View source for Routing

{{Short description|Process of selecting paths in a data communications network}}

'''Routing''' is the process of selecting a path for traffic in a [[Network theory|network]] or between or across multiple networks. Broadly, routing is performed in many types of networks, including [[Circuit switching|circuit-switched network]]s, such as the public switched telephone network (PSTN), and [[Computer network|computer network]]s, such as the [[Internet]].

In packet switching networks, routing is the higher-level decision making that directs [[Network packet|network packet]]s from their source toward their destination through intermediate [[Node (networking)|network nodes]] by specific packet forwarding mechanisms. [[Packet forwarding]] is the transit of network packets from one [[Engineering:Network interface controller|network interface]] to another.  Intermediate nodes are typically network hardware devices such as [[Engineering:Router (computing)|routers]], [[Engineering:Gateway (telecommunications)|gateways]], [[Firewall (computing)|firewalls]], or [[Network switch|switches]]. General-purpose [[Computer|computer]]s also forward packets and perform routing, although they have no specially optimized hardware for the task.

The routing process usually directs forwarding on the basis of [[Routing table|routing table]]s. Routing tables maintain a record of the routes to various network destinations. Routing tables may be specified by an administrator, learned by observing network traffic or built with the assistance of [[Routing protocol|routing protocol]]s.

Routing, in a narrower sense of the term, often refers to [[IP routing]] and is contrasted with [[Bridging (networking)|bridging]]. IP routing assumes that [[Network address|network address]]es are structured and that similar addresses imply proximity within the network. Structured addresses allow a single routing table entry to represent the route to a group of devices.  In large networks, structured addressing (routing, in the narrow sense) outperforms unstructured addressing (bridging). Routing has become the dominant form of addressing on the Internet. Bridging is still widely used within [[Engineering:Local area network|local area network]]s.

==Delivery schemes==
{{routing scheme}}<!--{{routing scheme}} links to this section-->
Routing schemes differ in how they deliver messages:
{{Routing scheme/details}}

Unicast is the dominant form of message delivery on the Internet.  This article focuses on unicast routing algorithms.

==Topology distribution==
With [[Static routing|static routing]], small networks may use manually configured routing tables. Larger networks have complex [[Network topology|topologies]] that can change rapidly, making the manual construction of routing tables unfeasible. Nevertheless, most of the public switched telephone network (PSTN) uses pre-computed routing tables, with fallback routes if the most direct route becomes blocked (see [[Routing in the PSTN|routing in the PSTN]]).

[[Dynamic routing]] attempts to solve this problem by constructing routing tables automatically, based on information carried by [[Routing protocol|routing protocol]]s, allowing the network to act nearly autonomously in avoiding network failures and blockages. Dynamic routing dominates the Internet. Examples of dynamic-routing protocols and algorithms include [[Routing Information Protocol]] (RIP), [[Open Shortest Path First]] (OSPF) and [[Enhanced Interior Gateway Routing Protocol]] (EIGRP).

===Distance vector algorithms===
{{main|Distance-vector routing protocol}}
Distance vector algorithms use the [[Bellman–Ford algorithm]]. This approach assigns a ''cost'' number to each of the links between each node in the network. Nodes send information from point A to point B via the path that results in the lowest ''total cost'' (i.e. the sum of the costs of the links between the nodes used).

When a node first starts, it only knows of its immediate neighbors and the direct cost involved in reaching them. (This information — the list of destinations, the total cost to each, and the ''next hop'' to send data to get there — makes up the routing table, or ''distance table''.) Each node, on a regular basis, sends to each neighbor node its own current assessment of the total cost to get to all the destinations it knows of. The neighboring nodes examine this information and compare it to what they already know; anything that represents an improvement on what they already have, they insert in their own table. Over time, all the nodes in the network discover the best next hop and total cost for all destinations.

When a network node goes down, any nodes that used it as their next hop discard the entry and convey the updated routing information to all adjacent nodes, which in turn repeat the process. Eventually, all the nodes in the network receive the updates and discover new paths to all the destinations that do not involve the down node.

