Course content:-
- Accelerating applications with CUDA C/C++: Write, compile, and run massively parallel CUDA C++ code
- Managing accelerated application memory with CUDA C/C++ unified memory: Perform efficient host-to-device and device-to-host memory transform using CUDA unified memory
- Asynchronous streaming, and visual profiling with CUDA C/C++: Use CUDA streams for concurrent kernel execution and do visual application profiling with Nsight Systems
Pre-requisites:
- Basic understanding of C/C++
- Your own laptop and charger
Food coupons will be given to attendees which can be used at all Taste Imperial outlets for lunch.
The course will result in a certificate from NVIDIA for all participants finishing the course.
The course is free of charge, but if you register and do not attend and/or cancel with insufficient notice, we will charge you as outlined in the POD website T&C