In today’s accelerated world the access time of network services has become an important factor among the Internet users, and particularly in enterprise environment. The other important requirement for network services is the extended availability time. For many applications even minimal service outage could be unacceptable. The access time on network services can be reduced and the availability time can be increased by using anycasting.
Anycasting is a communication paradigm, where multiple devices have a same address. When a client sent a packet to an anycast address, the network will deliver the packet to the nearest anycast host.
Because of the growth of the Internet, the precise design of the service implementation has become increasingly important. The analytic examination of the number of hops to the most nearby member of the anycast group provides help us in the network planning procedures. For the estimate of the hopcount has already existed a method, that is based on the distribution of the number of hops to the most nearby member of the anycast group.
In the first part of my thesis I present the details of anycast communication paradigm, some of the general use-case of anycasting, and I introduce the most important anycast routing protocols and anycast group membership management solutions. In the second part of this work I introduce the method, that deals with the estimation of the number of hops to the most nearby member of the anycast group, then I examine two anycast applications by using this method.