Post-Moore Technologies for Plasma Simulation: A Community Roadmap

2026-05-08

AI Summary

Plasma simulations are hindered by computational intensity, memory access bottlenecks, and high power consumption, challenges that conventional processors struggle to address sustainably. This work proposes a post-Moore co-design framework that systematically evaluates the applicability of reconfigurable accelerators, non-von Neumann architectures, and quantum computing across diverse plasma algorithms—including particle-in-cell, Vlasov, gyrokinetic, and magnetohydrodynamic (MHD) methods. The study introduces an innovative three-tiered technology roadmap: near-term kernel offloading and data service optimization, mid-term operator-level hardware acceleration, and long-term exploration of quantum approaches. It further underscores the critical importance of a modular software ecosystem and community-driven benchmarking. The findings reveal that no single universal solution exists; instead, progress must be pursued through staged, application-specific co-design strategies.

Abstract

Plasma simulations are among the most computationally demanding scientific workloads, combining high-dimensional kinetic evolution, particle-mesh coupling, field solves, and data-intensive communication. As general-purpose processor scaling slows, post-Moore technologies are being explored to address bottlenecks in data movement, memory access, and power consumption. This paper provides a community perspective on the role of these technologies in plasma simulation, assessing three major classes: reconfigurable and data-path accelerators, non-von Neumann architectures, and quantum computing. Each is evaluated, in a co-design approach, against representative plasma workloads spanning particle-in-cell, continuum Vlasov, gyrokinetic, fluid/MHD, hybrid, and warm dense matter methods. We find that no single technology can replace existing HPC platforms. Instead, three tiers of opportunity emerge: FPGA-class and data-path accelerators offer near-term kernel offload and workflow-level data services, non-von Neumann architectures represent medium-term directions for operator-level acceleration, and quantum computing, although the least mature, is potentially the most disruptive for warm dense matter and inertial confinement fusion microphysics. We outline best practices for selective adoption and identify focused demonstrators, benchmarking, and modular software ecosystems as immediate community priorities.

Problem

• plasma simulation
• post-Moore technologies
• computational bottlenecks
• data movement
• memory access

Innovation

• post-Moore computing
• plasma simulation
• co-design
• non-von Neumann architectures
• quantum computing