Fact-checked by Grok 2 weeks ago

Distance-vector routing protocol

A distance-vector routing protocol is a class of routing algorithms used in packet-switched networks, where each router maintains a table of distances (typically measured in hops) to all possible destinations and periodically shares this information, known as a distance vector, with its directly connected neighbors to compute optimal paths. These protocols operate on the principle of the distributed Bellman-Ford algorithm, in which routers iteratively relax path costs based on updates from neighbors until convergence is achieved. In practice, a router updates its routing table by selecting the path with the minimum total distance advertised by neighbors, adding the cost of the link to that neighbor. Key mechanisms in distance-vector protocols include periodic broadcasts of the entire —often every 30 seconds—to neighboring routers, as well as triggered updates when significant changes occur to reduce time. To mitigate issues like routing loops, techniques such as split horizon (omitting routes learned from a neighbor when advertising back to them) and poison reverse (advertising infinite for such routes) are commonly employed. The metric is usually a hop count, with a maximum value (e.g., 15 in ) to prevent infinite loops and limit the protocol's scope to smaller networks. Prominent examples include the (RIP), standardized in RFC 1058, which uses hop count as its sole and is suitable for small to medium-sized networks. Other implementations, such as Cisco's (IGRP) and its successor Enhanced IGRP (EIGRP)—often described as an advanced distance-vector protocol—incorporate composite metrics like and delay for more accurate path selection. While simple to configure and effective in stable environments, distance-vector protocols suffer from slow convergence in dynamic topologies, susceptibility to count-to-infinity problems during failures, and limited scalability beyond 15 hops, making them less ideal for large-scale or rapidly changing networks compared to link-state alternatives.

Fundamentals

Definition and Key Characteristics

A distance-vector routing protocol is a class of distributed routing algorithms used in computer networks where each router maintains a routing table that records the shortest distance (or cost) to every possible destination in the network and periodically advertises this table, known as a distance vector, to its directly connected neighboring routers. These protocols operate without requiring routers to possess a complete map of the global network topology, relying instead solely on information exchanged with immediate neighbors to iteratively compute and refine paths. Key characteristics of distance-vector protocols include their implementation as a distributed variant of the Bellman-Ford algorithm, which enables routers to calculate optimal paths by relaxing distance estimates through neighbor updates. They typically employ metrics such as hop count—where each link contributes a cost of 1—or composite metrics incorporating factors like and delay to evaluate route quality. Updates are disseminated periodically, for example every 30 seconds in the case of the (), allowing the network to adapt to changes without centralized coordination. These protocols offer advantages in simplicity and low computational overhead, making them straightforward to implement on resource-constrained devices such as early-generation routers or systems. However, they incur high usage due to the periodic broadcasting of entire tables to neighbors and are susceptible to loops in the absence of additional mitigation techniques like split horizon.

Underlying Algorithm

The serves as the mathematical foundation for distance-vector routing protocols, providing a relaxation-based approach to compute shortest paths from a single source to all destinations in a weighted . Originally developed to handle graphs with possible negative edge weights—as long as no negative cycles exist—the algorithm iteratively relaxes path estimates until optimal distances are achieved. In routing contexts, this is adapted to distributed environments without negative weights, where link costs represent positive metrics such as hop counts or delays, ensuring to loop-free shortest paths under stable conditions. The core principle stems from Bellman's dynamic programming framework, where the shortest path distance to a destination satisfies the . For a router i and destination network N, the distance d_i(N) is updated via relaxation as d_i(N) = \min_{j \in N(i)} \left[ c_{ij} + d_j(N) \right], where N(i) denotes the set of neighbors of i, and c_{ij} is the cost of the direct from i to neighbor j. This embodies the idea that the best path to N passes through one of the immediate neighbors, combining the local cost with the neighbor's reported distance to N. The original formulation appears in Bellman's work on problems, emphasizing successive approximations to solve the functional equations iteratively. In the distributed setting of distance-vector protocols, routers perform this relaxation asynchronously based on periodic exchanges of distance vectors from neighbors. Each router initializes its distances (e.g., infinite for unreachable destinations except itself at zero) and iteratively updates them upon receiving neighbor reports, propagating changes until no further relaxations are possible, at which point the system converges to the global shortest paths. This process mirrors Ford's computational procedure for network flows, which scans and relaxes inequalities across all edges in repeated passes—up to |V|-1 times for |V| vertices in the centralized case—though the distributed version relies on rather than global . Key assumptions underpin the algorithm's applicability in routing: updates occur asynchronously without a global clock, link costs are non-negative to prevent oscillations from negative cycles, and the network graph is connected to ensure finite distances for all destinations. These conditions, drawn from the original theoretical developments, enable the algorithm to function in dynamic but networks, with guaranteed in the absence of changes.

