SQL server 2008: Selecting over a million records

Question

Selecting over a million records, what would be the best approach with performance. I am going to make up a table we can work later. But i want to get some information first. What are all the resources I need as for as CPU etc..........? Lets start there.