Oracle Analytics, OAS & Data Warehousing – February 19th

Posted on By Adrian Ward

Welcome to another Oracle Analytics and Data Warehousing roundup — a focused summary of useful updates, blog posts, and technical resources across:

  • Oracle Analytics Cloud (OAC)
  • Oracle Analytics Server (OAS)
  • Autonomous Data Warehouse / 23ai
  • Fusion Data Intelligence (FDI) (soon to be renamed)
  • SQL & performance tuning content

🔷 Oracle Analytics (OAC & OAS) – Platform Updates & Governance

Recent updates across the Oracle Analytics ecosystem continue to focus on AI-assisted analysis and stronger governance controls.

🔹 AI-Enhanced Analytics

Oracle Analytics continues to expand AI-driven capabilities including:

  • AI-assisted insight discovery (clusters, trends, outliers)
  • Natural language interaction
  • Feedback loops to improve AI-generated responses
  • Embedded AI in the semantic layer

While most AI enhancements arrive first in OAC, OAS customers should monitor feature parity updates as Oracle continues aligning on-prem and cloud roadmaps.

🔹 Fine-Grained Semantic Model Permissions

A particularly important enhancement for enterprise deployments:

  • Separation of model creation, editing, and deployment permissions
  • Improved governance for larger BI teams
  • Better Dev → Test → Prod workflows

This is relevant for both:

  • OAC semantic models
  • OAS RPD-based environments (where governance and controlled deployment remain critical)

👉 Oracle Analytics Blog
https://blogs.oracle.com/analytics/

🔷 Oracle Analytics Server (OAS) Considerations

For organisations still running OAS (12c / 2023+ releases):

  • Continued emphasis on hybrid strategies (OAS + OAC coexistence)
  • Importance of RPD lifecycle governance
  • Patch and platform alignment with newer database versions (including 23ai)

Many customers are evaluating:

  • Lift-and-shift to OAC
  • Hybrid architecture
  • Retaining OAS for regulatory or data residency reasons

It’s worth keeping an eye on roadmap discussions in Oracle community forums.

👉 Oracle Analytics Community
https://community.oracle.com/products/oracleanalytics/categories/announcements

🔷 Fusion Data Intelligence (FDI) – Rebranding on the Horizon

Fusion Data Intelligence (FDI) continues to evolve and is expected to receive a new name as Oracle aligns branding around AI and data platform capabilities.

FDI remains central for:

  • Prebuilt analytics for Fusion Applications
  • Packaged KPIs and subject areas
  • Embedded AI and predictive capabilities
  • Tight integration with Oracle Autonomous Data Warehouse

Recent updates focus on:

  • Expanded subject areas
  • Enhanced AI-driven insight capabilities
  • Continuous data ingestion improvements
  • Stronger semantic layer modelling

As the rename approaches, expect closer alignment with Oracle’s broader AI Data Platform messaging.

If you’re running Fusion Apps, FDI (whatever the new name becomes) remains strategically important.

🔷 Autonomous Data Warehouse & 23ai

Recent data warehousing discussions focus on:

  • Performance optimisation in Autonomous Database
  • 19c → 23ai upgrade paths
  • AI-native database capabilities
  • Enhancements around caching and workload management

Oracle continues pushing toward:

AI-native database infrastructure with reduced administrative overhead

For analytics teams, this matters because:

  • Faster queries = better dashboard performance
  • Better caching = improved concurrency
  • Smarter automation = fewer DBA bottlenecks

👉 Oracle Data Warehousing Blog
https://blogs.oracle.com/datawarehousing/

💡 SQL Tip of the Week – Use Analytic Functions Instead of Self-Joins

When working with Oracle Analytics (OAC/OAS) against Autonomous Data Warehouse, performance often depends on how efficiently you write SQL.

A common anti-pattern in warehouse environments is using self-joins to calculate rankings or running totals.

Instead, use analytic functions.

Example: Ranking Sales per Region

Instead of:

SELECT a.region,
       a.sales_rep,
       a.sales_amount
FROM sales a
JOIN sales b
  ON a.region = b.region
 AND a.sales_amount <= b.sales_amount
GROUP BY a.region, a.sales_rep, a.sales_amount
HAVING COUNT(*) <= 3;

Use:

SELECT region,
       sales_rep,
       sales_amount,
       RANK() OVER (
           PARTITION BY region
           ORDER BY sales_amount DESC
       ) AS sales_rank
FROM sales;

Then filter:

SELECT *
FROM (
    SELECT region,
           sales_rep,
           sales_amount,
           RANK() OVER (
               PARTITION BY region
               ORDER BY sales_amount DESC
           ) AS sales_rank
    FROM sales
)
WHERE sales_rank <= 3;

Why this matters for OAC / OAS:

  • Analytic functions are optimised in Oracle Database
  • Better execution plans in ADW
  • Less data movement
  • Cleaner logical SQL in semantic models
  • Improved dashboard performance

If you’re modelling in OAS RPD or OAC Semantic Modeler, pushing ranking logic down to the database layer using analytic functions often results in better scalability.

🔎 TL;DR This Week

  • AI capabilities continue expanding in Oracle Analytics
  • Fine-grained semantic model governance improves enterprise control
  • OAS remains relevant in hybrid strategies
  • Fusion Data Intelligence (FDI) is evolving and will be renamed soon
  • 23ai and ADW performance enhancements continue
  • Use analytic SQL functions to improve warehouse query performance
Weekly News