Introduction: Beyond the Standards

In the evolving landscape of network routing, two main philosophies emerged: the simple, rumor-based Distance Vector (like RIP) and the complex, map-based Link-State (like OSPF). While these open-standard protocols provided a common language for routers from different manufacturers, they also represented a set of compromises. A major player in the networking industry, Cisco Systems, saw an opportunity to build something better, tailored to their own hardware and a vision of a more intelligent network.

This led to the development of proprietary protocols. A proprietary protocol is designed and owned by a single company, and was historically only usable on that company's equipment. While this limits interoperability, it gives the company the freedom to innovate beyond the constraints of public standards committees.

This page explores two of Cisco's most significant contributions to routing: the original Interior Gateway Routing Protocol (IGRP), created to solve RIP's critical flaws, and its revolutionary successor, the Enhanced Interior Gateway Routing Protocol (EIGRP), a sophisticated hybrid that combines the best of both routing worlds.

IGRP: A Smarter Distance Vector

The Interior Gateway Routing Protocol (IGRP) was Cisco's first major step in creating a superior routing protocol in the 1980s. At its heart, it was still a distance-vector protocol, meaning it followed the same "routing by rumor" principle as RIP. However, it was designed from the ground up to address RIP's most crippling weaknesses.

Fixing RIP’s Flaws

IGRP was created to solve two key problems:

• The Naive Metric: RIP's sole reliance on hop count was its biggest flaw, treating a slow 56kbps satellite link the same as a high-speed fiber link. IGRP introduced a far more sophisticated way to measure path quality.
• The Limited Scale: RIP's maximum hop count of 15 made it unsuitable for any network of moderate size. IGRP was designed for larger enterprise networks, with a much higher default maximum hop count (100, configurable up to 255).

The IGRP Composite Metric

The crowning achievement of IGRP was its advanced composite metric. Instead of just counting routers, IGRP could consider multiple characteristics of a path to calculate a more realistic "cost." It took into account five factors, known as K-values:

• Bandwidth (K1): The slowest bandwidth along the entire path. Like a highway's capacity being limited by its narrowest section.
• Delay (K3): The cumulative delay (latency) of all links in the path.
• Load (K2): The traffic utilization of the link (how busy it is).
• Reliability (K4): A measure of the link's error rate.
• MTU (K5): The Maximum Transmission Unit, or the largest packet size a link can handle. (This was not actually part of the metric formula but was tracked).

While all five values could be used, the default configuration for IGRP (and later EIGRP) only used Bandwidth and Delay for stability reasons. The formula was complex, but its goal was simple: to prefer paths with higher bandwidth and lower cumulative delay.

The Downfall of IGRP: A Classful Dead End

Despite its advanced metric, IGRP suffered from the same fatal flaw as RIPv1: it was a classful routing protocol. It did not send subnet mask information in its routing updates. This meant IGRP could not support VLSM and was fundamentally incompatible with the architecture of modern, efficiently addressed networks.

As the internet rapidly moved towards a classless model to conserve IP addresses, IGRP became obsolete. Cisco needed a successor that retained IGRP's intelligent metric but embraced the new classless world. That successor was EIGRP.

EIGRP: The Best of Both Worlds

The Enhanced Interior Gateway Routing Protocol (EIGRP) is Cisco's masterpiece of routing protocol design. It is often called a hybrid protocol because while it is built on the foundations of a distance-vector algorithm, it incorporates advanced features and rapid convergence mechanisms typically associated with link-state protocols. It is sometimes referred to as an "advanced distance-vector" protocol.

EIGRP solves all of IGRP's and RIP's problems while offering unparalleled performance and efficiency. It achieves this through several key innovations.

DUAL: The Brains of the Operation

The core of EIGRP's intelligence is its algorithm, the Diffusing Update Algorithm (DUAL). DUAL is responsible for ensuring every path is loop-free and for calculating backup paths.

Beyond a Simple Routing Table

To make intelligent decisions, EIGRP maintains three tables instead of just one:

• Neighbor Table: A list of all directly connected routers that have established an EIGRP adjacency. It is like a router's phone book of trusted contacts.
• Topology Table: This is the heart of DUAL. The topology table contains every single path to every destination learned from its neighbors, not just the best one. It is the router's complete collection of all "rumors" it has heard.
• Routing Table: Contains only the absolute best path to each destination, selected from the topology table. This is the final "road atlas" used for forwarding packets.

