About AATLAS
Project Overview
AATLAS is a 48-month project funded by the French National Research Agency (ANR). It develops hybrid analog–digital solvers to tackle NP-hard optimization problems such as Max-Cut, k-SAT, and the traveling salesman problem. The approach combines the speed and energy efficiency of analog reconfigurable cores on Field-Programmable Analog Arrays (FPAA) with the flexibility of digital processors on Field-Programmable Gate Arrays (FPGA). This hybrid strategy enables scalable demonstrators capable of handling problems ranging from 50 to 1000 nodes.
Goals and Impact
The project aims to design and implement hybrid FPAA–FPGA solvers, demonstrate scalability up to 1000 nodes, and develop algorithms adapted to hybrid hardware. AATLAS will provide open demonstrators and tools for the scientific community, advance unconventional computing, and offer energy-efficient solutions to real-world optimization challenges. It will also contribute to training young researchers at the crossroads of physics, computing, and mathematics.