Getting Started
Limited Availability
Anomaly Detection is currently in Limited Availability. Contact support@hydrolix.io to learn more about access.
This page covers how to enable Anomaly Detection with a Hydrolix cluster.
To enable Anomaly Detection for your Hydrolix deployment, contact customer success at support@hydrolix.io with as much of the following as possible. The Hydrolix customer success team can also work with you to answer these questions:
- Use case: Describe your monitoring objectives (For example: CDN performance)
- Metrics: Identify which metrics you want to monitor for anomalies
Summary Table Required
Anomaly Detection only works on summary table data as the amount of data in raw tables inflates the cost of LLM analysis. If you don't already have a summary table for your raw data, mention this when contacting customer success — the team can help you design an appropriate summary table structure for your data and use case.
Bring your own LLM⚓︎
If you're using your own LLM, please be ready to provide:
- API credentials for your AI model
- Model name/identifier
Supported LLMs⚓︎
- GPT-4o — OpenAI
- GPT-5 — OpenAI
- Claude Sonnet 4 — Anthropic
- Gemini 1.5 Pro — Google
- Llama 3 70B — Meta
- Qwen3 — Alibaba
- Mixtral 8x22B — Mistral AI
- DeepSeek R1 — DeepSeek
- Grok-1 — xAI
Your Customer Success Engineer will work with you to configure Anomaly Detection based on your specific requirements and use case, and will let you know when your anomaly detection dashboards are available ready.
See Using Anomaly Detection for how to use Anomaly Detection once it's enabled.