Operational Mechanics

Vector Exchange Process

In distance-vector routing protocols, neighbor discovery occurs passively through the reception of periodic update messages broadcast or multicast on directly connected network interfaces, allowing routers to identify adjacent peers without dedicated hello mechanisms. Upon startup, a router may actively solicit initial routing information by sending a general request message to all potential neighbors, prompting responses with complete routing tables. For instance, in the Routing Information Protocol (RIP), these exchanges utilize UDP port 520, with broadcasts to 255.255.255.255 in RIP version 1 or multicasts to 224.0.0.9 in version 2. Routers transmit distance vectors—summaries of their routing tables—periodically at fixed intervals to all discovered neighbors, ensuring consistent propagation of network reachability information. These updates typically include the full set of known routes, though some implementations support triggered updates sent immediately upon detecting changes, such as link failures or metric alterations, to accelerate while mitigating excessive traffic through random delays of 1 to 5 seconds. In , periodic updates occur every 30 seconds, and triggered updates are limited to modified entries only. The message format for distance vectors is compact and structured to carry multiple route entries efficiently within datagrams. Each message begins with a command field (e.g., request or response), followed by a identifier, and then up to 25 route entries per packet. A typical entry specifies the destination , a representing the (such as hop count, ranging from 1 to 15, with 16 indicating unreachable), and optional fields like next-hop router or subnet mask for enhanced precision. In version 2, additional fields include a route tag for route origin classification and an header in the first entry for . Upon receiving a distance vector message, the router validates it by confirming the source IP is from a directly connected and the port is correct (e.g., 520 for ), then checks for if enabled—such as simple or in version 2—discarding invalid packets to prevent unauthorized updates. Valid entries are parsed, with the reported incremented by the to the sending , and temporarily stored for potential integration into the local , though the actual route selection occurs separately. This process ensures only reliable exchanges contribute to the protocol's operation.

Route Computation and Selection

In distance-vector routing protocols, routers process incoming routing vectors by first adding the cost of the incoming interface to each advertised in the vector, then comparing the resulting to the current entry in their own for the same destination. If the new is lower than the existing one, or if the update is from the same neighbor (even if the new is equal or higher), the router updates the table entry with the new and resets the route's timeout . This computation occurs upon receipt of vectors periodically exchanged from neighboring routers. The next-hop determination follows directly from this process: the router selects the neighbor that provided the update with the minimum total distance as the next hop for forwarding packets to that destination, recording this neighbor's address in the entry. For directly connected networks, no next hop is required, as traffic is sent directly over the . Routing tables are maintained through periodic aging mechanisms to remove stale or invalid routes; for instance, in the (), entries are marked as unreachable (metric set to infinity, or 16 hops) after 180 seconds without an and fully flushed after an additional 120-second garbage-collection period. Expired entries trigger the removal of the route and may prompt triggered updates to inform neighbors. In cases of ties—such as multiple paths offering the same minimum distance—routers may employ tie-breaking rules like to prioritize routes from preferred protocols or sources, or enable load balancing across equal-cost paths to distribute traffic. For example, implementations support load balancing over up to four equal-cost paths by default, enhancing utilization without altering the primary metric-based selection.

