Both OSPF and ISIS are suitable for Service Provider environment, because of their link state nature. While ISIS is the more deployed, the predominant argument is to use which you are more comfortable or familiar with.
Both routing protocols ensures that you have a proper design, segmenting your network into areas and an hierarchal ip address design. Both have a concept of a backbone. However, in OSPF the backbone is an area, while the backbone in ISIS is a chain of routers. Because of the requirement to have all areas connected to the backbone, extending the network in ISIS is easier than in OSPF.
However, OSPF areas has more features such as stubby areas, NSSA, totally stubby etc. There are also more route types in OSPF (Internal/External, E1/E2).
OSPF also has more link types (PtP, Broadcast, NBMA, PmP) while ISIS has only two (LAN and PtP), hence OSPF is more preferred on NBMA networks like frame relay.
The general opinion is that ISIS supports more routers per area and is more stable.
as already posted both protocolls are very similar. Also both will allow for MPLS traffic engineering. The information, that ISIS allows for more routers in an area than OSPF is more like a myth. Both are so similar, that you can expect to have a similar amount of routers in one area/level (about 600 is a proven number - yet we do talk about 7200 up, in case you have 836 this might be too much). Max routers per area is a question of memory and CPU utilization - not such an issue with 7200VXR, 12k and the like.
So both protocolls are similar in scalability, ISIS is more easy to handle when you have multiple areas, but the importance of it depends on the size of the network you are planning.
Another point: OSPF is available in more IOS trains and feature sets than ISIS.
One other point to consider: What does your staff know about OSPF/ISIS. From an operational point of view you need to have the proper knowledge to achieve reasonable times to resolve problems. This usually favours OSPF in because you have more OSPF routers deployed than ISIS, i.e. easier to get "OSPF people" than "ISIS people".
Another key point to note is that ISIS metrics do not scale with link bandwdidth. The default operation therefore looks like it uses hop count as its metric. The value for each link metric need to individually configured.
Despite all these, I think the most important issue is still personell and support. Use the protocol that you are most comfortable and familiar with.
Not that it is more easy to be OSPF person, but that there are more people that know OSPF than there are that know ISIS. Because OSPF is more widely used, people are more familiar with it, and therefore you will get more support on OSPF. For example, you can check how many people have asked questions on ISIS on this forum as against OSPF.
With 20 routers, you do not have a problem for both protocols.
The two protocols are very similar, that is why for a lot of people the decision is made on which protocol they are more familiar with.
I agree with other respondents that the two protocols are similar but also do have their differences.
You need to consider what kind of services the network will need to provide and the size of it with a scalability factor, before making a decision.
Most Service Providers who deploy MPLS networks nowadays opt for ISIS with all routers being leve-2. The main reason being scalability over a single area due to features such as Partial Route Calculation (PRC). This is due to the fact that certain MPLS Traffice Engineering features can only be supported over a single area.
Also ISIS now supports wide metrics which allow for a 24bit interface metric and 32bit path metric. Some features such as MPLS-TE are supported only with wide metrics, so all routers have to be configured with wide metrics.
I will say that the same goes for OSPF. A lot of service providers deploying OSPF also uses a single area, the Area 0. MPLS traffic engineering would work in a single area, though there are new IOS features that allow it to work in multiple areas, but the operation is not optimal.
Also, because of route summarisation, you generally want to use a single area. route summarisation breaks LSPs which is a key point in MPLS deployment.
I truly wonder how many networks would be big enough to have a scalability issues with either protocol.
We have 3 identical switches configured by someone else and would like to claim some of the Gigabit ports(G1/G2/G3/G4) for use on servers. When we try to change the wiring and configuration, we run in to connectivity issues. Attached is a des...
This is actually a pretty cool feature, i didn't even know it existed until I was looking for a solution to advertise a subnet (prefix in BGP talk), only if a certain condition existed. This is exactly what conditional advertisements does
j ai une question j ai achete un routeur cisco 887VA-k9 , je le configuré avec la configuration ci- dessous
si je le lier avec mon pc portable sur l un de ses ports directement ça marche toute est bien ( la connexion internet + m...