SALT: Molecular Computing with Regular Crystals

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

SGER: SALT: Molecular Computing with Regular Crystals Abstract The research will investigate a massively parallel three-dimensional computing architecture implementing a new family of cellular automata called Salt. Transitions in the automata are governed by a 6-cycle state update process that alternates between two disjoint sets of cells in a regular lattice. It may be possible to construct such a regular crystal lattice structure at a molecular scale resulting in unprecedented density and parallelism. The physics and topology of such systems are suited for solving problems such as computational fluid dynamics and protein folding. In this work we will investigate computational models that exploit the Salt architecture, and conduct preliminary analysis of the engineering feasibility of Salt models at the physical level. Long-term efforts are directed towards nanoscale construction of these devices and identifying suitable programming models for Salt hardware.