Historical Development

Origins and Evolution

The distance-vector routing protocol traces its origins to the foundational work on shortest-path algorithms in the mid-20th century, particularly the Bellman-Ford algorithm, which was independently developed by Lester Ford Jr. in 1956 and Richard Bellman in 1958. This algorithm provided a method for computing optimal paths in networks by iteratively relaxing edge weights, forming the theoretical basis for distributed routing where nodes exchange distance estimates rather than full topology information. In 1969, these concepts were adapted for practical network use in the , the precursor to the modern , where the initial routing mechanism employed a distributed version of Bellman-Ford to manage dynamic path selection across packet-switched nodes. One of the earliest implementations occurred in 1973 with ' PARC Universal Packet (PUP) protocol, which utilized distance-vector techniques to enable among heterogeneous local area networks at PARC. This adoption highlighted the protocol's simplicity and suitability for emerging distributed systems, paving the way for broader application in computer networking. The protocol evolved significantly in the late 1980s and 1990s to address limitations in scalability and addressing efficiency. The (RIP), a canonical distance-vector implementation, was standardized in RFC 1058 in 1988, establishing hop count as the primary metric for route selection in interior gateway routing. By the late 1990s, RIP version 2 (RIPv2), detailed in RFC 2453 in 1998, introduced enhancements such as support for variable-length subnet masking (VLSM) to enable (CIDR), along with authentication mechanisms and updates for improved security and efficiency. These advancements facilitated the Internet's explosive growth, influencing variants like the (BGP), a path-vector extension of distance-vector principles that uses autonomous system paths instead of simple distances to manage inter-domain routing at scale.

Major Protocol Implementations

The (RIP) is one of the earliest and most widely implemented distance-vector protocols, standardized in RFC 1058 for version 1, which operates in a classful manner and uses broadcast updates to exchange routing tables every 30 seconds, limiting networks to a maximum of 15 hops to prevent infinite loops. Version 2 of RIP, defined in RFC 2453, extends this by supporting classless routing through subnet masks, switching to multicast updates on address 224.0.0.9 for efficiency, and incorporating authentication as specified in RFC 2082 to secure exchanges against unauthorized modifications. The (IGRP), developed by Systems in the mid-1980s as a enhancement to , employs a composite that incorporates , delay, load, reliability, and (MTU) size, with and delay weighted by default to better reflect real-world . Unlike 's hop count, IGRP uses a 32-bit value, allowing for larger networks up to 255 hops, and broadcasts updates every 90 seconds while supporting multiple protocols including , IPX, and . Enhanced Interior Gateway Routing Protocol (EIGRP), also Cisco proprietary and introduced as an evolution of IGRP, functions as a hybrid protocol blending distance-vector principles with link-state features through the Diffusing Update Algorithm (DUAL), which ensures loop-free paths by maintaining a table and computing feasible successors as backup routes. EIGRP enables partial, triggered updates rather than periodic full-table broadcasts, reducing usage, and uses the same composite as IGRP but with finer via a 32-bit or 64-bit representation. The protocol's reliance on reliable via RTP for acknowledgments further distinguishes it from pure distance-vector approaches. Other notable implementations include open-source routing daemons like GateD, which supports alongside other protocols in a unified for systems, facilitating multi-protocol environments. Successors such as Zebra and its fork provide modular implementations of distance-vector protocols including versions 1 and 2, emphasizing kernel integration and redistribution capabilities for diverse network topologies. Additionally, the (BGP), while primarily path-vector for inter-domain routing, extends distance-vector concepts by propagating AS-path attributes to detect loops and enforce policies, as outlined in 4271.

Limitations

Count to Infinity Issue

The count to infinity problem in protocols occurs when routers engaged in a repeatedly increment the metric to an unreachable destination based on each other's outdated information, causing the metric to rise slowly until it reaches a predefined maximum value treated as . This issue highlights a key limitation of the protocol's reliance on periodic exchanges of vectors from neighbors. The problem typically arises after a change, such as a link failure, where updates propagate asynchronously and routers fail to immediately invalidate routes learned from affected neighbors. For instance, in a simple with routers A, B, and C connected in sequence, where A originally reaches destination X directly with 1 (B has 2 via A, C has 3 via B), if the A-X link fails and A advertises to B, but before B processes this, C advertises its stale route with 3 to X. B then updates to 4 via C and advertises this back to C, who updates to 5 via B. This cycle continues, with metrics incrementing alternately (e.g., B: 4, C: 5; then B: 6, C: 7; and so on) until reaching . In the worst case, this can take O(N) update cycles, where N is the . The impacts include prolonged convergence times, where the network remains in an unstable state for many update cycles, and excessive consumption from the frequent exchange of these escalating route advertisements. Additionally, during the looping period, packets destined for the unreachable network may circulate indefinitely among the affected routers until their time-to-live () expires, effectively dropping them and causing temporary service disruption. To detect and resolve the issue, distance-vector protocols define a finite maximum metric—such as 16 in the ()—beyond which a destination is declared unreachable, triggering route removal and halting the increment cycle. This threshold prevents indefinite counting while signaling the network failure state to all routers.

