Newly released AiiDA v2.0 improves on performance and usability while maintaining a stable API

This was published on May 10, 2022

The developers of AiiDA have released AiiDA v2.0. The culmination of more than two years of work, the update focused on re-designing the codebase to make it more performant, user-friendly and maintainable, with minimal changes to the API.

AiiDA, a joint effort of NCCR MARVEL, the Theory and Simulation of Materials laboratory (THEOS) at EPFL and Bosch Research Center in Cambridge MA (USA), is an open-source Python workflow management system that enables the design and execution of computational workflows, ranging from simple pipelines of cheap data operations to sequences of expensive simulations connected through sophisticated logic. Its development is supported by a number of supporting partners and institutions and data generated with AiiDA are disseminated through

Besides many small improvements in the user interface and usability, the recent update features an improved file repository, simplification of the storage backend and more efficient export archives. These improvements make AiiDA more performant in certain areas (such as exporting and importing data, and creating backups of AiiDA installations) and make it possible to integrate AiiDA with alternative data storage methods, such as object stores on remote servers or in the cloud. Details on the update can be found in the AiiDA press release here. Additional information and commentary on AiiDA can be found in the Nature Physics Review article written by Sebastiaan P. Huber, post-doctoral Research Scientist at THEOS (currently active as the Data Officer at NCCR Catalysis).

Stay in touch with the MARVEL project

Low-volume newsletters, targeted to the scientific and industrial communities.

Subscribe to our newsletter