Hi Everyone..!
Because I had difficulty finding all IP routing protocols in a single document, my goal with this essay is to present all IP routing protocol methods in this article. I hope that simplifies things for you and helps you comprehend it better. However, if anyone finds any errors or has fresh information, please let me know.
Importance of IP Routing Protocols
IP routing protocols can identify to use the route where IP data travel across to follow multiple networks from its source to the particular destination. Mostly, two forms of IP routing implemented either statically or dynamically to build up an IP routing table in router or layer-3 devices. Static routing is manually-configured by adding fixed route and do not update routing table in case of network changed. Static routing is also more supportive in a stub networks where only single route exists. Dynamic routing, also called adaptive routing, this allows the router to maintain routing table automatically. It has two basic categories of dynamic iGP and eGP routing protocols for IP packet switching networks, as shown in figure 1.
Figure 1 Hierarchical View of Dynamic Routing Protocol
Interior Gateway Protocols (IGP)
IGP reside within an ISP’s network and update routing table inside an autonomous system (AS). The most widely adopted IGPs are Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-IS, Routing Information Protocol (RIP) and Enhanced Interior Gateway Routing Protocol (EIGRP). Whereas Exterior Gateway Protocols (EGPs) are routing protocols which are used in the Internet for exchanging routing information between Autonomous Systems (AS), such as Border Gateway Protocol (BGP).
IPv4 and IPv6 both supports these routing protocols but with some necessary upgradation brought into the IPv6. The detail of each routing protocols is shown in the table 1. The focus of this article is mainly review of routing protocols namely RIP, EIGRP, IS-IS, OSPF and BGP.
Table 1 IP Routing Protocols
RIP Protocol Overview
Routing Information Protocol (RIP) is the oldest routing protocol, originally described in RFC 1058 and standardized in 1988. RIP is known as (IGP), it is typically preferred in small to medium-sized networks, throughout years RIP evolved into three versions RIPv1, RIPv2 and RIPng. The initial RIPv1 version had limited features and supports only classful feature, it could not update routing from subnet masks. The major problem faced in implementing RIPv1 was advertising complete IP block and therefore several IP addresses were going to waste. Hence, RIPv2 was created in 1993 as an enhancement of RIPv1 and published in 1998 in RFC 2453. It resolved many limitations of RIP version 1 in addition to support of the VLSM and CIDR, due to shortage of IPv4 addresses space. RIPv1 and RIPv2 versions did not have a support for IPv6. Another adaptation of RIP was needed to support IPv6 and Routing Information Protocol next generation (RIPng) was developed. In IPv2 has a maximum hop count limitation, although RIPng utilizes the reserved multicast IPv6 address FF02::9 for each RIPng updates.
EIGRP Overview
Enhanced Interior Gateway Routing Protocol (EIGRP) also type of distance vector, it is an upgraded version of IGRP created by Cisco. IPv6 robust EIGRP is known as EIGRP for IPv6 and is like EIGRP utilized with IPv4. EIGRP give features, such as expanded network width of 224s hops in distinction with 15 hops of RIP and straightforward hello technique for neighbor revelation. EIGRP give rapid convergence, which permits speedy routing updates and it can support large scales network. EIGRP for IPv6 gives route filtering and furthermore has a convention subordinate module for IPv4, IPv6. Main disadvantage of EIGRP is its Cisco proprietary routing protocol.
IS-IS Protocol Overview
Intermediate System to Intermediate System (IS-IS) protocol was initially designed by Digital Equipment Corporation (DEC) in late 1989, it was implemented by the International Standards Organization (ISO). The present standard of ISIS protocol is ISO/IEC 10589:2002. IS-IS runs on data link layer similar to Connectionless Network Protocol (CLNP) based network and deals with the upper layer protocols, which allowed IS-IS very easy adoption of IPv6.
OSPF Protocol Overview
Open Shortest Path First (OSPF) is routing protocol designed by IETF, it can be found formally in RFC 1131 which was first standardized in 1998. It is a link-state protocol using Dijkstra's algorithm, also consider as shortest path first (SPF) algorithm, which is same kind of routing algorithm used in IS-IS. Later on, OSPF version 2 was upgraded and defined in RFC 1247, RFC 2178 and RFC 2328. Further OSPF version 3 has several additional features and described in RFC 2740. The OSPFv3 the protocol is extension of OSPFv2, where some necessary changes have been added in LSA in order to support of IPv6 and implementing IPv6 addresses that is defined in RFC 5340. It is open standard (not proprietary) routing protocol implemented by different network vendors, mostly used in enterprises and ISPs networks for their internal IP routing protocol.
