June 16, 2022 | 7AM PDT / 10AM EDT / 4PM CEST
Join us to learn how to port CUDA kernels for GPUs to the Vitis™ unified software platform and its high-level synthesis (HLS) technology.
The Vitis software platform offers a unified design methodology and programming model for deploying accelerated applications on all Xilinx programmable logic-based platforms, allowing the developer to use C++ for kernel design. This webinar shows how a simple and concise coding style can deliver over 2X performance on Versal® devices vs. GPUs
During the session, we’ll demonstrate the Vitis HLS tool and interpret the results based on a dot product kernel example that we compare to CUDA.
Product Manager - Software Development Flows and Competitive, AMD