Data Architecture
Overview of how data is stored and materialized within Method.
Overview
The Data Architecture in Method is designed to serve a number of high level technical use cases:
- Construct and maintain an Ontology, or connected knowledge graph, from heterogeneous input sources
- Organize and materialize data in different form factors to equip AI to be productive and reliable
- Track history of resources, specifically property and relationship changes
- Deliver interoperable input and output boundaries so developers can add custom sources and export data to their own systems
The following architecture diagram shows how data flows from Ledger (bottom), up to Ontology (knowledge graph), to then serve various functions within Method Platform.
Importantly, Ledger is an append only transaction store, that is very general in nature. This helps with heterogeneous input sources and tracking history. The architecture materializes data in other form factors (e.g. graph database, vector embeddings), that serve data to AI and users alike for very specific functions. Moving from bottom to top, data goes from a general state to a complex state; everything is kept in sync to support the real time nature of security work.
