New Computing Paradigm for Analyzing Increasingly Complex Social Infrastructure Systems — Ising Computer

Masanao Yamaoka, Ph.D. Chihiro Yoshimura Masato Hayashi Takuya Okuyama Hidetaka Aoki Hiroyuki Mizuno, Ph.D. OVERVIEW: The optimization of social infrastructure systems will be among the requirements of Hitachi’s Social Innovation Business in the future, therefore, there will be a need to solve combinatorial optimization problems. Hitachi has devised a computing technology based on a new paradigm that is capable of solving combinatorial optimization problems efficiently using an Ising model, and has built a prototype 20,000-spin Ising chip using a 65-nm process. An Ising chip represents a combinatorial optimization problem by mapping it onto an Ising model based on the spin of magnetic materials, and solves the problem by taking advantage of the system’s natural tendency to converge. This convergence is implemented using a CMOS circuit. In addition to demonstrating its ability to solve combinatorial optimization problems and operate at 100 MHz, the prototype chip consumes approximately 1,800 times less power to obtain the solution than required by a conventional Neumann-architecture computer.