Convergence Delays and Loops

In distance-vector routing protocols, convergence delays refer to the time required for all routers in a network to update their routing tables and agree on optimal paths following a topology change, such as a link failure or addition. These delays arise primarily from the protocol's reliance on periodic exchange of distance vectors among neighboring routers, typically every 30 seconds in protocols like , which propagates information incrementally across the network. In large networks, this process can take several minutes, as updates must traverse multiple hops before stabilizing, potentially leading to suboptimal routing during the interim period. Temporary routing loops often form during this update propagation phase, where inconsistent information causes packets to cycle briefly among routers before the tables converge. For instance, if router A initially routes traffic to destination X via router B, and B then updates to route via C after a failure, but C temporarily routes back through A due to outdated information from B, a short-lived loop (A → B → C → A) may occur, forwarding packets indefinitely until the next update cycle resolves the inconsistency. Such loops are transient but can disrupt traffic flow, especially in networks with high latency or frequent changes. Several factors exacerbate these convergence delays and loop formations, including network size, which limits effective propagation (e.g., RIP's 15-hop ), update frequency that introduces lags, and metric sensitivity where small changes trigger widespread recomputations. The protocol's "routing by rumor" nature—wherein routers depend solely on neighbors' indirect reports without global knowledge—further amplifies spread, slowing overall stabilization. In extreme cases, these delays can manifest as prolonged issues akin to the count-to-infinity problem, where metrics increment unboundedly. Additionally, temporary loops during can have implications, such as amplifying DDoS attacks by repeatedly generating response floods in looped paths involving vulnerable middleboxes.

Mitigation Strategies

Split Horizon Technique

The split horizon technique is a fundamental mitigation strategy in distance-vector routing protocols designed to prevent routing loops by enforcing a simple rule: a router does not advertise a route back to the from which it originally learned that route. This approach addresses the vulnerability in basic distance-vector exchanges where mutual misinformation between adjacent routers can perpetuate incorrect paths. In operation, when a router prepares and sends its routing update to a specific neighbor via an interface, it omits any routes for which that neighbor is the designated next hop, effectively "splitting" the horizon of information flow to avoid echoing back learned data. For instance, if router A learns a path to network X from router B, A will exclude the route to X in its updates sent back to B, ensuring that B does not mistakenly adopt A's potentially outdated or looped information as an alternative path. This omission occurs selectively per interface during periodic or triggered updates, integrating seamlessly with the standard vector exchange process without requiring additional fields in protocol messages. The primary benefits of split horizon include the immediate prevention of two-router loops, where adjacent nodes might otherwise reinforce each other's erroneous routes, thus enhancing stability in simple topologies. Additionally, by excluding unnecessary routes from updates, it reduces the size of messages and conserves , particularly in networks with frequent advertisements. These advantages make split horizon a lightweight enhancement that promotes faster for bilateral failures without complicating the core distance-vector computation. Despite its effectiveness against pairwise loops, split horizon has notable limitations, as it cannot prevent routing loops involving three or more routers, where information can propagate indirectly through alternate paths and still lead to count-to-infinity scenarios. For example, in a triangular , a failure may allow distant nodes to advertise inflated distances that bypass the split horizon restriction, prolonging convergence delays. This constraint underscores the need for complementary techniques in larger or more complex networks to fully address the inherent challenges of distance-vector protocols.

