Solving sparse linear systems with approximate inverse preconditioners on analog devices

Vasileios Kalantzis, Anshul Gupta, Lior Horesh, Tomasz Nowicki, Mark S. Squillante, Chai Wah Wu, Tayfun Gokmen, Haim Avron

Abstract: Sparse linear system solvers are computationally expensive kernels that lie at the heart of numerous applications. This paper proposes a preconditioning framework that combines approximate inverses with stationary iterations to substantially reduce the time and energy requirements of this task by utilizing a hybrid architecture that combines conventional digital microprocessors with analog crossbar array accelerators. Our analysis and experiments with a simulator for analog hardware show that an order of magnitude speedup is readily attainable despite the noise in analog computations.