Attended Hands-On Tutorial: Introduction to oneAPI with Intel FPGAs at FPGA 2021 Conference

In one of old posts I have written about Data Parallel C++ book. This year in February 2021 I had the opportunity to attend the 29th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA 2021). It is the premier conference for presentation of advances in FPGA technology. In this conference there were couple of tutorials. I had an opportunity to attend the a hands-on tutorial on oneAPI with Intel FPGAs. I was excited and it was worth it.

The tutorial was how to use the oneAPI software model to create Data Parallel C++ programs to target supported FPGA acceleration cards. It was a good learning experience that how a source code is interpreted by the compiler to build a custom hardware datapath.

The tutorial was about the 3-step flow for development: (1) emulation, (2) using the static optimization report to fine tune your implementation, and (3) compiling a bitstream for the FPGA. As a tutorial attendee we were introduced to optimization concepts including pipelining loop iterations and architecting kernel memory, which are important when targeting FPGAs. In the second hour of the tutorial, we had practiced the multi-step development flow, and progress through code emulation and several stages of optimization with sample code implementing the Hough Transform.

And the most interesting part was the lab portion where one could use JupyterLab on the Intel DevCloud for oneAPI — which required only a web browser to access an environment to work with the latest toolset.

The community also has very active Slack Channel where one can post technical questions and ask for help.

Considering the emerging heterogenous architecture, I can already foresee the growing importance of frameworks like OneAPI.

I will post more about my learning outcomes regarding OneAPI.