Poison Reverse and Timers

Poison reverse is an enhancement to the split horizon technique in distance-vector routing protocols, where routes learned from a neighboring router are not simply omitted in updates sent back to that , but instead advertised with an infinite metric value to explicitly indicate unreachability. In protocols like , this infinite metric is set to , ensuring that the recognizes the route as and avoids using it to propagate potentially looping information. This method provides stronger protection against routing loops compared to basic split horizon by actively poisoning the route, though it slightly increases the size of update messages. In distance-vector protocols like , routes are managed using to detect and handle invalidation. The invalidation is set to 180 seconds; if no update is received within this period, the route is marked unreachable by setting its to (16 in ). At this point, a garbage-collection of 120 seconds is started, during which the route remains in the table with the infinite and is advertised as such to neighbors, allowing time for . If a valid update with a finite arrives before the garbage-collection expires, the route is updated accordingly and are reset; otherwise, the route is removed upon expiration of the garbage-collection . Some implementations add a hold-down mechanism to ignore potentially erroneous updates during this period for further stabilization, though this is not specified in RFC 1058 for . Triggered updates complement these techniques by enabling immediate propagation of routing changes rather than waiting for periodic broadcasts, thereby accelerating in dynamic environments. In , when a route changes, the router sends an update containing only the affected entries almost immediately, subject to a random delay of 1 to 5 seconds to avoid overload from synchronized transmissions. Split horizon and poison reverse rules are applied to these updates to maintain prevention. Together, poison reverse, the invalidation and garbage-collection timers, and triggered updates enhance loop detection and speed in distance-vector protocols by proactively disseminating unreachability information and managing route lifetimes. However, in implementations with additional hold-down features, they can introduce temporary unreachability if a legitimate alternate route is advertised during the suppression period and ignored, potentially delaying recovery in certain failure scenarios.

Applications and Examples

RIP Protocol Illustration

The (RIP) serves as a foundational example of a distance-vector routing protocol, employing periodic vector exchanges to propagate routing tables among routers using counts as the distance . operates over on port 520, with routers broadcasting or multicasting their entire routing tables every 30 seconds to neighboring devices. An entry in a router's table becomes invalid after 180 seconds without an update, after which a garbage collection timer of 120 seconds triggers its removal from the table and announcements to neighbors. The protocol's messages consist of a fixed header followed by up to 25 routing entries; the header includes a 1-octet command field (e.g., 1 for request, 2 for response), a 1-octet field (1 for RIP1, 2 for RIP2), and reserved fields, while each entry comprises a 2-octet address family identifier (2 for ), a 4-octet , and a 1-octet (padded to 4 octets), with RIP2 adding fields for subnet mask, next hop, and route tag. Metrics range from 1 to 15 s, with 16 denoting infinity for unreachable networks. Upon , a RIP router initializes its by sending request messages (command=1) as broadcasts on all directly connected interfaces, soliciting complete routing tables from . Neighboring routers respond with their full tables (command=2), which the booting router processes entry by entry: for each destination, it selects the route with the lowest plus one to itself, discarding any with metric 16, and installs valid routes with the responding neighbor as next hop. If no responses arrive within a timeout period, connected networks are added directly with metric 1; the router then begins periodic updates. When a link failure or increase occurs, triggers an immediate update to notify neighbors, with a random 1- to 5-second delay to avoid storms. The affected route's is incremented accordingly in the update; if a route becomes unreachable, the router sets its metric to 16 () and propagates this value, preventing further use until . In , configuration begins in global configuration mode with the "router rip" command to enter router configuration mode, followed by "" statements specifying networks (e.g., "network 192.168.1.0") to enable on all interfaces belonging to those networks, advertising routes learned or connected therein. Interfaces can be fine-tuned with commands like "ip rip send version 2" under interface mode to control message versions sent or received.

