PIELake
OSDU-Native Data Foundation for Subsurface AI
An OSDU-native scientific data foundation built on Databricks, designed to unify, govern and expose complex subsurface datasets through interoperable services, domain data models and AI-ready access patterns.
Overview
Developed by Petrabytes, PIELake provides OSDU-aligned schemas, entitlements and domain data management services for complex subsurface information. It sits between the underlying Databricks platform and PIEScale applications.
PIELake and PIEScale are complementary offerings: PIELake supplies the scientific data foundation, while PIEScale supplies specialized applications, workflows and user experiences. Neither is developed by Databricks or IesBrazil.
Key capabilities
OSDU-Native Data Foundation
OSDU-aligned schemas, core services and entitlements organize scientific information with domain data management services. Domain-service availability and testing status vary.
Scientific Data Integration
Bring seismic, wells, geology, reservoir, production and geospatial datasets into a shared scientific data foundation using supported formats and domain models.
Data Quality & Governance
Validate source headers, metadata and values before transformation, assess data quality, and control access to governed scientific information.
Search & Discovery
Discover datasets through metadata search, spatial exploration and domain-aware queries, connecting wells, seismic and other subsurface information in context.
AI-Ready Data Access
Expose governed information through OSDU-compatible REST APIs, Model Context Protocol (MCP) and agent-oriented integration patterns for applications and AI workflows.
Databricks-Native Architecture
Petrabytes documents integration with Unity Catalog, Lakebase, Lakeflow and Genie as part of the underlying Databricks data intelligence platform.
Capabilities and integrations depend on the selected deployment, enabled services and commercial configuration; not every capability is available in every deployment.
Applications
Subsurface data modernization.
OSDU-based data platform architectures.
Seismic and well data integration.
Scientific data governance and discovery.
Integration with Databricks lakehouse environments.
Data preparation for analytics and domain AI agents.
A complementary architecture
Petrabytes describes three complementary architectural layers, with ready-to-use capabilities above the application layer.
- Foundation
- PIELake — OSDU-native schemas, entitlements, domain data management and governed access to subsurface information.
- Applications
- PIEScale® — Specialized subsurface data operations, configurable workflows and user experiences built on the scientific data foundation.
Databricks is the underlying data intelligence platform, providing processing, governance and analytics capabilities. PIELake and PIEScale remain Petrabytes offerings, not Databricks products.
OSDU compatibility
Petrabytes currently identifies the following OSDU release milestones for PIELake. These are manufacturer-stated support statuses, not independent certification or external validation.
- M25
- Supported
- M26
- Under testing
- M27
- Under testing
Versions under testing are not presented as fully supported. Petrabytes also marks Reservoir DDMS and RAFS as under testing; confirm required releases and domain services for the intended deployment.
Developer
Petrabytes develops PIEScale®, a scientific data platform for subsurface and energy data.
Official websiteTechnical resources
Interested in PIELake?
Speak with one of our specialists about PIELake and your project requirements.