===Link-state algorithms===
{{main|Link-state routing protocol}}
When applying link-state algorithms, a [[Graph (discrete mathematics)|graphical map]] of the network is the fundamental data used for each node. To produce its map, each node floods the entire network with information about the other nodes it can connect to.  Each node then independently assembles this information into a map. Using this map, each router independently determines the least-cost path from itself to every other node using a standard [[Shortest path problem|shortest paths]] algorithm such as [[Dijkstra's algorithm]]. The result is a [[Tree (graph theory)|tree graph]] rooted at the current node, such that the path through the tree from the root to any other node is the least-cost path to that node. This tree then serves to construct the routing table, which specifies the best next hop to get from the current node to any other node.

===Optimized Link State Routing algorithm===
{{main|Optimized Link State Routing Protocol}}
A link-state routing algorithm optimized for [[Engineering:Mobile ad hoc network|mobile ad hoc network]]s is the optimized Link State Routing Protocol (OLSR).<ref>RFC 3626</ref> OLSR is proactive; it uses Hello and Topology Control (TC) messages to discover and disseminate link-state information through the mobile ad hoc network. Using Hello messages, each node discovers 2-hop neighbor information and elects a set of ''multipoint relays'' (MPRs). MPRs distinguish OLSR from other link-state routing protocols.

===Path-vector protocol===
{{main|Path-vector routing protocol}}
Distance vector and link-state routing are both intra-domain routing protocols. They are used inside an [[Autonomous system (Internet)|autonomous system]], but not between autonomous systems. Both of these routing protocols become intractable in large networks and cannot be used in [[Inter-domain|inter-domain]] routing. Distance vector routing is subject to instability if there are more than a few hops in the domain. Link state routing needs significant resources to calculate routing tables. It also creates heavy traffic due to flooding.

Path-vector routing is used for inter-domain routing. It is similar to distance vector routing. Path-vector routing assumes that one node (there can be many) in each autonomous system acts on behalf of the entire autonomous system. This node is called the ''speaker node.'' The speaker node creates a routing table and advertises it to neighboring speaker nodes in neighboring autonomous systems. The idea is the same as distance vector routing except that only speaker nodes in each autonomous system can communicate with each other. The speaker node advertises the path, not the metric, of the nodes in its autonomous system or other autonomous systems.

The path-vector routing algorithm is similar to the distance vector algorithm in the sense that each border router advertises the destinations it can reach to its neighboring router. However, instead of advertising networks in terms of a destination and the distance to that destination, networks are advertised as destination addresses and path descriptions to reach those destinations. The path, expressed in terms of the domains (or confederations) traversed so far, is carried in a special path attribute that records the sequence of routing domains through which the reachability information has passed. A route is defined as a pairing between a destination and the attributes of the path to that destination, thus the name, path-vector routing; The routers receive a vector that contains paths to a set of destinations.<ref>{{IETF RFC|1322}}</ref>

==Path selection==
Path selection involves applying a [[Metrics (networking)|routing metric]] to multiple routes to select (or predict) the best route. Most routing algorithms use only one network path at a time. [[Multipath routing]] and specifically [[Equal-cost multi-path routing|equal-cost multi-path routing]] techniques enable the use of multiple alternative paths.

In computer networking, the metric is computed by a routing algorithm, and can cover information such as [[Bandwidth (computing)|bandwidth]], [[Network delay|network delay]], hop count, path cost, load, [[Maximum transmission unit|maximum transmission unit]], reliability, and communication cost.<ref>{{citation |url=http://rainer.baumann.info/public/tik262.pdf |first1=Rainer |last1=Baumann |first2=Simon |last2=Heimlicher |first3=Mario |last3=Strasser |first4=Andreas |last4=Weibel |title=A Survey on Routing Metrics |date=February 10, 2007 |access-date=2020-05-04}}</ref> The routing table stores only the best possible routes, while [[Link-state routing protocol|link-state]] or topological databases may store all other information as well.

In case of overlapping or equal routes, algorithms consider the following elements in priority order to decide which routes to install into the routing table:
#''Prefix length'': A matching route table entry with a longer subnet mask is always preferred as it specifies the destination more exactly.
#''[[Metrics (networking)|Metric]]'': When comparing routes learned via the same routing protocol, a lower metric is preferred. Metrics cannot be compared between routes learned from different routing protocols.
#''[[Administrative distance]]'': When comparing route table entries from different sources such as different routing protocols and static configuration, a lower administrative distance indicates a more reliable source and thus a preferred route.