Network Scenario Demonstration

To illustrate the operation of a distance-vector routing protocol, consider a simple consisting of three routers, A, B, and C, connected in a linear : A—B—C. Each router is attached to its own local —network X at A, Y at B, and Z at C—with link costs of 1 each. Initially, all links are operational, and routers exchange periodic updates containing their distance vectors (routing tables) with directly connected neighbors, using hop count as the metric. In normal operation, the network converges quickly to optimal paths. Router A directly knows network X at 0 and learns network Y at 1 via B, and network Z at 2 via B. Router B knows Y at 0, X at 1 via A, and Z at 1 via C. Router C knows Z at 0, Y at 1 via B, and X at 2 via B. These are established through iterative update exchanges: for example, C initially advertises Z at 0 to B, which updates its and propagates to A with an added hop, resulting in the converged tables shown below.
RouterDestinationDistanceNext Hop
AX0Local
AY1B
AZ2B
BX1A
BY0Local
BZ1C
CX2B
CY1B
CZ0Local
Packets flow efficiently in this state; for instance, a packet from X to traverses A → B → C in 2 . Now simulate a link where the B—C connection goes down, making Z unreachable from B and A. Without mitigations, B detects the and sets its to Z as (e.g., 16 in protocols like ), advertising this to A. However, A still holds Z at 2 (learned via B) and advertises it back to B in the next . B then incorrectly updates its to Z as 3 (A's 2 + 1), advertising this to A, which updates to 4 via B. This ping-pong effect continues, with distances incrementing stepwise (5, 6, etc.) until reaching , a process known as counting to that delays and can create temporary loops. Applying the split horizon mitigation changes the outcome. Split horizon prevents a router from advertising a route back to the neighbor from which it was learned. Thus, A never advertises Z back to B (since A learned Z from B), and after the failure, when B advertises for Z to A, A immediately updates its table to without relearning a looped path from B. B and A both converge to Z as unreachable within one or two update cycles, avoiding the incremental counting. The final converged tables reflect this:
RouterDestinationDistanceNext Hop
AX0Local
AY1B
AZ-
BX1A
BY0Local
BZ-
CX-
CY-
CZ0Local
Post-convergence, packets from X or Y to Z are dropped at A or B, respectively, as no valid route exists, ensuring no persistent loops. C, isolated, cannot reach X or Y. This demonstrates how split horizon accelerates in distance-vector protocols during failures.

