The Interoperable Material to Device (IM2D) simulation box is the core outcome of the INTERSECT project. It is meant to provide material properties on-demand through robust and automatized workflows for advanced device modelling. In the various stages of the IM2D simulation box, data are received, processed, and stored.
In D2.6, we describe how the automated infrastructure is able to handle, store, and share data, which is the core of the IM2D data hub. We start with an introduction to the components of the IM2D box and how they interact, then we give a complete description of the data pipelines.

An essential piece of information required to perform Density Functional Theory (DFT) simulations on a material is its atomistic crystal structure. This contains information on the periodicity of the lattice, its chemical composition, and the arrangement of the atoms inside the unit cell. Up to now, we have implemented data pipelines to give access to the IM2D users materials available to almost all the major open-access materials databases, such as Materials Cloud, Materials Project, Crystallography Open Database (COD), and many more.

An important step to make this access as much universal as possible, has been the development of the OPTIMADE (Open Databases Integration for Materials Design) REST (REpresentational State Transfer) API (Application Programming Interface), which unifies the way various databases expose their data on the web.
We present our effort towards the (OPTIMADE) and the tools we help to develop to make it user-friendly and open access. Furthermore, we demonstrate how we implement queries to the OPTIMADE API within the IM2D framework.

Info & Download: F. J. dos Santos, F. Benedetto, C. Rosati, M. Bertocchi, F. Aquilante, A. Calzolari, and N. Marzari (2021): Pipelines to/from external repositories. Deliverable D2.6 of the H2020 project INTERSECT (final version as of 28/07/2021). EC grant agreement no: 814487, EPFL, Lausanne, Switzerland.

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!

The first papers is “AiiDA 1.0, a scalable computational infrastructure for automated reproducible workflows and data provenance“, by Talirz, L., Kumbhar, S., Passaro, E. et al., including or PI and WP2 coordinator Nicola Marzari from our EPFL partner.

AiiDA is a “(…) robust open-source high-throughput infrastructure addressing the challenges arising from the needs of automated workflow management and data provenance recording“.

The paper introduces developments and capabilities required to reach sustained performance, with AiiDA supporting throughputs of tens of thousands processes/hour, while automatically preserving and storing the full data provenance in a relational database making it queryable and traversable, thus enabling high-performance data analytics. AiiDA is a constituent part of the INTERSECT IM2D box, an interoperable materials-to-device simulation platform for the development of disruptive electronics.

The second paper is “Materials Cloud, a platform for open computational science“, by Huber, S.P., Zoupanos, S., Uhrin, M. et al., including Nicola Marzari.

Materials Cloud is a “(…) platform designed to enable open and seamless sharing of resources for computational science, driven by applications in materials modelling“.

The paper illustrates the Materials Cloud platform ecosystem, with its LEARN, WORK, DISCOVER, EXPLORE, and ARCHIVE sections, and aiming to support researchers throughout the life cycle of a scientific project and to make their research output FAIR and reproducible. AiiDA also manages the interconnection of our INTERSECT IM2D from/to external materials databases, such as Materials Cloud.

Read more about the papers on https://nccr-marvel.ch/highlights/2020-09SciData.
To discover more our INTERSECT advancements, take a look at the most recent published deliverables (M18+1) on our website.

References:
→ Huber, S.P., Zoupanos, S., Uhrin, M. et al. AiiDA 1.0, a scalable computational infrastructure for automated reproducible workflows and data provenance. Sci Data 7, 300 (2020).
→ Talirz, L., Kumbhar, S., Passaro, E. et al. Materials Cloud, a platform for open computational science. Sci Data 7, 299 (2020).