Need Advice On Most Performant Option...

Hi there,I'm looking for advices from people with more experience using SQL Server.  I have a situation where from my perspective i have 3 ways to accomplish my task, but i dont know which way would be more efficient and performant.Here's the deal, i have a table which holds millions of records.  This table will eventually receives between 30-40 insertion a second.  So it's pretty busy.  To display the data correctly in reports over the web, i need to manipulate the data in such a way that certain transaction are modified and others are completly eliminated depending on a set of preferences choosen at the moment of the request for the report.  Obviously i dont see any ways to do this with just one SQL query, so my choices are the following:1.  Create a second table, which would hold the manipulated data. Create a trigger and on each insertion manipulate the data and modify the second table.2.  Create a stored procedure, from the stored procedure i query my records, create a temp table, manipulate my data and return it to the user3.  Simply fetch the data, return it to the IIS server and process it there before returning the result to the client.The manipulation process consists of querying about 1000 to 2000 records, eliminating duplicate results (or similar results within a chosen range) by comparing them to the last one, and to indicate for each record how many other records were similar.  In other words, it's a basic loop and i compare each record to the last one to see if it changed.Personally i think choice #2 would be better, but i could be wrong.  I'd like to hear your comments and suggestions.Thanks,