Because a routing metric is specific to a given routing protocol, multi-protocol routers must use some external [[Heuristic|heuristic]] to select between routes learned from different routing protocols. [[Company:Cisco|Cisco]] routers, for example, attribute a value known as the [[Administrative distance|administrative distance]] to each route, where smaller administrative distances indicate routes learned from a protocol assumed to be more reliable.

A local administrator can set up host-specific routes that provide more control over network usage, permits testing, and better overall security. This is useful for debugging network connections or routing tables.

In some small systems, a single central device decides ahead of time the complete path of every packet. In some other small systems, whichever edge device injects a packet into the network decides ahead of time the complete path of that particular packet. In either case, the route-planning device needs to know a lot of information about what devices are connected to the network and how they are connected to each other. Once it has this information, it can use an algorithm such as [[A* search algorithm]] to find the best path.

In high-speed systems, there are so many packets transmitted every second that it is infeasible for a single device to calculate the complete path for each and every packet. Early high-speed systems dealt with this with [[Circuit switching|circuit switching]] by setting up a path once for the first packet between some source and some destination; later packets between that same source and that same destination continue to follow the same path without recalculating until the circuit teardown. Later high-speed systems inject packets into the network without any one device ever calculating a complete path for packets.

In large systems, there are so many connections between devices, and those connections change so frequently, that it is infeasible for any one device to even know how all the devices are connected to each other, much less calculate a complete path through them. Such systems generally use [[Hop (networking)#Next hop|next-hop]] routing.

Most systems use a deterministic [[Dynamic routing|dynamic routing]] algorithm. When a device chooses a path to a particular final destination, that device always chooses the same path to that destination until it receives information that makes it think some other path is better.

A few routing algorithms do not use a deterministic algorithm to find the best link for a packet to get from its original source to its final destination. Instead, to avoid congestion hot spots in packet systems, a few algorithms use a randomized algorithm—Valiant's paradigm—that routes a path to a randomly picked intermediate destination, and from there to its true final destination.<ref>{{citation |author1=Michael Mitzenmacher |author2=Andréa W. Richa |author3=Ramesh Sitaraman |url=http://www.eecs.harvard.edu/~michaelm/postscripts/handbook2001.pdf |title=The Power of Two Random Choices: A Survey of Techniques and Results |section=Randomized Protocols for Circuit Routing |page=34 |url-status=live |archive-url=https://web.archive.org/web/20231213082735/https://www.eecs.harvard.edu/~michaelm/postscripts/handbook2001.pdf |archive-date= Dec 13, 2023 }}</ref><ref>{{citation |author=Stefan Haas |url=http://inspirehep.net/record/887357/files/cer-002474543.pdf |website=INSPIRE |title=The IEEE 1355 Standard: Developments, Performance and Application in High Energy Physics |date=1998 |page=15 |quote=To eliminate network hot spots, ... a two phase routing algorithm. This involves every packet being first sent to a randomly chosen intermediate destination; from the intermediate destination it is forwarded to its final destination. This algorithm, referred to as Universal Routing, is designed to maximize capacity and minimize delay under conditions of heavy load. |url-status=live |archive-url=https://web.archive.org/web/20190516092754/http://inspirehep.net/record/887357/files/cer-002474543.pdf |archive-date= May 16, 2019 }}</ref> In many early telephone switches, a randomizer was often used to select the start of a path through a multistage switching fabric.

Depending on the application for which path selection is performed, different metrics can be used. For example, for web requests one can use minimum latency paths to minimize web page load time, or for bulk data transfers one can choose the least utilized path to balance load across the network and increase throughput. A popular path selection objective is to reduce the average completion times of traffic flows and the total network bandwidth consumption. Recently, a path selection metric was proposed that computes the total number of bytes scheduled on the edges per path as selection metric.<ref>{{Cite web |title= Poster Abstract: Minimizing Flow Completion Times using Adaptive Routing over Inter-Datacenter Wide Area Networks |first1= M. |last1=Noormohammadpour |first2=C. S. |last2=Raghavendra |via=ResearchGate |doi=10.1109/INFCOMW.2018.8406853 |date= Apr 2018 |url=https://www.researchgate.net/publication/323411017}}</ref> An empirical analysis of several path selection metrics, including this new proposal, has been made available.<ref>{{Cite web |title= Minimizing Flow Completion Times using Adaptive Routing over Inter-Datacenter Wide Area Networks |first1=M |last1=Noormohammadpour |first2=C. S. |last2=Raghavendra |via=ResearchGate |doi=10.13140/RG.2.2.36009.90720 |doi-access=free |date= Apr 2018 |url=https://www.researchgate.net/publication/323723167}}</ref>

