Welcome to Zarrtraj’s documentation!

Zarrtraj is a proof-of-concept for a new paradigm of trajectory storage and analysis in MDAnalysis by allowing users to store and read and write trajectory files in the H5MD format to and from the cloud.

This means users can interact with massive trajectory files without ever storing them on their own disk.

Contents:

• Getting Started
  • Installation
• YiiP Protein Example
• Walkthrough
  • Writing trajectories to cloud storage
  • Reading your H5MD file
• API Documentation
  • Example: Loading a .h5md file from disk
  • Example: Reading from cloud services
  • Example: Writing directly to cloud storage
  • Classes
• Performance Considerations
  • Set n_frames to the number of frames being written
  • Set precision=3
  • Use compressor=numcodecs.Blosc(cname="zstd", clevel=9)
  • Example
• ZarrMD specification
• Benchmarks