| ||
There are multiple programs on the market for tuning and optimizing your application. The two most popular performance monitoring applications are the VTune program by Intel for use with Intel's chips, available at http://developer.intel.com/software/products/eval, and the AMD Code Analyst for use with AMD's chips, available at http://www.amd.com.
These programs allow you to balance your pipelining and monitor the overall performance of your application and make recommendations as to how to balance your code to make it as tight as possible. By setting up monitoring zones indicated by start and stop markers, the application can be benchmark between those zones.