Target and Host Setup

NVIDIA Nsight 2.2 allows you to debug your applications in two different ways:

• Local debugging, in which the host and target are on the same machine.
• Remote debugging, in which the host and target are on two different machines.
  
  

Local Debugging

When using NVIDIA Nsight to debug certain types of applications (such as most graphics applications), it's generally recommended to use a remote debugging configuration, as local debugging can adversely impact the performance of your application. In addition, some components of NVIDIA Nsight, such as shader debugging, will not work at all with a local debugging configuration.

If you configure your system for local debugging, you will use a single computer to build the project and run the application to be debugged. This means that you will need to have one or more GPUs in your computer, and must meet minimum requirements. (See System Requirements for a list of supported GPUs.) Note that CUDA debugging and analysis only requires one GPU, while graphics debugging requires at least two GPUs.

Remote Debugging

If you choose to configure your systems for remote debugging, you will use one computer as the host machine. The host machine will run Visual Studio to build your project, as well as to launch debugging sessions.

Meanwhile, a separate computer is configured as the target machine. The target will do the following:

• Run the NVIDIA Nsight Monitor. The monitor detects incoming requests from the host to execute an application on the target machine.
• Run the application to be debugged or analyzed. NVIDIA Nsight software manages the synchronization of files between the host and target machines.  
  

Performing remote debugging this way can be advantageous because the Visual Studio environment will continue to run on the host machine, even if the target machine has to be rebooted because of an application crash. Remote debugging also allows you to debug over a network, which means that a target machine can be made easily available to a development team, or to members who do not have physical access to the target GPU.