Wednesday, 30 March 2016

Migration Methods

In this post, we will discuss various options to migrate customer's network from one service provider to other service provider. 

Let's assume that you work for an enterprise customer with three sites. All the sites are currently connected through a Layer 3 IPVPN solution provided by service provider A.



Now lets say there is a new service provider "B" which can offer a similar Layer 3 VPN solution with better SLA and reduced cost and your task is to migrate all three sites from existing provider to new provider as seamless as possible. 

What are the ways to do this?

1. Big Bang:- The first method of doing this is what I called a "Big Bang" approach. In this method, you will have to make sure that the circuits from SP2 are installed on all the sites. After that you can schedule a migration time windows during which you will have to disconnect SP1 circuits from all the sites and start using new circuits. This migration technique may be used in smaller networks but not ideal for medium to large networks. 



2. Transit Site:- The second method of doing this is by using either your HQ or DC as a transit site. As shows in the diagram we will have to exchange routes between old and the new network by using either static routes or dynamic routing protocols. 
Once a site migrates from SP1 to SP2, the traffic between migrated and non migrated sites will traverse through this transit site. The benefits of using this method is that you can plan the migration according to their requirements. You, the customer will be in charge of routing so minimum input is required from the both the service providers. 



However during the migration phase, all the traffic between migrated and non migrated sites will trombone through the transit site so you will have to spec up the bandwidth of the transit site accordingly. Also lets say in this example, if the HQ is in Germany which is a transit location and both the remote offices are in London. In normal condition, one remote office will talk to the other remote office directly. If one site gets migrated then the traffic between both the sites in London will traverse through Germany which means increase in RTD. So the scheduling of migration is also very important factor.


3. SP NNI:- The third method of migration is by using an NNI between service providers. In this method, both the service providers will setup a connection or connections, at a common location to exchange the routes. As the site migrates, the NNI will be used as a transit point for the customer sites. It is very similar to the previous option we saw. 


The key difference here is that we are not using customer site for the connectivity which means customer will have to provider less input however it also means that customer will have less control and visibility of the traffic during migration. Obviously customer will also have to consider the additional costs of this circuit.


In future posts, we will see how we can write migration plan.

4 comments:

  1. Once again, excellent blog. Thanks. I did not know that there is NNI for migration purpose, but with additional cost. I guess, above options are when you have no backup link from the site (Like DSL) or back link BW is not enough?

    Let's say if we have backup link and normally used only during failover. In this case, we can failover traffic to backup link, do transition of primary. Once primary up with new SP, we can router the traffic through it (so switched back from backup link). Am I correct in saying this? Or any other issues we have to be aware of?

    ReplyDelete
  2. My family always say that I am wasting my time here at net, but I know I am getting experience all the time by reading such pleasant posts.
    Free open vpn

    ReplyDelete
  3. Great post Thank you for sharing your insights on this topic.

    Server Virtualisation Consulting
    Fortinet Services

    ReplyDelete
  4. Thank you for sharing your thoughts and knowledge on this topic.

    Networking Consultancy

    ReplyDelete