Approximate and Transprecision Computing Lectures Series
In the last 10 years, the demand for new computing strategies driven by energy-efficiency has grown exponentially. Flop-per-watt (thus, per-euro) has become de-facto a driving model in hardware design. Results in this direction have been significant, leveraging first multi-core parallelism and then recently moving toward heterogeneous architectures (e.g., multicore CPU coupled with GP-GPUs). However, these evolutions will not be sufficient in thelong term. To maintain an exponential increase in computational efficiency, we will need to rely either on an unlikely breakthrough discovery in hardware technology, or on a fundamental change in computing paradigms.
OPRECOMP is a 4-year research project,funded under the EU Framework Horizon 2020, that aims to build an innovative, reliable foundation for computing based on transprecision analytics.
OPRECOMP project is also committed in the dissemination of research results non only in scientific journal but also with educational events. Due to the COVID-19 epidemic emergency the Summer School foreseen for 2020 has been canceled. In place of the scheduled event we propose a set of lectures on key topics of the project, ranging from fundamental energy limitsto architectures, algorithms and code optimization.
We hope you enjoy these letures. Do not hesitate to contact the teachers, who will be happy to provide further clarifications or suggestions to deepen the subject.