Red Hat OpenShift AIRed Hat OpenShift AI
Monitor and optimize AI models and costs on Red Hat OpenShift AI.
Technology
- Product information
Overview
Combine Dynatrace end-to-end observability powered by AI with Red Hat OpenShift AI to track the performance, quality, and cost of your models and AI services at scale. You get precise, real-time answers across your generative AI applications — automatically and in full context.
This is intended for users who:
- Build, serve, or operate AI and ML models on Red Hat OpenShift AI.
- Need to control the cost and reliability of AI services in production.
Use cases
- Auto-discover and monitor AI models, tracking latency, throughput, token usage, and resource consumption.
- Correlate model consumption, GPU utilization, and cost data to optimize AI infrastructure spend.
- Analyze prompts, outputs, and responses to detect drift, bias, and hallucinations.
- Enforce SLOs and error budgets to keep AI services reliable.
Get started
Set up full-stack observability for your GenAI applications using OpenLLMetry (built on OpenTelemetry) to feed end-to-end insights into Dynatrace. Follow the setup steps in Dynatrace Documentation.
Details
Works with generative AI applications built and served on Red Hat OpenShift AI. Ingests traces, logs, and business events through OpenTelemetry, so it applies to any model output without changes to your application logic.