==Multiple agents==
In some networks, routing is complicated by the fact that no single entity is responsible for selecting paths; instead, multiple entities are involved in selecting paths or even parts of a single path. Complications or inefficiency can result if these entities choose paths to optimize their own objectives, which may conflict with the objectives of other participants.

A classic example involves traffic in a road system, in which each driver picks a path that minimizes their travel time. With such routing, the [[Nash equilibrium|equilibrium]] routes can be longer than optimal for all drivers. In particular, [[Braess's paradox]] shows that adding a new road can ''lengthen'' travel times for all drivers.

In a single-agent model used, for example, for routing [[Engineering:Automated guided vehicle|automated guided vehicle]]s (AGVs) on a terminal, reservations are made for each vehicle to prevent simultaneous use of the same part of an infrastructure. This approach is also referred to as context-aware routing.<ref>{{cite web |first1=Jonne |last1=Zutt |first2=Arjan J.C. |last2=van Gemund |first3=Mathijs M. |last3=de Weerdt |first4=Cees |last4=Witteveen |date=2010 |url=http://www.st.ewi.tudelft.nl/~mathijs/publications/intinfra09.pdf |title=Dealing with Uncertainty in Operational Transport Planning |url-status=dead |archive-url=https://web.archive.org/web/20170922001342/http://www.st.ewi.tudelft.nl/~mathijs/publications/intinfra09.pdf |archive-date= Sep 22, 2017 }} In R.R. Negenborn and Z. Lukszo and H. Hellendoorn (Eds.) Intelligent Infrastructures, Ch. 14, pp. 355–382. Springer.</ref> 

The Internet is partitioned into [[Autonomous system (Internet)|autonomous systems]] (ASs) such as internet service providers (ISPs), each of which controls routes involving its network. Routing occurs at multiple levels. First, AS-level paths are selected via the [[Border Gateway Protocol|BGP]] protocol that produces a sequence of ASs through which packets flow. Each AS may have multiple paths, offered by neighboring ASs, from which to choose. These routing decisions often correlate with business relationships with these neighboring ASs,<ref>Matthew Caesar and [[Biography:Jennifer Rexford|Jennifer Rexford]]. "[http://www.cs.princeton.edu/~jrex/papers/policies.pdf BGP routing policies in ISP networks]". IEEE Network Magazine, special issue on Interdomain Routing, Nov/Dec 2005.</ref> which may be unrelated to path quality or latency. Second, once an AS-level path has been selected, there are often multiple corresponding router-level paths to choose from. This is due, in part, because two ISPs may be connected through multiple connections. In choosing the single router-level path, it is common practice for each ISP to employ [[Hot-potato routing|hot-potato routing]]: sending traffic along the path that minimizes the distance through the ISP's own network—even if that path lengthens the total distance to the destination.

For example, consider two ISPs, ''A'' and ''B''. Each has a presence in {{wipe|New York City}}, connected by a fast link with latency {{val|5|ul=ms}}—and each has a presence in [[Place:London|London]] connected by a 5&nbsp;ms link. Suppose both ISPs have trans-Atlantic links that connect their two networks, but ''A''<nowiki/>'s link has latency 100&nbsp;ms and ''B''<nowiki/>'s has latency 120&nbsp;ms. When routing a message from a source in ''A''<nowiki/>'s London network to a destination in ''B''<nowiki/>'s New York network, ''A'' may choose to immediately send the message to ''B'' in London. This saves ''A'' the work of sending it along an expensive trans-Atlantic link, but causes the message to experience latency 125&nbsp;ms when the other route would have been 20&nbsp;ms faster.

