Shifting sands of hardware and software in exascale quantum mechanical simulations

TL;DR

The paper discusses the challenges and necessary adaptations in hardware and software for quantum mechanical simulations at exascale computing, emphasizing heterogeneous architectures and the need for new algorithms and standards.

Contribution

It highlights the limitations of traditional software in exascale environments and proposes strategies for adapting quantum chemistry software to heterogeneous architectures.

Findings

Traditional algorithms are insufficient for exascale heterogeneous systems.

Efficient GPU-tailored algorithms and reduced precision methods are crucial.

Standardized libraries and programming models are needed for software scalability.

Abstract

The era of exascale computing presents both exciting opportunities and unique challenges for quantum mechanical simulations. Although the transition from petaflops to exascale computing has been marked by a steady increase in computational power, it is accompanied by a shift towards heterogeneous architectures, with graphical processing units (GPUs) in particular gaining a dominant role. The exascale era, therefore, demands a fundamental shift in software development strategies. This Perspective examines the changing landscape of hardware and software for exascale computing, highlighting the limitations of traditional algorithms and software implementations in light of the increasing use of heterogeneous architectures in high-end systems. We discuss the challenges of adapting quantum chemistry software to these new architectures, including the fragmentation of the software stack, the need for more efficient algorithms (including reduced precision versions) tailored for GPUs, and the importance of developing standardized libraries and programming models.