Heterogeneous Computing

This work is about applying machine learning to performance modeling of applications on different targets.

Improve performance and power efficiency of regular computational kernels in ES and HPS systems, by efficient runtime mapping of computations to specialized CGRA, while ensuring the most transparency to the application developer.

SMILES is one of the main Research Lines of the Norte 2020 project Tec4Growth, involving all the 4 centres from the Computer Science cluster and 5 other centres from INESC TEC.

This project will develop, implement and evaluate a novel compilation and synthesis system approach for FPGA-based platforms.