Quick Start

Get Aether running in 5 minutes.

1. Create Configuration

Create aether.yaml in your working directory:

services:
  torch:
    base_url: "https://your-torch-server.org"
    username: "your-username"
    password: "your-password"

  dimp:
    url: "http://your-dimp-server:32861/fhir"

pipeline:
  enabled_steps:
    - torch
    - dimp

jobs_dir: "./jobs"

Replace URLs and credentials with your actual server details.

2. Run a Pipeline

aether pipeline start your-query.crtdl

Aether will:

1. Send your CRTDL query to TORCH
2. Wait for extraction to complete
3. Download the FHIR data
4. Pseudonymize it via DIMP
5. Save results in jobs/<job-id>/pseudonymized/

Output files use .ndjson.zst extension (zstd compressed).

3. Check Progress

# List all jobs
aether job list

# Check a specific job
aether pipeline status <job-id>

4. Resume if Needed

If a pipeline fails or pauses, resume it:

aether pipeline continue <job-id>

Alternative: Direct TORCH URL

If you already have a TORCH extraction URL, skip the CRTDL submission:

aether pipeline start "https://torch.example.com/fhir/extraction/result-123"

Aether auto-detects URLs containing /fhir/extraction/ or /fhir/result/ and polls them directly. See TORCH Integration for details.

Alternative: Local Import

Process local FHIR files instead of TORCH extraction:

services:
  local_import:
    dir: "/path/to/fhir/data"
  dimp:
    url: "http://your-dimp-server:32861/fhir"

pipeline:
  enabled_steps:
    - local_import
    - dimp

# Use config directory
aether pipeline start

# Or override with --dir flag
aether pipeline start --dir /other/path

Next Steps

• Configuration - All configuration options
• Pipeline Steps - Available steps
• CLI Commands - Command reference