Atomio technical guide

Atomio is a syndication server which can hold multiple Atom syndication feeds. It has an API to create/delete feeds and entries, and provides resource level security compatible with Ontoserver’s resource level security.

It is designed as a companion to Ontoserver to:

  • Keep a static record of content loaded/available to a server, and;
  • Keep static snapshots of releases from an authoring server to make available for publication.

Being a static source of truth for releases enables “production” read-only Ontoserver instances to be ephemeral, removing the need to backup and maintain state.

System Requirements

Atomio uses a simple in-memory database which is stored on disk to record the feed state, along with the artefacts in the feeds. As a result it can require a large amount of disk space if it stores many very large artefacts. However typically it serves very few requests infrequently, and the performance of these requests is not critical.

Memory512MB minimum
1GB recommended
Disk5GB minimum*
CPU1 CPU
*Required storage depends greatly on the number and size of artefacts to store. This is usually heavily influenced by SNOMED CT binary indexes which are by far the largest artefact at 1.5GB each

Core concepts

Creating release candidates

Configuration

API

Deploy new Atomio version

Change log

Infrastructure and operations