Publications

This report documents the creation of four datasets for modelling and benchmarking computational methods. They are dedicated to further investigations of systems, which have been used for demonstrations in work package (WP) 5, i.e. hydrogen under pressure, protonated water hexamer and molecular interactions in excited-state organic dimers. The datasets have been developed in the groups of Michele Casula (CNRS), Matthias Rupp (LIST), Kasia Pernal (TUL), and Michal Hapka (University of Warsaw). ...Read more

The TREX EU Centre of Excellence investigates implementations of Quantum Monte Carlo (QMC) calculations optimized for exascale high-performance computing. These calculations are high-accuracy quantum-chemical and materials simulations that are inherently parallelizable and computationally demanding. Thus, they are uniquely positioned to utilize and explore the upcoming exascale supercomputer architectures. TREX focuses on the development and promotion of an open-source, high-performance software platform of inter-operable flagship codes and exascale-ready libraries. ...Read more

This deliverable adresses the critical concern of floating-point accuracy in numerical simulations and computation-intensive codes, emphasizing the need for early detection and resolution of numerical bugs. It introduces Verificarlo-CI, a continuous integration workflow that monitors and optimizes numerical accuracy during code development within QMCkl. It also details the design of optimized versions of Sherman-Morrison-Woodbury (SMWB) kernels and show they achieve a good tradeoff between accuracy and performance. The report also explores the benefits of mixed precision in the CHAMP code. ...Read more

The document is a direct output of WP7 whose key role in the project is to raise awareness about project results, disseminate major outputs, and engage key stakeholders. The aim of this document is twofold: ...Read more

The QMCkl library represents a significant advancement in the field of quantum chemistry, particularly in Quantum Monte Carlo (QMC) simulations. Distributed under the 3-clause BSD license, the library aims for broad adoption across both academic and commercial platforms by offering a permissive opensource licensing model. The development methodology of QMCkl is rooted in literate programming principles, ensuring a seamless integration of documentation and code. The library’s API is designed in C, ensuring cross-language compatibility and universal adaptability. ...Read more

The TREX project is a collaborative initiative aimed at enhancing Quantum Monte Carlo (QMC) simulations by developing and optimizing flagship codes tailored for exascale computing. This report provides a comprehensive overview of the advancements and achievements under Work Package (WP)2, Code Modularization and Interfacing, and highlights its essential role in accomplishing the TREX project’s primary objectives. The TREX project focuses on three key goals: ...Read more

The current document D6.2 describes and details the structure of the training and education strategy and the plans of the events that have taken place at the time of writing and those planned for the remaining project lifetime. Various criteria are used to sort most of the planned activities: a distinction is made based on the subject (HPC, quantum chemistry), between the planned, upcoming and future events, between events internal to the consortium and events open to external audiences. Document D6.2 is structured into 6 Chapters. ...Read more

This deliverable documents the initial performance analysis results obtained for all 6 TREX flagship applications. The focus was, in particular, the scaling of the application and the ability to exploit parallelism at all the different levels of modern HPC architectures. This ranges from the efficient use of SIMD instructions to the use of highly parallel compute accelerators like Graphics Processing Units (GPUs). ...Read more

The TREX project is in the field of high-accuracy quantum chemical and materials simulations with a special focus on Quantum Monte Carlo (QMC) approaches to the solution of the quantum many-body problem at the heart of atomistic physics, chemistry, and materials science. Importantly, due to their inherent parallelizability and high computational cost, QMC approaches, and thus TREX is uniquely positioned to fully exploit the massive parallelism of the upcoming exascale supercomputer architectures. ...Read more

In this first document, after a brief Introduction (Section 1), we are describing the rationale for QMCkl design guidelines (Section 2). In particular, objectives and general methodology will be highlighted. Then, more precise details on the methodology will be given by walking through a detailed DGEMM example (Section 3). Finally, some preliminary performance results on a key QMCkl routine (JASTROW computation will be given). ...Read more