Skip to main content

Interactive Format Transformation Explorer

See format transformation in action! Use the interactive explorer below to step through 4 stages of converting JSON to analytics-optimized formats (Avro, Parquet). Watch how columnar storage reduces costs by 90% and makes queries 10x faster.

How to Use This Explorer​

  1. Navigate using arrow keys (← →) or click the numbered stage buttons
  2. Compare the Input (left) and Output (right) showing transformations
  3. Observe size reductions and query performance improvements
  4. Inspect the YAML code showing format conversion logic
  5. Learn from the stage description explaining the benefits

Interactive Format Transformation Explorer​

JSON Input

Most APIs and logs produce JSON, but JSON is inefficient for analytics: no schema enforcement, large file size (verbose keys/values), slow parsing, poor compression.

Use ← → arrow keys to navigate
📥Input
{
"sensor_id": "sensor-42",
"temperature_celsius": 23.5,
"humidity_percent": 45.2,
"timestamp": "2024-01-15T10:30:00Z"
}
📤Output
❌ JSON Drawbacks:
Size: 145 bytes (verbose keys)
Schema: none (typos allowed)
Compression: ~50% (gzip)
Query: full scan required
Added/Changed
Removed
Completed Step
Current Step
Not Done Yet
📄New Pipeline Stepstep-0-json.yaml
# JSON input (verbose, uncompressed)

Try It Yourself​

Next: Set up your environment to build format transformation pipelines