A structured approach to high availability and fault tolerance isessential in a production-grade service delivery network, where delaysand faults can occur for a multitude of reasons. In this thesis, weconsider the high-level scheduling and load sharing propertiesoffered by the Domain Name System, as implemented in popular DNSsoftware packages. At this level, the scheduling mechanism can accountfor server availability, geographical proximity, time zones, etc. Weexplore the performance and capabilities of high-level DNS-based loadbalancing, where we draw special attention to the choice ofcaching policy (time-to-live) for DNS data. Our findings confirm thehigh performance of modern DNS server implementations, but questionthe use of DNS as a suitable load balancing mechanism in itself.Further, we analyse the use of a database-supported DNS service forallowing highly dynamical query responses, and show that this approachhas both potentially negative (single point of failure) and positive(improved balancing flexibility) properties.