[Paper Review] Eurolab-4-HPC Long-Term Vision on High-Performance Computing

Theo Ungerer, Paul Carpenter|arXiv (Cornell University)|Jul 11, 2018

Scientific Computing and Data Management2 references4 citations

TL;DR

This paper presents a long-term vision for High-Performance Computing (HPC) in Europe through the Eurolab-4-HPC initiative, proposing a strategic roadmap for emerging technologies such as 3D chip stacking, photonics, quantum computing, neuromorphic systems, and energy-efficient architectures. It integrates insights from leading European researchers to guide future HPC innovation, with a focus on sustainability, resilience, and hardware-software co-design across diverse emerging computing paradigms.

ABSTRACT

Radical changes in computing are foreseen for the next decade. The US IEEE society wants to "reboot computing" and the HiPEAC Vision 2017 sees the time to "re-invent computing", both by challenging its basic assumptions. This document presents the "EuroLab-4-HPC Long-Term Vision on High-Performance Computing" of August 2017, a road mapping effort within the EC CSA1 Eurolab-4-HPC that targets potential changes in hardware, software, and applications in High-Performance Computing (HPC). The objective of the Eurolab-4-HPC vision is to provide a long-term roadmap from 2023 to 2030 for High-Performance Computing (HPC). Because of the long-term perspective and its speculative nature, the authors started with an assessment of future computing technologies that could influence HPC hardware and software. The proposal on research topics is derived from the report and discussions within the road mapping expert group. We prefer the term "vision" over "roadmap", firstly because timings are hard to predict given the long-term perspective, and secondly because EuroLab-4-HPC will have no direct control over the realization of its vision.

Motivation & Objective

• To define a strategic, long-term vision for the future of High-Performance Computing in Europe, aligning research and innovation across emerging technologies.
• To identify and prioritize key technological pathways—such as 3D chip stacking, photonics, quantum computing, and neuromorphic systems—that will shape the next generation of HPC systems.
• To address critical challenges in energy efficiency, system resilience, and hardware-software co-design to ensure sustainable HPC evolution.
• To foster collaboration across European institutions and research communities through coordinated roadmapping and feedback from workshops and expert input.
• To guide policy and funding decisions by synthesizing expert insights into a unified vision for HPC beyond 2030.

Proposed method

• The vision was developed through a collaborative effort involving 20+ leading European researchers and institutions across diverse HPC technology domains.
• Contributors provided expert input on specific technologies including die stacking, 3D-chips, photonics, quantum and resistive computing, memristors, and nanomaterials.
• The process included structured roadmapping workshops hosted by HiPEAC and EXDCI, with feedback from working groups and technical experts.
• A comprehensive synthesis of technical, system-level, and sustainability challenges was compiled from contributions across multiple institutions and research areas.
• The final document was compiled and edited by researchers at the University of Augsburg, integrating diverse perspectives into a unified strategic vision.
• The output was published on arXiv as a preprint to ensure broad dissemination and community feedback.

Experimental results

Research questions

• RQ1 What are the most promising emerging technologies that will define the future of High-Performance Computing in Europe?
• RQ2 How can 3D chip stacking and advanced packaging techniques improve HPC system performance and energy efficiency?
• RQ3 What role can photonics, neuromorphic, and quantum computing play in overcoming current limits of traditional HPC architectures?
• RQ4 How can hardware-software co-design and resilience strategies be integrated into next-generation HPC systems?
• RQ5 What systemic and collaborative approaches are needed to ensure sustainable and coordinated HPC innovation across Europe?

Key findings

• 3D chip stacking and die-stacking technologies are identified as critical enablers for improving performance and reducing power consumption in future HPC systems.
• Photonics is highlighted as a transformative technology for interconnects, offering higher bandwidth and lower energy per bit than traditional electrical interconnects.
• Quantum and resistive computing are seen as emerging paradigms with potential to solve specific classes of HPC workloads beyond classical capabilities.
• Neuromorphic and in-memory computing approaches are positioned as key solutions for energy-efficient, low-latency processing of data-intensive and AI-driven workloads.
• Memristors and novel nanomaterials such as graphene and diamond transistors show promise for next-generation non-volatile and low-power computing components.
• A strong emphasis is placed on green ICT, system resilience, and hardware impact, with multiple contributors addressing sustainability and reliability across the HPC stack.