How do I handle high-volume flows (1M+ messages/day)?
IBM Monitoring Profiles support high throughput with minimal overhead (<3% latency impact). For 1M messages/day (694 messages/minute average, 2,000 peak):
MQ Queue: MAXDEPTH 500,000 messages (~10GB) buffers 6-8 hour peak volumes. Configure queue on SSD storage for performance.
Nodinite Agent: Deploy on dedicated server (4-core, 16GB RAM typical). Agent processes 100-500 messages/second (MQGET batch size 100, parallel workers 8-16).
Monitoring: Configure [Non-Events Monitoring][Non Events Agent] to alert on volume anomalies: "Expected 40,000 messages/hour 9AM-5PM, seeing only 5,000" = possible broker stopped. Or: "Expected 40,000/hour, seeing 120,000" = possible duplicate processing bug.
Scalability: Tested to 50M events/day per Nodinite instance (distributed across multiple agent instances). For higher volumes, deploy multiple Nodinite environments (by business unit, by geography).
Related Questions
See all FAQs: [Troubleshooting Overview][]