I’ve been searching for information on how Full text indexing uses hardware
resources (Memory, IO, CPU etc) and how I can plan for scaling up data
volumes and search numbers. I have found that there is more information
regarding improving the actual indexing process but this is a minor aspect to
the project as it will be largely read only after the initial populate.

I realise that full text index data volumes are very dependant on the data
content but is there percentage of overhead that's generally accepted as an
average?

Is there a way to get performance statistics that show how particular
indexes are performing and how they are using the available resources?

Are there performance configurations that apply to searching rather than the
population performance (i.e. sacrifice population performance in favour of
increased search performance)?

Has anyone looked into these issues in the SQL 2008 CTPs?