Intermediate System - Intermediate System Routing Protocol is a protocol link state similar to OSPF is used with great enterprise and ISP customers. An intermediate system is a router and IS-IS is the routing protocol that routes packets between intermediate systems. IS-IS uses a database link state and runs the Dijkstra SPF algorithm to select the shortest path. Neighbor routers point to point and point to multipoint link adjacencies by sending Hello packets and exchange databases link state. IS-IS routers on broadcast networks NBMA select a designated router, which establishes adjacencies with all neighboring routers on the network. The designated router and each router will establish neighbor adjacency with all neighboring routers by advertising link multicast state in the network itself. This is different from OSPF, which establishes adjacencies between R & D and each router only neighbor. IS-IS uses a hierarchical structure of the area with the Level 1 and Level 2 types of router. Level 1 routers are similar to intra-area OSPF routers, which have no direct connections outside its region. Level 2 routers set up by the back area which connects the different areas similar to OSPF area 0. With IS-IS a router can be a L1/L2 router is like an OSPF Area Border Router (ABR), which has links with the region and the dorsal region. The difference with the IS-IS is that the links between routers include the border area and not the router. Each IS-IS router must be assigned an address that is unique to this area routing. An address format is used, which consists of an area of identification and an identification system. The region is the number assigned area and the system identifier is a MAC address of a router's interfaces. There is support for masks subnet variable length, which is standard with all protocols, link state. Note that IS-IS affects the process of routing to an interface instead of a network.

Features

· Link State

· Roads IP, CLNS

· Routing Advertisements: Partial when routing changes occur

· Metric: variable cost (cost per default 10 assigned to each interface)

· Hop Count: None (limited by the network)

· Variable Length Subnet Masks

· Summarization of Network Address class or sub-network boundary

· Load Balancing of 6 equal cost paths

· Timers: Interval Hello, Hello Multiplier

* Types of surface: Hierarchical Topology similar to OSPF

* Types Router: Level 1 and Level 2

* Types LSP: Internal L1 and L2, L2 outside

· Designated Router election, no BDR

Border Gateway Protocol (BGP)

Border Gateway Protocol is an exterior gateway protocol, which is different from interior gateway protocols discussed so far. The distinction is important because the term "autonomous system is used somewhat differently with protocols such as EIGRP is that with BGP. External routing protocols such as BGP route between autonomous systems, which are assigned a specific number AS. AS numbers can be assigned to an office with one or more BGP routers. The BGP routing table consists of destination IP addresses, an AS-Path associated to reach this destination and next hop address of the router. The AS-Path is a collection of AS numbers, which represent each of the offices involved in routing packets. This contrasts with EIGRP uses autonomous systems as well. The difference is their autonomous systems refer to a logical grouping of routers within the same administrative system. A network can configure EIGRP many autonomous systems. They are all managed by the company to define an abstract way, redistribution and filtering. BGP is used a lot by Internet Service Providers (ISPs) and companies of large companies that have Internet connections with dual-hosted routers single or dual-homed to the same or different Internet Service Providers. BGP will route packets through a network of ISPs, which is a separate routing domain is managed by them. The ISP has its own assigned the number that is assigned by InterNIC. New customers can either request an assignment as their duties from the ISP or InterNIC. A unique number that the assignment is provided for customers when they log in using BGP. There are 10 attributes that defined a particular order or sequence, which uses BGP as parameters to determine the best path to a destination. Companies with a single circuit connection to an ISP will implement a default route on their router, which forwards all packets that are destined to an external network. BGP router will redistribute routing information (peering) with all routers on the network IGP (EIGRP, RIP, OSPF, etc.) that involve exchange of full routing tables. Once it is finished, incremental updates are sent with topology changes. Each BGP router can be configured to filter routing emissions with road maps rather than send / receive the entire Internet routing table.

BGP routing table Components

· Destination IP Address / Subnet Mask

· AS-Path

· Next Hop IP Address

The book Network Planning and Design Guide is available on Amazon.com

Aucun commentaire:

Enregistrer un commentaire