Abstract: INTERSECT is conceived to provide new instruments and services to the community of materials and device modeling in the form of data, codes, expertise and interoperable solutions for the development of disruptive electronics. To this end, INTERSECT aims at establishing a high-level informatics infrastructure to manage the data produced by the Interoperable Materials-To-Device (IM2D) simulation box through the interconnection of three main engine codes: Quantum ESPRESSO (QE) and SIESTA, both being software tools for quantum mechanical modeling of materials, and Ginestra^TM, for atomistic and continuum modeling of electronic devices. IM2D combines these tools to achieve a predictive level of accuracy for the performance of the actual device from first-principles quantum mechanical simulations. The AiiDA and SimPhoNy softwares are at the core of such interoperability Hub (iHub). The former is a Python infrastructure supporting different codes through plugins for automated design and implementation of complex workflows and task tracking, while capable to store the full provenance of each object in a tailored database. The latter is a Python-framework offering semantic interoperability to third party tools, such as simulation-, data-storage- and data-transformation-backends. Ontology domains in common Resource Description Framework (RDF)-formats can be installed in the core component (osp-core), so that their entities can be instantiated on a script-basis, interconnected among each other and deliver information to syntactic data structures through osp-wrappers. The connection to public repositories and to a repository of repositories (a catalogue) is handled by a gateway to the Application Programming Interface (API) known as OPTIMADE, briefly described in the present deliverable.

Within this multilevel computational framework, data and data pipeline are of paramount relevance. Most relevant definitions, formats and data schema implemented within the INTERSECT project have been described in the First Data Management Plan (FDMP), Deliverable D4.2, submitted at M7 (July 2019). The present deliverable focuses on updates and new implementations not included in the first report. In the following, we first describe the aspects related to the access of data from public repositories by means of the OPTIMADE API gateway-client though AiiDA. Furthermore, the details of interfacing SimPhoNy and Ginestra^TM have been worked out more in detail in order to semantically enrich the IM2D toolbox. By differentiating the workflow parameters into multiple degrees of difficulty, the intersection will also provide features for upscaling the Ginestra^TM– Graphical User Interface (GUI) in terms of higher flexibility and usability for operators with varying scientific background. Finally, we briefly describe the recent actions adopted to assure the security of data and code sources in the development of the IM2D code.

Info & Download: F. Aquilante, et al. (2021): Data Management Plan Assessment and Revision. D4.6 of the H2020 project INTERSECT (final version as of 29/01/2021). EC grant agreement no: 814487, CNR, Modena, Italy.

Find all the INTERSECT deliverables developed so far in our Report page, and subscribe to our newsletter to keep up-to-date with our project activities!

F. J. Dos Santos, et al. (2021): QE and SIESTA workflows for advanced materials parameters (Task 2.1.3-5). Part 1, Deliverable D2.3 of the H2020 project INTERSECT (final version as of 29/01/2021). EC grant agreement no: 814487, EPFL, Lausanne, Switzerland. Download.

Rosati C. et al., (2021): Interoperable Material-to-Device simulation box for disruptive electronics. Deliverable D2.4 of the H2020 project INTERSECT (final version as of 29/01/2021). EC grant agreement no: 814487, EPFL, Lausanne, Switzerland. Download.

F. Aquilante, et al. (2021): Data Management Plan Assessment and Revision D4.6 of the H2020 project INTERSECT (final version as of 29/01/2021). EC grant agreement no: 814487, CNR, Modena, Italy. Download.

All project deliverables, MM6+1, M12+1 and M18+1 included, are available in our report section.

Executive Summary: Over the last two decades, the field of computational science has seen a dramatic shift towards incorporating high-throughput computation and big-data analysis as fundamental pillars of the scientific discovery process. This has necessitated the development of tools and techniques to deal with the generation, storage and processing of large amounts of data. In this work we present an in-depth look at the workflow engine powering AiiDA, a widely adopted, highly flexible and database-backed informatics infrastructure with an emphasis on data reproducibility.

We detail many of the design choices that were made which were informed by several important goals: the ability to scale from running on individual laptops up to high-performance supercomputers, managing jobs with runtimes spanning from fractions of a second to weeks and scaling up to thousands of jobs concurrently, and all this while maximising robustness. In short, AiiDA aims to be a Swiss army knife for high-throughput computational science. As well as the architecture, we outline important API design choices made to give workflow writers a great deal of liberty whilst guiding them towards writing robust and modular workflows, ultimately enabling them to encode their scientific knowledge to the benefit of the wider scientific community.

Info & Download: https://doi.org/10.1016/j.commatsci.2020.110086.