Exabyte.io - accessible and collaborative environment for accelerated materials development.
Exabyte.io is a cloud-based platform for materials modeling combining physics-based simulations techniques such as Density Functional Theory and Molecular Dynamics with data infrastructure, collaboration tools, and high-performance computational resources. The goal of the platform is to streamline the discovery and selection of new materials during the early-stage research and development.
Low-level simulations made accessible
Exabyte.io provides an accessible software environment that allows complex sequence encoding of low-level simulations into repeatable scientific workflows for enterprise applications, augmenting the currently standard command-line usage.
High fidelity and high throughput
The platform was previously scaled to tens of thousands of computing cores per customer within minutes to perform the calculations in a high-throughput rapid fashion, or, alternatively, improve the fidelity of the results using distributed memory calculations.
Role-based and team-based collaboration and professional networking tools allowing users to design materials and workflows together, review and curate each others work, and securely share results to make intelligent decisions faster.
Exabyte.io platform contains all required tools for computational materials design from nanoscale:
- Computer-aided design tools, allowing users to visualize and design new atomic arrangements, including surfaces, supercells and more.
- Atomistic modeling engines. The platform presently supports Vienna Ab-initio Simulations package (VASP), Quantum ESPRESSO, through its modular workflow designer interface, and LAMMPS, GROMACS, CP2K, NWChem via command-line.
- Structured data and statistical modeling. All data, including materials, their properties, simulation workflows, and jobs, is structured and organized to aid with further extraction. The platform supports Materials Project and the Inorganic Crystal Structure Database (ICSD) as the sources for structural information and has a "proof-of-concept" support for statistical predictive models.
- Representational state transfer (RESTful) application programming interface (API) for users to analyze the data produced by calculations and interact with the platform programmatically
Currently Exabyte.io has early use cases for electronic materials (semiconductor), alloys, catalysts, polymers, and composite materials. The company is actively working to support other material classes and simulation techniques in the near term.