Guaranteed Loop-Free Paths: The Feasibility Condition

The magic of DUAL lies in its ability to select backup routes that are guaranteed to be loop-free. It does this by using a simple but powerful rule called the Feasibility Condition. To understand this, we need two definitions:

• Feasible Distance (FD): This is my own personal, lowest-cost route to the destination. It's the metric of the path currently in my routing table.
• Reported Distance (RD): This is the distance my neighbor tells me they have to the destination. It is the neighbor's Feasible Distance.

The Feasibility Condition states: For a path advertised by a neighbor to be considered a valid, loop-free backup path, the neighbor's Reported Distance (RD) to the destination must be strictly less than my own current Feasible Distance (FD) to that same destination.

Analogy: Imagine you are at point A trying to get to a destination Z. Your current best path has a total travel time of 60 minutes (your FD). A friend at point B (your neighbor) tells you, "I know how to get to Z, and from my location, it only takes 40 minutes" (B's RD). Since 40 is less than 60, this is a valid backup. It means your friend is logically "closer" to the destination than you are, so you can safely use them without fear of them sending you back towards yourself. However, if another person at point C tells you, "I can get to Z in 70 minutes," you will ignore them as a backup, because their reported time is greater than your current best. Following their path could lead you in a loop: they might be trying to use your route!

Successors and Feasible Successors

This powerful condition allows DUAL to categorize the paths in its topology table:

• Successor: This is the route with the overall lowest metric (the best path). The successor is installed in the routing table.
• Feasible Successor (FS): This is an alternative path that meets the Feasibility Condition $(\text{RD} < \text{FD})$. It is a guaranteed loop-free backup path that is kept in the topology table, ready for immediate use.

This is the key to EIGRP's incredible speed. If the successor path fails, the router does not need to send out queries or wait for timers. If a Feasible Successor is present in the topology table, the router can promote it to become the new successor instantly. Convergence can be almost immediate. Only if no Feasible Successor exists does the router need to enter an "active" state and send out queries to its neighbors to find a new path.

Other EIGRP Advantages

Beyond its DUAL algorithm, EIGRP includes several other features that make it a modern, efficient protocol.

• Reliable Transport Protocol (RTP): EIGRP doesn't use UDP and hope for the best. It has its own lightweight, reliable protocol (not to be confused with the Real-time Transport Protocol for voice/video) to ensure that its packets (Hellos, Updates, Queries, Replies, Acks) are delivered and acknowledged by its neighbors.
• Partial and Bounded Updates: Unlike RIP's periodic broadcast of the entire table, EIGRP is far more efficient. It sends updates only when a change occurs in the network topology (a link goes up or down, or a metric changes). Furthermore, these updates are partial (containing only the information that changed) and bounded (sent only to the routers that are affected by the change). This drastically reduces the amount of routing traffic on the network.
• Classless by Design: EIGRP was designed from the start to be classless. It fully supports VLSM and sends subnet mask information with every route update.
• Equal and Unequal Cost Load Balancing: EIGRP can distribute traffic across multiple paths to the same destination. It natively supports load balancing over up to 4 equal-cost paths (configurable for more). Uniquely, it can also be configured to perform unequal-cost load balancing using a feature called Variance, allowing traffic to be sent over backup paths that are not as good as the primary, as long as they meet the Feasibility Condition.

Conclusion: EIGRP's Place in Modern Networking

IGRP is an obsolete protocol, a crucial historical bridge between the simplicity of RIP and the sophistication of modern routing. Its legacy lives on in EIGRP, which inherited and enhanced its composite metric.

EIGRP itself remains a highly respected and capable protocol. For many years, its biggest drawback was its proprietary nature, which limited its use to all-Cisco networks. In 2013, Cisco released the core functionality of EIGRP as an open standard to allow for greater interoperability. Today, it stands alongside OSPF as one of the two major IGPs. The choice between them often comes down to network requirements and engineer preference. EIGRP is frequently praised for its ease of configuration in simple networks and its exceptionally fast convergence times, while OSPF is valued for its robust, open-standard nature and highly structured hierarchical design for massive-scale networks.