2 Features of Oracle AI Data Platform

Oracle AI Data Platform is a modern data platform designed to simplify data ingestion, processing, and analytics at scale. It provides a seamless integration of compute, storage, and cataloging capabilities to enable efficient data management.

Key features of AI Data Platform include:

Workspace

A workspace in AI Data Platform acts as an isolated environment where users can manage and organize their data lake resources, including workflows, notebooks, and libraries. Workspaces enable efficient collaboration and governance by keeping resources grouped logically.

Compute

AI Data Platform provides scalable CPU and GPU compute resources for executing data processing and analytics workloads. Users can leverage Spark-based execution environments for high-performance processing, supporting batch and interactive workloads.

AI Compute

AI compute provides scalable resources for agents, both for testing and production. Once an agent is attached to an AI compute, agent developers can create sessions and start interacting with their agent. Multiple agents can be either attached (for testing) or deployed (for production) on a given AI compute.

Agents

Agents are end-to-end agentic applications. Agents are defined through a graph of steps represented by different types of nodes, such as agents and tools. Agents are defined through a no-code visual flow builder and through code via third-party libraries, such as LangGraph.

Notebook

AI Data Platform includes notebooks as an interactive development environment for writing and executing code. It supports Python and SparkSQL enabling users to transform, analyze, and visualize data directly within AI Data Platform.

Workflow

The workflow component allows users to define and orchestrate data pipelines made of notebooks, Python tasks, if-else, and other job tasks. Users can create, schedule, and monitor workflows for ETL, data transformations, and analytics automation.

Master Catalog

The Master Catalog serves as the central metadata repository for all structured and unstructured datasets within an AI Data Platform. It provides unified governance and data discovery, allowing users to search and manage datasets across different schemas and storage locations.

Catalog

A catalog in an AI Data Platform is a logical grouping of schemas, tables, volumes, and models, providing a structured way to organize datasets. Users can create multiple catalogs for different projects or teams to ensure effective data segmentation.

Knowledge Bases

Knowledge bases leverage Oracle AI Database 23ai Vector Search capability to store vector embeddings from documents stored in an AI Data Platform , such as PDF, DOCX, and HTML files. Through Oracle AI Database 23ai’s vector search capabilities, knowledge bases empower AI agents to perform semantic searches and retrieve semantically relevant documents.

Schema

A schema defines the structure within a catalog, organizing tables and views under a common namespace. Schemas help in logically structuring data for different applications and analytics workloads.

Table

A table in an AI Data Platform represents structured datasets that can be queried and processed. Tables support various storage formats, including Delta Uniform, ensuring compatibility with multiple query engines.

View

A view is a virtual table in an AI Data Platform that provides a queryable representation of data stored in underlying tables. Views allow for simplified access to transformed datasets without requiring data duplication.

Volume

A volume is a storage abstraction in an AI Data Platform that provides a managed space for persisting raw, processed, and curated data. It supports efficient data access and integration with Object Storage.

Auto Populate

The Auto Populate feature simplifies metadata management by automatically detecting and registering new datasets located in OCI Object Storage. This reduces manual effort in keeping data catalogs up to date.

Role-Based Access Controls (RBAC)

AI Data Platform implements RBAC to enforce fine-grained access control across different resources. Users can define roles and permissions for workspaces, catalogs, datasets, agents, and knowledge bases to ensure secure collaboration.

Audit Log

Audit logs in AI Data Platform capture detailed records of user activities. These logs help monitor usage, ensure compliance, and investigate issues such as unauthorized access or configuration changes.

Three-Part Namespace

AI Data Platform adopts a three-part namespace (Catalog.Schema.Table) for accessing datasets, enabling a structured and consistent way to reference data across the platform. This standardization improves interoperability and ease of access.