Additionally, a similar routing challenge can be observed in cellular networks, where different packets are destined for various endpoints, and each link exhibits varying spectral efficiency. In this context, the selection of the ''optimal'' path involves considering latency and packet error rate. To address this, multiple independent entities, one for each base station, play a crucial role in path selection while striving to optimize overall network performance.<ref>Shahaf Yamin and Haim H. Permuter. "[https://www.sciencedirect.com/science/article/pii/S1570870523002676 Multi-agent reinforcement learning for network routing in integrated access backhaul networks]". ''Ad Hoc Networks'', Volume 153, 2024, 103347, {{ISSN|1570-8705}}, {{doi|10.1016/j.adhoc.2023.103347}}.</ref>

A 2003 measurement study of Internet routes found that, between pairs of neighboring ISPs, more than 30% of paths have inflated latency due to hot-potato routing, with 5% of paths being delayed by at least 12&nbsp;ms. Inflation due to AS-level path selection, while substantial, was attributed primarily to BGP's lack of a mechanism to directly optimize for latency, rather than to selfish routing policies. It was also suggested that, were an appropriate mechanism in place, ISPs would be willing to cooperate to reduce latency rather than use hot-potato routing.<ref>Neil Spring, Ratul Mahajan, and Thomas Anderson. "[http://www.cs.washington.edu/research/networking/rocketfuel/papers/sigcomm2003.pdf Quantifying the Causes of Path Inflation]". Proc. SIGCOMM 2003.</ref> Such a mechanism was later published by the same authors, first for the case of two ISPs<ref>Ratul Mahajan, David Wetherall, and Thomas Anderson. "[http://research.microsoft.com/en-us/um/people/ratul/papers/nsdi2005-nexit.pdf Negotiation-Based Routing Between Neighboring ISPs]". Proc. NSDI 2005.</ref> and then for the global case.<ref>Ratul Mahajan, David Wetherall, and Thomas Anderson. [http://research.microsoft.com/en-us/um/people/ratul/papers/nsdi2007-wiser.pdf Mutually Controlled Routing with Independent ISPs]. Proc. NSDI 2007.</ref>

==Route analytics==
As the Internet and IP networks have become [[Mission critical|mission critical]] business tools, there has been increased interest in techniques and methods to monitor the routing posture of networks. Incorrect routing or routing issues cause undesirable performance degradation, [[Route flapping|flapping]] or downtime. Monitoring routing in a network is achieved using [[Route analytics|route analytics]] tools and techniques.<ref>{{Cite book|last1=Santhi|first1=P.|title=An Efficient Security Way of Authentication and Pair wise Key Distribution with Mobile Sinks in Wireless Sensor Networks|last2=Ahmed|first2=Md Shakeel|last3=Mehertaj|first3=Sk|last4=Manohar|first4=T. Bharath|citeseerx=10.1.1.392.151}}</ref>

==Centralized routing==
In networks where a logically centralized control is available over the forwarding state, for example, using [[Software-defined networking|software-defined networking]], routing techniques can be used that aim to optimize global and network-wide performance metrics. This has been used by large internet companies that operate many data centers in different geographical locations attached using private optical links, examples of which include Microsoft's Global WAN,<ref>{{cite web |url=https://azure.microsoft.com/en-us/blog/how-microsoft-builds-its-fast-and-reliable-global-network/ |title=How Microsoft builds its fast and reliable global network |last=Khalidi |first=Yousef |date=March 15, 2017}}</ref> Facebook's Express Backbone,<ref>{{cite web |url=https://code.fb.com/data-center-engineering/building-express-backbone-facebook-s-new-long-haul-network/ |title=Building Express Backbone: Facebook's new long-haul network |date=May 1, 2017}}</ref> and Google's B4.<ref>{{cite web |url=https://www.networkcomputing.com/networking/inside-googles-software-defined-network/512240144 |title=Inside Google's Software-Defined Network |date=May 14, 2017}}</ref> 

