atomate

Official Resources

• Homepage: https://atomate.org/
• Documentation: https://atomate.org/
• Source Repository: https://github.com/hackingmaterials/atomate
• License: BSD 3-Clause License

Overview

Atomate is a library of pre-defined workflows for computational materials science. It is built on top of Pymatgen, FireWorks, and Custodian. Atomate provides robust, production-ready workflows for running VASP, multiple FEFF, and other calculations, handling everything from input generation to error correction and database insertion.

Scientific domain: Workflow automation, high-throughput materials science
Target user community: VASP users, Materials Project data contributors

Capabilities (CRITICAL)

• Pre-built Workflows: Standard workflows for band structure, elastic tensor, dielectric constant, Gibbs free energy, etc.
• VASP Integration: Extensive support for VASP calculations with best-practice parameters.
• Error Handling: Automatic error correction via Custodian (e.g., restarting divergent SCF).
• Database: Automatic parsing and insertion of results into a MongoDB database.
• Proven: Used to generate the Materials Project database.

Sources: Atomate documentation, Comp. Mater. Sci. 131, 106 (2017)

Inputs & Outputs

• Input formats: Pymatgen Structure objects, FireWorks workflow objects
• Output data types: MongoDB documents containing calculation details, properties, and provenance

Interfaces & Ecosystem

• FireWorks: The workflow engine
• Pymatgen: The analysis code
• Custodian: The error handler
• VASP: The primary DFT engine supported (also FEFF, LAMMPS, etc.)

Workflow and Usage

1. Setup FireWorks LaunchPad.
2. Add workflow: wf = get_wf_bandstructure(structure)
3. Add to LaunchPad: lpad.add_wf(wf)
4. Run: qlaunch (on cluster)
5. Query results from MongoDB.

Performance Characteristics

• Highly scalable (tested on millions of calculations)
• robust error handling saves computational time

Application Areas

• Database generation
• High-throughput screening
• Reproducible property calculation

Community and Support

• Developed by Materials Project team
• Active Google Group
• Large user base

Verification & Sources

Primary sources:

1. Homepage: https://atomate.org/
2. GitHub: https://github.com/hackingmaterials/atomate
3. Publication: K. Mathew et al., Comp. Mater. Sci. 131, 106 (2017)

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• Website: ACTIVE
• Documentation: COMPREHENSIVE
• Source: OPEN (GitHub)
• Development: ACTIVE (Materials Project)
• Applications: VASP workflows, high-throughput