Learn about best practices and new developments in OpenCL on NVIDIA platforms. We'll share our learnings and describe the best practices to follow when using OpenCL on NVIDIA platforms. We'll also touch upon new developments and performance improvements in NVIDIA OpenCL software and how they address developer concerns we understood in the past year. In particular, we'll describe how to manage memory efficiently on NVIDIA GPUs to get better OpenCL performance. We'll also introduce users to our new extension for memory allocation and also cover the performance characteristics of various memory management APIs.