I'm referring to parent's point "Sometimes you just have to do joins across large tables.". Indexes and query hints can make a world of difference once you prof

I'm referring to parent's point "Sometimes you just have to do joins across large tables.". Indexes and query hints can make a world of difference once you profile how the query is being executed. Another option is to create a stored procedure which first extracts a smaller amount of data from first table to a temporary table and then join it to the other large table. Measuring and testing different optimization methods takes a lot of effort and time. We should only do that when faced with a real problem of slowness. "The first rule of optimization is: Don't do it. The second rule of optimization (for experts only) is: Don't do it yet. Measure twice, optimize once."