Guru,
These parameters affect the total amount of resources consumed by the application server so you will need to verify that the hardware available (steady state and peak) can handle the additional traffic on the server. This scaling is very hardware specific and also tied closely to what each user session is actually trying to do (some are small tasks and others have larger implications).
Based on the above comments you are not talking about a huge jump in resources but do you really have 1000 concurrent HTTP requests coming into the server at peak volumes and if so what does your availble CPU %, Available RAM, and Disk Utilization look like when this is happening?
Sam
PS: Have a look here for some addiitonal insight starting on page 19...
Executive Summary of the Performance Tests (PDF)*:* A high level overview of the hardware involved along with configuration of each scenario and a summary of the measured results from each test.