OSPF Protocols Operation and Neighbors
OSPF exchange routing information with its neighbor routers consist of state links and interfaces. Link state connection is depiction of that interface association with neighboring routers. The interface contains important data, for example, the IPv4 and IPv6 prefix of the interface, the system subnet cover, cost of neighbors and the types of network. It is connected either point-to-point or point-to-multi-point to build routing tables from other routers in the network. Router's accumulation of link state advertisement information takes placed of link state data. The substance of the data, further applied the “Dijkstra algorithm” calculation in order to create OSPF routing table. Dissimilarity within data and routing table is that the LSA contains an entire accumulation of basic information; whereas routing table holds complete information of the shortest routes that typical known destination through intermediate router interface ports.
OSPF Version 3 (OSPFv3)
OSPFv3 supports IPv6 and IPv4 Address Families (AFs) which is described in RFC 5838, with the addition of address families it also provides several benefits while reducing complexity, enhancing mobility, easing migration, and improving scalability. Since OSPFv3 uses instance numbers to provide multiple options for deployment of OSPF. Service Providers (SPs) can run both IP address families IPv4 and IPv6 into a single OSPFv3 process.
OSPFv3 For IPV6
On Non-Broadcast Multi-Access (NBMA) networks, LSA flooding carried out by the Designated Router (DR) or backup DR (BDR). On point-to-point networks, LSA propagating through an immediately neighbor router’s interface. OSPF utilizes the Hello packet, intermittently sending hello packets out every interface and neighbors established based hello packet. Once routers established its neighbor ship, OPSF router may keep continue to exchange LSA databases, which makes a peering. Although not each neighboring router has a peering. OSPF on communicate with NBMA portions only, in order to diminishes the number of routing update that is being traded on a segment by picking one router as DR and a BDR router. Along with this, these routers will responses to maintain essential routes information on the segment. As opposed to particular flooding OSPF directing updates with each router on the fragment, the DR and BDR routers are passing on the data to alternate routers.
BGP Routing Protocol
Border Gateway Protocol (BGP) is created by IETF in 1989, its original description is given in RFC 1105, released with BGP version 1. The initial purpose of the designed exterior gateway protocol was to provide facility to ISPs for exchanging their routing data with other autonomous systems (AS) on the Internet. Over the time, several incremental upgradations have been made in BGP by IETF to allow ISPs control to routing and policy manipulations through the BGP attributes. Two more versions of BGP have been described, version 2 with RFC 1163 and version 3 in RFC 1267. In early 1990s, due to the World Wide Web (WWW) everyone was getting on the Internet. Due to this, Internet routing tables grew rapidly up to 20,000 routes in 1994, that become too large for the routers to maintain more and more networks on IP addresses. While the evolution of routing table issue has been solved by introducing BGP-4 in 1994 (RFC 1654, later RFC 1771 and then RFC 4271). BGP-4 adds the concept of Classless Inter-Domain Routing CIDR and route summarization / aggregation which support to eliminates the problem of wasted addresses from three classes of IPv6 address. BGP first started to support of IPv6, in 1995 described on RFC 1883, it was improved to RFC2283 in 1998. The new RFC permit BGPv4 extension to support a wide range of IPv4 and IPv6 "address families" including multicast, IPv6, IPv6 multicast, VPNv6, and so on. BGP process sometimes handle single session of Multiprotocol Extensions which is Multiprotocol BGP (MP-BGP).
iBGP and eBGP Overview
BGP is considered important routing protocol by ISP, as it is multiple module protocol and it supports IPv4, IPv6, VPNv4, VPNv6, and MPLS. BGP is also path-vector protocol, uses TCP protocol on port 179 to establish connection between adjacent BGP routers before exchange routes. And working on a reliable transportation protocol that reduce and support BGP to implement update fragmentation, retransmission, acknowledgment, and sequencing. BGP uses two types of adjacency relationships between neighboring routers. When BGP peering, session develop two different distinct ASs is known as external (eBGP) while BGP peering session develop inside known as internal (iBGP). The below figure 2 shows the concept of eBGP and iBGP peering between two different service providers. In the figure 2 an ISP using AS 100 for their iBGP peering within same autonomous system, and service provider establishing an eBGP peering with another ISP through using AS 200 for the exchange of routes with each other.
Figure 2 iBGP and eBGP connections
IBGP Peering Session
By default, BGP follows split-horizon rule that makes it possible to avoid routing loops. BGP learn routes via iBGP session and these routes will not normally be advertised to other iBGP neighbors. To accomplish this task, BGP require to develop iBGP adjacency relation with all other iBGP neighboring node in form of a full mesh. This approach is not recommended for the ISP production environment. Logically, as the connection of routers rises, so does quantity of BGP required to establish a full mesh of iBGP peer’s increments. That can create complex with a large number of iBGP peers, and overall potential of the network may has affected on bases of maintaining of huge collection of iBGP adjacency relation that might overuse significant resources of network, such as hardware, bandwidth. CPU etc.
