Massively parallel split-step Fourier techniques for simulating quantum systems on graphics processing units

Mark Richter, Binh Trieu, G. W. Arnold +1 more

Bharat S. Rawal

Computing today requires a lot of processing power, an immersive environment, and excellent precision. In this paper, we propose a hybrid multiprocessor architecture with quantum integration for data-intensive applications like a metaverse. Regular non-computationally intensive applications are handled by the central processing unit (CPU) and video rendering, graphically intensive programs by the graphics processing unit (GPU), and enormous, parallel, and computationally intensive applications by the quantum processing unit (QPU). To achieve higher permanence computing, we provide combined CPU, GPU, and QPU architecture in this work. This paper provides a roadmap for truly high-performance computing for massive real-time systems. In addition, we present an onsite and cloud-based architecture for scalability. Also, we present multimodality security protection systems for internal and external communication based on the split computing paradigm.

Solvita Zariņa, Rūsiņš Freivalds

Benjamin Himberg, Adrian Del Maestro

An exact quantum Monte Carlo algorithm for interacting particles in the spatial continuum is extended to exploit the massive parallelism offered by graphics processing units. Its efficacy is tested on the Calogero-Sutherland model describing a system of bosons interacting in one spatial dimension via an inverse square law. Due to the long range nature of the interactions, this model has proved difficult to simulate via conventional path integral Monte Carlo methods running on conventional processors. Using Graphics Processing Units, optimal speedup factors of up to 640 times are obtained for N = 126 particles. The known results for the ground state energy are confirmed and, for the first time, the effects of thermal fluctuations at finite temperature are explored.

Eladio Gutiérrez, Sergio Romero, Marı́a A. Trenas +3 more

Due to the increasing computing power of the graphics processing units (GPU), they are becoming more and more popular when solving general purpose algorithms. As the simulation of quantum computers results on a problem with exponential complexity, it is advisable to perform a parallel computation, such as the one provided by the SIMD multiprocessors present in recent GPUs. In this paper, we focus on an important quantum algorithm, the quantum Fourier transform (QTF), in order to evaluate different parallelization strategies on a novel GPU architecture. Our implementation makes use of the new CUDA software/hardware architecture developed recently by NVIDIA.

Koen De Raedt, Kristel F. L. Michielsen, Hans A. De Raedt +6 more