References

  1. [1]
    Chapter: Routing Overview - Cisco
    Jul 16, 2018 · The Routing Information Protocol (RIP) is a distance-vector protocol that uses a hop count as its metric. RIP is widely used for routing ...About Layer 3 Unicast Routing · Routing Algorithms · Summary of Layer 3 Unicast...
  2. [2]
    RFC 1058 - Routing Information Protocol - IETF Datatracker
    RFC 1058 describes the Routing Information Protocol (RIP), a protocol for exchanging routing information among gateways and hosts, based on the 'routed' ...
  3. [3]
    Introduction to EIGRP - Cisco
    This document describes the Interior Gateway Routing Protocol (IGRP) suite of routing protocols designed and developed by Cisco Systems.
  4. [4]
  5. [5]
    3.4 Routing - Computer Networks: A Systems Approach
    The starting assumption for distance-vector routing is that each node knows the cost of the link to each of its directly connected neighbors. These costs may be ...
  6. [6]
    What Is Distance-Vector Routing? - JumpCloud
    Mar 7, 2025 · One basic method is distance-vector routing, which calculates the shortest path (or fewest hops) between the source and destination. Routers ...
  7. [7]
  8. [8]
    [PDF] Formal Verification of Standards for Distance Vector Routing ...
    The aim of this paper is to study how methods of automated reasoning can be used to prove properties of network routing protocols. We carry out three case.
  9. [9]
  10. [10]
    [PDF] THE BELLMAN-FORD ALGORITHM AND “DISTRIBUTED ...
    The Bellman-Ford algorithm finds shortest paths in a graph, based on Bellman's equations and Ford's 1956 work.
  11. [11]
    [PDF] richard bellman - on a routing problem
    The purpose of this paper is to show that the functional equation technique of dynamic programming, [1, 2], combined with the concept of approximation in policy.
  12. [12]
    Network Flow Theory. - RAND
    Year: 1956; Document Number: P-923. Citation. RAND Style Manual. Ford, L. R., Network Flow Theory. RAND Corporation, P-923, 1956. As of November 7, 2025: https ...Missing: PDF | Show results with:PDF
  13. [13]
    RFC 1058: Routing Information Protocol
    - Periodically, send a routing update to every neighbor. The update is a set of messages that contain all of the information from the routing table. It ...
  14. [14]
  15. [15]
    IP Routing: RIP Configuration Guide - Cisco
    Feb 16, 2016 · Routing Information Protocol (RIP) is a routing protocol using a distance-vector algorithm, using broadcast UDP data packets to exchange  ...
  16. [16]
    RIP - Cisco Community
    Routing Information Protocol (RIP) is a distance-vector routing protocol. RIP sends the complete routing table out to all active interfaces every 30 seconds ...
  17. [17]
    Bellman-Ford Algorithm - an overview | ScienceDirect Topics
    The Bellman-Ford algorithm is widely used in network routing protocols, notably in distance vector routing and the Routing Information Protocol (RIP), where it ...Introduction · Algorithm Description and... · Applications and Distributed...
  18. [18]
    [PDF] 6.02 Lecture 23: A brief history of the Internet - MIT OpenCourseWare
    Dec 10, 2012 · vector routing; time sync. • 1973: Ethernet (MAC protocol inspired by Aloha -- CSMA). • 1973-74: Xerox PUP (used distance-vector protocol).
  19. [19]
    RFC 2453: RIP Version 2
    This document specifies an extension of the Routing Information Protocol (RIP), as defined in [1], to expand the amount of useful information carried in RIP ...
  20. [20]
    Operation of BGP > Introduction to BGP - Cisco Press
    Nov 19, 2016 · BGP is a vector protocol, although called a path vector rather than distance vector because it sees the route to a destination as a path ...
  21. [21]
    RFC 2453 - RIP Version 2 - IETF Datatracker
    This document specifies an extension of the Routing Information Protocol (RIP), as defined in [1], to expand the amount of useful information carried in RIP ...
  22. [22]
    RFC 2082 - RIP-2 MD5 Authentication - IETF Datatracker
    RIP-2 MD5 Authentication · RFC - Proposed Standard January 1997. Report errata. Obsoleted by RFC 4822. Was draft-ietf-ripv2-md5 (ripv2 WG) · 03 · RFC 2082.
  23. [23]
    An Introduction to IGRP - Cisco
    This document introduces Interior Gateway Routing Protocol (IGRP). It has two purposes. One is to form an introduction to the IGRP technology.Goals for IGRP · Summary of IGRP · Packet Routing · Reception of Routing Updates
  24. [24]
    Cisco's Enhanced Interior Gateway Routing Protocol (EIGRP)
    EIGRP is a routing protocol based on Distance Vector technology. The specific algorithm used is called DUAL, a Diffusing Update Algorithm.
  25. [25]
    Gated - freesoft.org
    Gated, maintained by the Merit Gated Consortium, is a UNIX implementation of all routing protocols currently in common use, including RIP versions 1 and 2, ...
  26. [26]
    RFC 4271 - A Border Gateway Protocol 4 (BGP-4) - IETF Datatracker
    This document discusses the Border Gateway Protocol (BGP), which is an inter-Autonomous System routing protocol.Missing: vector | Show results with:vector
  27. [27]
    [PDF] Routing Loops as Mega Amplifiers for DNS-based DDoS Attacks
    Transient loops appear when topology changes and the routing protocol has not yet converged. Such loops do not require man- ual intervention to be resolved ...
  28. [28]
  29. [29]
  30. [30]
    IP Routing: RIP Configuration Guide, Cisco IOS XE Everest 16.6
    Aug 15, 2017 · Cisco supports two modes of authentication on an interface on which RIP is enabled: plain-text authentication and message digest algorithm 5 ( ...
  31. [31]
    RIP Distance Vector Routing Protocol - NetworkLessons.com
    RIP (Routing Information Protocol) is a simple distance vector routing protocol. In this lesson, I explain RIP and how to configure it.Missing: authoritative sources
  32. [32]
    What is split horizon and how does it work? - TechTarget
    Aug 11, 2021 · Split horizon is a method used by distance vector protocols to prevent network routing loops. ... The following figure shows an example of three ...Missing: ABC | Show results with:ABC
  33. [33]
    Route Poisoning and Count to infinity problem in Routing
    Sep 30, 2021 · The Count to Infinity problem is caused by routing loops. Route poisoning uses an infinite metric to signal a failed route.Missing: RFC | Show results with:RFC