Table of Contents
IS-IS routing protocol overview is the target of the first lesson of the IS-IS course. In this section we will have an overview of the ISIS routing protocol and the main differences from the OSPF routing protocol.
IS-IS routing protocol is a link-state routing protocol like OSPF. However, IS-IS is mainly used in service provider environments.
This course is dedicated to discuss and configure the topics related to IS-IS routing protocol. it is assumed that you already know the concepts and configuration of OSPF routing protocol.
We will implement all the configurations of this course in cisco devices.
IS-IS Routing Protocol Overview
IS-IS routing protocol, like the OSPF, is considered as a link-state routing protocol and therefore the rules governing ISIS are those defined in the link-state routing protocols.
How Link State Routing Protocols Work? A Brief Overview
In link-state routing protocols, each router transmits the information of the links and routers connected to itself throughout the network, and therefore each router is informed of the entire network topology.
Then, each router extracts the best route to each link in the network by running the SPF Algorithm and places the best route in the routing table.
Topology changes are immediately propagated into the network by the connecting routers. Then the routers update their routing table by re-running the SPF algorithm.
What is defined here also applies to the OSPF routing protocol. Then the question arises: What are the differences between these two routing protocols?
IS-IS versus OSPF
If I want to briefly mention some important differences in the overview section, the first is the usage of these two protocols.
IS-IS and OSPF Usage
Nowadays, the OSPF protocol is mainly used in the enterprise environment, but the IS-IS protocol is mainly seen in the service providers, which is due to the differences in the structure of the area and also the way of sending link state information in these two routing protocols.
Areas Structure in IS-IS and OSPF
In the ISIS protocol, the structure of the areas is more flexible, and the limitation in the OSPF protocol that you cannot communicate directly between areas does not exist in the ISIS protocol.
In OSPF, communication between areas is only possible via the backbone area. But in IS-IS any two areas can communicate with each other directly if there is a physical connection between them.
Link State Information Types in IS-IS and OSPF
The link state information sent in OSPF is called LSA and it varies widely including LSA type 1, Type 2, Type 3, Type 4, Type 5, Type 7 and some other types which are less used. This diversity brings flexibility and also complexity to the OSPF.
But in ISIS, link state information is called LSP and there are only two types, LSP L1 and LSP L2. Link state information in IS-IS is much simpler and has less overhead than OSPF.
This lower overhead is too significant in a way that ISIS can be used for very large environments such as service providers.
According to Cisco’s old documentation, a maximum of 50 routers could be tolerated in one area of OSPF, and if the number of routers is larger, The routers must be placed in different areas. In ISIS this limitation is 1000 in the same documentation.
OSPF limitation in MPLS Traffic Engineering
This feature of ISIS that you can place 1000 routers in one area is very attractive for service providers. The reason is due to the MPLS Traffic Engineering service.
Most ISPs use MPLS traffic engineering service to maximize the capacity of their network infrastructure. You can take full advantage of Traffic Engineering Service if your network has only one area and this is not usually possible in OSPF.
Due to IS-IS scalability, it offers service providers the possibility to place the entire network in just one area and in this way fully utilize the traffic engineering service in the network, which also reduces costs.
OSPF is mostly implemented by enterprises as they usually have fewer routers and don’t use MPLS traffic engineering service. In OSPF there is a lot of flexibility in area types and LSA types and therefore gives network engineers more flexibility and freedom in implementation.
IS-IS Extensibility
In OSPF it is not easy to extend the protocol to send new information. Therefore, to enable OSPF for IPv6, a new protocol, OSPFv3, is designed and developed with some modifications.
But in IS-IS without changes, just with adding a new TLV, new information related to the IPv6 network is announced.
Therefore only one IS-IS protocol is enough to advertise both IPv4 and IPv6 information, but in OSPF two different protocols, OSPFv2 and OSPFv3 are used to advertise information of IPv4 and IPv6 networks.
IS-IS protocol was initially designed for CLNS networks, but since the thought and design of this protocol was based on extensibility, with a few changes, now it can be used easily in IP networks.
IS-IS over IP network
IS-IS protocol differs from OSPF also in the context of IP networks.
ISIS protocol was originally developed for CLNS networks using the CLNP protocol and was later extended to include the IP functionality as well.
Even if IS-IS is only used on IP networks, they still need CLNS addresses for their proper operation.
Using CLNS addresses on IP networks is kind of annoying for network administrators and probably this is one reason that enterprises have less tendency to use ISIS.
ISIS versus OSPF Summary
| ISIS | OSPF |
|
Area Structure is more flexible easy to extend and direct communications between Areas |
Area Structure is not flexible. difficult to extend and communication between Areas through backbone Area |
| Over Layer 2 | Over Layer 3 (IP) |
| needs CLNS Adrress in addition to IP | need only only IP addresseing |
| Extensible (ISIS for both IPv4 and IPv6) | not Extensible (OSPFv2 for IPv4 and OSPFv3 for IPv6) |
| scalable even with a single Area | scalable with multiple Area |
| LSP L1 and LSP L2 | LSA Type 1,2,3,4,5,7, ... |
| mostly used in Service providers | mostly used in Enterprises |
IS-IS Basic Terminologies
IS stands for Intermediate System, which means router, and IS-IS is a protocol that is activated between two IS or two routers, which means routing protocol.
In CLNS networks, which no longer exist today, the computers were called ES or end system, and the protocol between end nodes and routers was called ES-IS.
The IS-IS routing protocol is the only routing protocol that runs over layer 2, or data-link layer. If you remember EIGRP and OSPF routing protocols run over IP network, RIP routing protocol runs over UDP and BGP routing protocol runs over TCP.
IS-IS in the next Sections
after discussing IS-IS routing protocol overview, In the next few sections, we’ll learn about the concepts that are somehow new in the ISIS routing protocol and slightly different from OSPF. We will also discuss the hierarchical structure of the area as well as the addressing system in ISIS.
Thanks man ,
you are welcome
Thanks man! I would love to see more, this really helps and I really appreciate your efforts in propagating this knowledge!!!
thanks. you are welcome. I am happy that you like the content.