Scalable Scientific Data Ingestion Pipeline with Laravel

Design a robust Laravel microservice for ingesting large-scale scientific datasets from multiple instrument sources (mass spectrometry, genomic sequencers, electron microscopy). The system must handle concurrent data streams, validate complex JSON/XML payloads, implement idempotent processing, and provide automatic error recovery. Include a comprehensive logging mechanism that tracks each data point's provenance, transformation steps, and potential anomalies. Implement rate limiting, circuit breaker patterns, and ensure the solution can process minimum 500GB/day of scientific instrument data with less than 100ms latency.