Global performance metrics to optimize include maximizing network utilization, minimizing traffic flow completion times, maximizing the traffic delivered prior to specific deadlines and reducing the completion times of flows.<ref>{{cite journal|last1=Noormohammadpour|first1=Mohammad|last2=Raghavendra|first2=Cauligi|title=Datacenter Traffic Control: Understanding Techniques and Tradeoffs|journal=IEEE Communications Surveys and Tutorials|date=16 July 2018|volume=20|issue=2|pages=1492–1525|doi=10.1109/COMST.2017.2782753|arxiv=1712.03530|s2cid=28143006}}</ref> Work on the later over private WAN discusses modeling routing as a graph optimization problem by pushing all the queuing to the end-points. The authors also propose a heuristic to solve the problem efficiently while sacrificing negligible performance.<ref>{{cite journal|last1=Noormohammadpour|first1=Mohammad|last2=Srivastava|first2=Ajitesh|last3=Raghavendra|first3=Cauligi|title=On Minimizing the Completion Times of Long Flows over Inter-Datacenter WAN|journal=IEEE Communications Letters|volume=22|issue=12|pages=2475–2478|date=2018|url=https://www.researchgate.net/publication/328008697|doi=10.1109/LCOMM.2018.2872980|arxiv=1810.00169|bibcode=2018arXiv181000169N|s2cid=52898719}}</ref>

==See also==
{{div col|colwidth=20em}}
* [[Black hole (networking)]]
* [[Collective routing]]
* [[Deflection routing]]
* [[Edge disjoint shortest pair algorithm]]
* [[Flood search routing]]
* [[Fuzzy routing]]
* [[Geographic routing]]
* [[Heuristic routing]]
* [[Path computation element]] (PCE)
* [[Policy-based routing]]
* Wormhole routing
* [[Small-world routing]]
* [[Turn restriction routing]]
{{Div col end}}

==References==
{{Reflist}}

==Further reading==
{{Refbegin}}
* {{cite book | author=Ash, Gerald | title=Dynamic Routing in Telecommunication Networks | publisher=McGraw–Hill | year=1997 | isbn=978-0-07-006414-0}}
* {{cite book |author1=Doyle, Jeff  |author2=Carroll, Jennifer  |name-list-style=amp | title=Routing TCP/IP, Volume I, Second Ed. | publisher=Cisco Press | year=2005 | isbn=978-1-58705-202-6}}[http://www.ciscopress.com/title/1587052024 Ciscopress] {{ISBN|1-58705-202-4}}
* {{cite book | author1=Doyle, Jeff | author2=Carroll, Jennifer | name-list-style=amp | title=Routing TCP/IP, Volume II | publisher=Cisco Press | year=2001 | isbn=978-1-57870-089-9 | url=https://archive.org/details/routingtcpip00jeff }}[http://www.ciscopress.com/title/1578700892 Ciscopress] {{ISBN|1-57870-089-2}}
* {{cite book | author=Huitema, Christian | title=Routing in the Internet, Second Ed. | publisher=Prentice–Hall | year=2000 | isbn=978-0-321-22735-5}}
* {{cite book |author1=Kurose, James E.  |author2=Ross, Keith W.  |name-list-style=amp | title=Computer Networking, Third Ed. | publisher=Benjamin/Cummings | year=2004 | isbn=978-0-321-22735-5}}
* {{cite book |author1=Medhi, Deepankar  |author2=Ramasamy, Karthikeyan  |name-list-style=amp | title=Network Routing: Algorithms, Protocols, and Architectures | publisher=Morgan Kaufmann | year=2007 | isbn=978-0-12-088588-6}}
{{Refend}}

==External links==
* [https://web.archive.org/web/20061209093728/http://wiki.uni.lu/secan-lab/Count-To-Infinity+Problem.html Count-To-Infinity Problem]
* {{cite web |url=http://wwwlehre.dhbw-stuttgart.de/~schulte/htme/55024.htm#HDR3 |title=Stability Features |archive-url=https://web.archive.org/web/20150925085557/http://wwwlehre.dhbw-stuttgart.de/~schulte/htme/55024.htm#HDR3 |archive-date=2015-09-25}}, ways of avoiding the count-to-infinity problem
* [https://web.archive.org/web/20070216114435/http://www.cisco.com/web/about/ciscoitatwork/case_studies/routing.html Cisco IT Case Studies] about Routing and Switching
* {{cite web |url=http://www.eventhelix.com/Realtimemantra/Networking/ip_routing.htm |title=IP Routing and Subnets |access-date=2018-04-28 |website=www.eventhelix.com}}

{{Routing protocols}}

[[Category:Routing| ]]
[[Category:Internet architecture]]

{{Sourceattribution|Routing}}