BGP Route Reflectors (BGP-RR)
It is quite difficult for the service provider to implement logical full-mesh topology for iBGP built configuration of routers. Due to the increasing numbers of iBGP peering sessions there is the possibilities of less scalability. Therefore, in a large network, the full-mesh design for iBGP can be a key challenge for ISPs. BGP implement alternative possible to solve the complexity of multiple peering configurations. Route reflectors and BGP confederations.
Route Reflectors (RRs) is support to aggregate route information and therefore keep various alternative routes inside PoPs or inside a specific geographic area. Just the best route, as picked by the best way choice procedure of the RR, it permitted to be appropriated inside the AS. A RR is a BGP speaker that reflects iBGP updates from RR-clients to different clients and also different RRs, furthermore flood iBGP updates from RRs to clients (however not to other RRs). RRs can be sorted out in progressive form and full-mesh is just required on the top level of the topology. However, routing loops can't happen within a AS, to avoid single points of failures, redundant RRs can be designed. Shown in figure 3.
Figure 3
Conclusion
This Article provided the overview of different IPv4 and IPv6 routing protocols including IGP and EGP is also presented. The fundamental concept of this post is to overview of IP routing protocols and also identify the new features and benefits bring with IPv6 for world., in order to have a better understand the changes made to these protocols to incorporate IPv6 support.
References:
Murhammer, Martin W., Orcun Atakan, Stefan Bretz, Larry R. Pugh, Kazunari Suzuki, and David H. Wood. “TCP / IP Tutorial and Technical Overview, Book,” IBM Corp. Int. Tech. Support Organzation, vol. 1, p. 1004, 2006.
Mak, Terrence, Peter YK Cheung, Kai-Pui Lam, and Wayne Luk. “Adaptive Routing in Network-on-Chips Using a Dynamic-Programming Network,” vol. 58, no. 8, pp. 3701–3716, 2011.
Cisco Systems Inc., “Cisco Networking Academy's Introduction to Routing Dyna-mically.” March 2014. [Online]. Available:https://www.ciscopress.com/articles/article.asp?p=2180210&seqNum=7. [accessed 07 Junly 2021].
Comparative analysis of various routing protocols." International Journal of modern Engineering Research., vol. 2015–April, no. March, pp. 315–319, 2015.
Norvor, Gideon Evans, Michael Asante, and Anokye Acheampong. "Dynamic Routing Implementation Decision between OSPFv3 and IS–IS in IPv6 Networks." Communications on Applied Electronics, vol. 6, no. 3, pp. 25–34, 2016.
Lindem, Acee, Sina Mirtorabi, Michael Barnes, Abhay Roy, and Rahul Aggarwal. "Support of address families in OSPFv3". No. RFC 5838, IETF, April 2010. [Online]. Available: https://tools.ietf.org/rfc/rfc5838.txt.
Cisco Systems Inc., “OSPFv3 Support for Address Families,” White paper, 2011. [Online].Available:https://www.cisco.com/c/en/us/products/collateral/ios-nx-os-software/open-shortest-path-first-ospf/whitepaper_c11-668030.pdf. [accessed 08 Junly 2021].
Gamess, Eric, Jesús Urbáez, and Arturo Palacios. "A visual teaching and learning environment for the open shortest path first routing protocol." Proceedings of the Ninth LACCEI. Medellin, Colombia. Lat. Am. Caribb. Conf., pp. 1–10, 2011.
Lougheed, K., and Y. Rekhter. "Border Gateway Protocol (BGP)”. Network Information Center. No. RFC 1105. June 1989. [Online]. Available: https://tools.ietf.org/pdf/rfc1105.pdf.
Fuller, Vince, Tony Li, Jessica Yu, and Kannan Varadhan. “Classless inter-domain routing (CIDR): an address assignment and aggregation strategy”. Network Information Center No. RFC 1519. September 1993. [Online]. Available: https://tools.ietf.org/pdf/rfc1519.pdf.
Teare, Diane, Bob Vachon, and Rick Graziani. “Implementing Cisco IP routing (ROUTE) foundation learning guide”. Foundation Learning Guide. Cisco Press, 2015.
Talk Mro. “EBGP-iBGP”, Illustration of eBGP and iBGP connections. Wikimedia Commons.24 May 2009. [Online]. Available: https://commons.wikimedia.org/wiki/File:EBGP-iBGP.svg.
Bates, Tony, Enke Chen, and Ravi Chandra. “BGP route reflection: An alternative to full mesh internal bgp (IBGP)”. Network Working Group. No. RFC 4456. April 2006. [Online]. Available: https://tools.ietf.org/rfc/rfc4456.txt.
Traina, Paul, Danny McPherson, and John Scudder. “Autonomous system confederations for BGP”. Network Working Group. No. RFC 5065 August 2007. [Online]. Available: https://tools.ietf.org/rfc/rfc5065.txt.
T. Griffin et G. Wilfong, “On the Correctness of IBGP Configuration,” Computer Communication Review, vol. 32, pp. 17– 29, October 2002.
I hope it will be beneficial for you, however in case of any query, please do let me know in comments.
