Skip to main content

Kernel Tuner simplifies the software development of   optimized and auto-tuned GPU programs, by enabling Python-based unit  testing of GPU code and making it easy to develop scripts for   auto-tuning GPU kernels. This also means no extensive changes and no new dependencies are 
required in the kernel code.
The kernels can still be compiled and used as normal from any host  programming language.

Kernel Tuner provides a comprehensive solution for  auto-tuning GPU programs, supporting auto-tuning of user-defined  parameters in both host and device code, supporting output verification 
of all benchmarked kernels during tuning, as well as many optimization  strategies to speed up the tuning process.


Used by CoE: ESiWACE3


Main uses: Performance optimization of GPU applications