Share This Article
Updated on May 9, 2025
Intermediate System to Intermediate System (IS-IS) is a link-state routing protocol that helps routers share information and deliver packets efficiently within a single autonomous system (AS). It uses the Shortest Path First (SPF) algorithm to find the best routes by creating a full map of the network. This guide breaks down its key concepts, how it works, its features, and how it’s used in modern networks.
Definition and Core Concepts
IS-IS is designed to handle large and complex networks, making it a top choice for service provider networks and large enterprises. To understand its operation, one must first grasp its fundamental elements:
Link-State Routing Protocol
IS-IS operates as a link-state routing protocol. Unlike distance-vector protocols, which rely on hop counts, link-state protocols require each router to maintain a complete view of the network topology. This approach ensures more accurate and efficient routing.
Autonomous System (AS)
An autonomous system is a collection of IP networks and routers managed as a single entity. IS-IS operates within the boundaries of an AS, facilitating efficient routing and network management.
Intermediate System (IS)
The term “intermediate system” refers to a router. These systems communicate with one another using IS-IS to share routing information and construct a consistent view of the network.
Link-State Packet (LSP)
LSPs are the building blocks of IS-IS. These packets carry information about a router’s directly connected neighbors and their link metrics, enabling the construction of a topology database.
Shortest Path First (SPF) Algorithm
Using the SPF algorithm (commonly referred to as Dijkstra’s algorithm), IS-IS computes the most efficient paths for data to traverse through the network. This algorithm calculates routes based on link costs, ensuring optimal performance and reliability.
Topology Database
Routers running IS-IS maintain a topology database, also known as the link-state database. This repository contains details about all routers, links, and metrics in the network, which are used to calculate routing paths.
Levels (Level 1 and Level 2)
IS-IS employs a two-level hierarchy to increase scalability:
- Level 1 routers form routing tables within a single area.
- Level 2 routers connect multiple areas, enabling inter-area communication.
Areas
An area is a logical subdivision of an autonomous system. Dividing a network into areas reduces the amount of routing information each router must process, improving efficiency and scalability.
How It Works
IS-IS achieves reliable routing through a series of highly efficient processes. Here’s how it functions step by step:
Neighbor Discovery
Routers running IS-IS identify neighboring routers and maintain adjacencies through the exchange of “hello” packets at regular intervals. These hello packets also serve to detect neighbor failures.
LSP Flooding
Once neighbors are discovered, routers generate and share LSPs. These packets are flooded throughout the network. Level 1 Link-State Packets (LSPs) are flooded within their specific area, while Level 2 LSPs are flooded throughout the entire autonomous system, ensuring all routers maintain a consistent and complete view of the network topology relevant to their level.
Topology Database Synchronization
Routers store received LSPs in their topology database. This synchronization ensures that every router within the network maintains an identical database, crucial for consistent routing.
SPF Calculation
With the help of Dijkstra’s algorithm, routers analyze the topology database to compute the shortest paths to all destinations. These calculations are based on link costs, which are typically determined by bandwidth, delay, or other metrics.
Routing Table Population
After calculating the shortest paths, routers populate their routing tables with next-hop information. This enables the efficient forwarding of packets across the network.
Key Features and Components
The adoption of IS-IS in large and dynamic networks is driven by several key features:
Link-State Algorithm
IS-IS leverages a link-state approach, maintaining a detailed map of the network to enable precise and efficient routing decisions.
Fast Convergence
When network changes occur, IS-IS quickly updates its topology database and recalculates routes, minimizing downtime and ensuring uninterrupted connectivity.
Scalability
The protocol’s hierarchical design, including areas and levels, allows it to scale effectively in large and complex networks.
Support for VLSM and CIDR
IS-IS supports Variable Length Subnet Masking (VLSM) and Classless Inter-Domain Routing (CIDR), enabling efficient IP address allocation and usage.
Two-Level Hierarchy
The distinction between Level 1 and Level 2 routers enhances scalability by limiting the scope of network changes and reducing processing requirements.
Use Cases and Applications
IS-IS is widely used across various network types due to its robustness and scalability. Common applications include:
Large Enterprise Networks
Enterprises with extensive, geographically distributed networks benefit from IS-IS’s scalability and efficient routing mechanisms.
Service Provider Networks
Service providers rely on IS-IS to manage complex topologies and ensure stable connectivity for their customers.
Networks Requiring Fast Convergence
Organizations that demand minimal downtime, such as financial institutions and healthcare providers, turn to IS-IS for its rapid convergence capabilities.
Networks with Complex Topologies
IS-IS is ideal for networks with intricate layouts and high volumes of routing information, thanks to its link-state approach and hierarchical structure.
Key Terms Appendix
- IS-IS (Intermediate System to Intermediate System): A link-state routing protocol used for exchanging routing information in an AS.
- Link-State Routing Protocol: A protocol that maintains a comprehensive view of network topology.
- Autonomous System (AS): A collection of IP networks under a common administration.
- Intermediate System (IS): A router participating in IS-IS.
- Link-State Packet (LSP): A packet carrying information about a router’s neighbors and link metrics.
- Shortest Path First (SPF) Algorithm: An algorithm used to calculate the most efficient routing paths.
- Topology Database: A repository of network topology information maintained by routers.
- VLSM (Variable Length Subnet Mask): A technique for efficient IP address allocation.
- CIDR (Classless Inter-Domain Routing): A method for IP address aggregation and route optimization.
- Convergence: The state in which all routers have an identical and updated view of the network.
