Design and acceleration of a quantum genetic algorithm through the matlab GPU library

Abstract

The potential processing power of a quantum computer is quantum parallelism, but significant disadvantages of quantum simulators are processing speed and memory. In this work, we illustrate with a Quantum Genetic Algorithm (QGA) the advantages of using the software platform of Compute Unified Device Architecture (CUDA) from NVIDIA, in special, the Matlab Graphic Processing Unit (GPU) library was used. The original software for Matlab named Quack!, which is a quantum computer simulator, was modified with the aim of speeding up a QGA. Experimental results that show advantages of using a QGA, as well as comparative experiments of the sequential implementation versus implementations that use the CUDA cores for different NVIDIA cards are presented.