Program Summary
The Von Neumann architecture has significantly aided the rapid advancement of computing over the past seven decades. However, moving data between the processors and memory components of this architecture requires significant time and high-energy consumption, which constrains the computing performance and workload. Overcoming this bottleneck requires new computing architectures and devices that can significantly advance the computing performance beyond the traditional practice of transistor scaling (i.e., Moore’s Law).
The Foundations Required for Novel Compute (FRANC) program aims to develop innovative approaches to advance compute technologies beyond the Von Neumann topology. Leveraging recent advances in materials, devices, and integration technology, the program seeks to develop novel memory-centric compute topologies that break the traditional separation of processors and memory components to realize dramatic advances in compute efficiency and throughput of the workload, especially for applications constrained by size, weight, and power (SWaP). Innovative compute architectures and new, fast non-volatile storage and memory-centric computing devices will be explored under FRANC to enable low latency compute near or inside the data storage elements. Such approaches are particularly suited for applications relevant to artificial intelligence (AI) where in-memory computation provides unique advantages over traditional Von Neumann computation.