Exabyte.io is the fastest way to design and develop new materials from concept to product. It is an all-in-one highly scalable framework for atomistic design, high fidelity simulations and large-scale data analytics. The links section below can direct reader to sources of more in-depth explanation about what we are building and why [1], quick demonstration of the scalability of our product [2] and example case studies [3].

This documentation contains information about how our product works. You can search by using the search field in the top left corner or just read on. In case you find that something is missing or if you still have questions after reading this documentation, please contact us.

Quickstart

In case you feel impatient, you can skip straight to our first tutorial summarizing first steps with exabyte.io. You will learn how to create and save your first material and run a simulation that predicts this material's electronic bandstructure.

Contents

Our users compute characteristic properties of materials through high-fidelity modeling. Therefore, the content of this documentation is divided into three main categories:

Materials

  • how to design, upload or import material geometries
  • supported characteristic properties (eg. band gap, formation energy, electron mobility) supported
  • advanced topics:
    • creation of combinatorial sets of materials for high-throughput screening,
    • overview of materials analytics (eg. how to make sense of all the characteristic properties)

Models & Methods

  • explanation of the models used to extract the properties of materials and their particular implementations (methods),
  • how to apply these models and run simulations
  • advanced topics:
    • editing simulation input,
    • dissecting convergence and relaxation algorithms,
    • calculation of compound characteristic properties (eg. formation energy)

Compute

  • how to submit a simulation job
  • how to set compute resources
  • job submission queues and urgency factors
  • advanced topics:
    • benchmarks and scaling
    • limits per compute queue
    • alternate connection methods:
      • secure shell terminal:
      • web-browser terminal:
      • remote desktop via VNC

Other parts explain data security, sharing and collaboration, billing/pricing and our terminology.

Highlighted features

Materials Design/Upload/Import

We let one develop material concepts from scratch by designing new geometries or reuse/extend prior works by uploading structures from disk or importing from trusted database sources. More information on the topic available on this page.

High performance and massive scalability

Materials simulations are often computationally demanding. We put extreme attention to performance and scalability of our product. More information available here and here.

Combinatorial screening

We enable automation of materials screening processes to make it easier to create leads, set-up and monitor simulations, and analyze the results. This page contains more information about combinatorial materials sets.

You can view our upcoming features here also.

How to get help

We have support team that is active 24/7 and will do our best to reply to your requests within 24 hours. Our support team can be contacted by phone, email, or the web during working hours Pacific Time:

Links

  1. What is materials discovery cloud, article
  2. How big is "cloud-scale", article
  3. High-throughput computer-aided discovery of new metallic alloys, case study