Applications of patchwork dynamics for glassy systems

We present work on ``patchwork dynamics'' as a technique for studying the nonequilibrium properties of glassy systems. In patchwork dynamics, we replace local Monte Carlo simulations, which require exponentially long times to equilibrate at a given length scale, with exact equilibration on patches at a given length scale, which can be done rapidly in models such as the 2D Ising spin glass and disordered dimer models. We have demonstrated some interesting applications of patchwork dynamics to such systems: 1) as a heuristic ground state algorithm for the 2D Ising spin glass on a torus (for which there are no known fast exact algorithms) and the 3D Ising spin glass; 2) as a method to study aging effects, persistence, and memory in 2D and 3D Ising spin glasses; 3) as a sampling procedure to study the nonequilibrium properties of disordered dimer models at finite temperatures.