← All stories
● Covered by 1 source · 1 reportMedium impact1 neutral

Netflix Develops Knowledge Graph for AI-Driven Observability to Accelerate Incident Investigations

🔄 Updated 8d ago
New to BrevFeed? We gather this story from every outlet covering it into one summary — ranked by real-world impact, not just the latest headline — so you never miss what matters. What is BrevFeed? →

Key points

  • Netflix is developing an operational knowledge graph.
  • It connects applications, incidents, deployments, alerts, and infrastructure.
  • Aims to reduce investigation time from hours to minutes.
  • Uses RDF, OWL, SHACL, SPARQL, and QuipuDB for storage.

Unifying Operational Data for Faster Investigations

Netflix is creating an ontology-driven knowledge graph to improve its operational observability. This system is designed to accelerate production investigations by providing AI agents with a unified view of its operational environment. The goal is to connect disparate information about applications, incidents, deployments, alerts, and infrastructure dependencies, which are currently scattered across various systems.

Addressing Time-Consuming Incident Resolution

Netflix processes over 2 billion requests daily and 38 million real-time logging events per second, supporting up to 65 million concurrent streams. Despite this scale, investigating an application regression can take approximately four hours, with a significant portion of that time spent manually reconstructing relationships between systems. The knowledge graph aims to make these relationships explicit, laying the groundwork for automated root-cause analysis and self-healing infrastructure.

Leveraging Existing Data with a New Model

While Netflix already collects metrics, events, logs, and traces (MELT), these signals are distributed across different systems like alerting, incident management, and deployment tools. The company identified a need for a consistent method to describe how individual components relate to each other. The Observability Ontology provides this common model, defining operational entities, their properties, and their interrelationships.

Technical Implementation and Data Ingestion

The knowledge graph utilizes established standards including RDF, OWL, SHACL, and SPARQL. The resulting graph data is stored in QuipuDB, Netflix's distributed RDF database built on Rust's redb storage engine. The graph is populated by 'enrichers,' which are components that consume information from operational systems and transform it into structured subject-predicate-object triples. Some enrichment processes are deterministic, while others use LLMs and probabilistic techniques for less structured information, with strict validation applied to outputs.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →

The daily brief

One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.

One email a day. Unsubscribe in one click, any time.

Today's brief

Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.

~4 min · 3 stories · Nov 21

▶ Play today's brief Listen on Spotify

New every morning, and the back catalogue is archived by date.

Primary sources

GitHub intigraph/quipu

Reporting from

Netflix is building an ontology-driven knowledge graph to unify operational data, aiming to reduce the time engineers spend investigating production issues. This system connects information from various operational tools, providing a shared understanding for AI agents and enabling automated root-cause analysis. The initiative addresses the challenge of manually correlating data scattered across different systems during incident resolution.