Antoine Kaufmann is a research group leader at the Max Planck Institute for Software Systems (MPI-SWS). In his research, he builds computer systems at the intersection of software and hardware with a focus on performance and efficiency. Antoine received his PhD from the University of Washington in 2018, and his MSc and BSc from ETH Zürich in 2014 and 2012 respectively.
Satisfying the growing demand for more compute as processor performance continues to stagnate in the post-Moore era requires radical changes throughout the systems stack. A proven strategy is to move from today’s general purpose platforms to post-Moore systems, specialized systems comprising tightly integrated and co-designed hardware and software components. Currently, designing and implementing these systems is a complex, laborious, and risky process, accessible to few and only practical for the most computing intensive applications. In this talk, I discuss the nascent challenges in building post-Moore systems and illustrate them through specific systems I have built. As a first step to address these challenges, I present Simbricks, a modular simulation framework that flexibly combines existing simulators for computers, custom hardware, and networks, into complete virtual post-Moore systems, enabling developers to compare and evaluate designs earlier and quicker. I conclude with a look towards future opportunities in abstractions, tooling, and methodology to further simplify the development of post-Moore systems.
Please contact MPI-SWS Office for link information.