How to Deploy Oracle 23ai Database on OCI – Complete DBA Implementation Guide.

Introduction

Oracle 23ai is Oracle’s latest AI-enabled database release designed for modern enterprise workloads, AI applications, vector search, JSON-relational architecture, and autonomous operations. Deploying Oracle 23ai on Oracle Cloud Infrastructure (OCI) provides DBAs with high performance, scalability, and enterprise-grade security.

In this guide, we will deploy Oracle 23ai Database on OCI from scratch using:

• Virtual Cloud Network (VCN)

• Public and Private Subnets

• Network Security Groups (NSG)

• Oracle DB System

• SSH Access Configuration

This article is written from a real DBA operational perspective instead of a basic cloud tutorial.

Architecture Overview

Before deployment, let us understand the architecture.

Components Used

Component	Purpose
VCN	Private cloud network for OCI resources
Public Subnet	Used for Bastion / external access
Private Subnet	Used for Database deployment
NSG	Security-level traffic control
Internet Gateway	Internet connectivity
Route Table	Controls network routing
Oracle DB System	Hosts Oracle 23ai database
SSH Key Pair	Secure server access

Target Architecture

The deployment architecture follows enterprise security standards.

Laptop
   |
SSH
   |
Public IP
   |
OCI Bastion / Public Subnet
   |
Private Subnet
   |
Oracle 23ai DB System

Prerequisites

Before starting deployment ensure:

• OCI account is active

• User has Administrator privileges

• SSH client installed

• Oracle Cloud region selected

• OCI Compartment created

Step 1 – Login to OCI Console

Open OCI Console.

Navigate to:

Oracle Cloud Console → Identity & Security → Compartments

Create a new compartment.

Example:

Name: PROD-DBA-LAB
Description: Oracle 23ai Deployment Lab

Compartments help isolate cloud resources.

Step 2 – Create Virtual Cloud Network (VCN)

What is VCN?

VCN acts like a private data center network inside OCI.

Navigate:

Networking → Virtual Cloud Networks → Create VCN

VCN Configuration

Parameter	Value
Name	OCI-23AI-VCN
CIDR Block	10.0.0.0/16
DNS Resolution	Enabled

Click:

Create VCN

Step 3 – Create Internet Gateway

Navigate:

VCN → Internet Gateways → Create Internet Gateway

Configuration:

Parameter	Value
Name	OCI-IGW
Enabled	Yes

Purpose:

• Enables internet communication

• Required for package downloads and updates

Step 4 – Create Route Table

Navigate:

VCN → Route Tables → Create Route Table

Add route rule:

Destination	Target
0.0.0.0/0	Internet Gateway

This allows outbound internet access.

Step 5 – Create Subnets

We will create:

1. Public Subnet

2. Private Subnet

Step 5A – Create Public Subnet

Navigate:

VCN → Subnets → Create Subnet

Configuration:

Parameter	Value
Name	PUBLIC-SUBNET
CIDR	10.0.1.0/24
Type	Regional
Public IP	Enabled
Route Table	OCI-RT

Purpose:

• Bastion access

• SSH connectivity

• Jump server access

Step 5B – Create Private Subnet

Configuration:

Parameter	Value
Name	PRIVATE-DB-SUBNET
CIDR	10.0.2.0/24
Public IP	Disabled
Route Table	OCI-RT

Purpose:

• Secure Oracle Database deployment

• Internal traffic only

Step 6 – Create Network Security Group (NSG)

Why NSG?

NSG provides instance-level traffic filtering.

Navigate:

Networking → Network Security Groups

Create NSG:

Name: OCI-23AI-NSG

Step 7 – Configure NSG Rules

Ingress Rules

Source	Port	Protocol	Purpose
Your Public IP	22	TCP	SSH Access
Application Server	1521	TCP	Oracle Listener
Internal Network	5500	TCP	EM Express

Egress Rules

Allow all outbound traffic.

Step 8 – Generate SSH Key Pair

Linux / macOS

Run:

ssh-keygen -t rsa -b 4096

Windows (PowerShell)

ssh-keygen

Files generated:

File	Purpose
id_rsa	Private Key
id_rsa.pub	Public Key

Important:

Never share private keys.

Step 9 – Create Oracle 23ai DB System

Navigate:

Oracle Database → Base Database Service → Create DB System

Step 10 – Configure DB System

Basic Information

Parameter	Value
DB System Name	OCI23AI-DB
Compartment	PROD-DBA-LAB
Availability Domain	AD-1
Shape	VM.Standard.E5.Flex
OCPU	2
Memory	32 GB

Step 11 – Configure Networking

Parameter	Value
VCN	OCI-23AI-VCN
Subnet	PRIVATE-DB-SUBNET
NSG	OCI-23AI-NSG
Hostname	oci23ai-db

Step 12 – Database Configuration

Database Details

Parameter	Value
Database Version	Oracle 23ai
Database Name	ORCL23AI
PDB Name	ORCLPDB1
Character Set	AL32UTF8
Workload Type	OLTP

Step 13 – Upload SSH Public Key

Paste contents of:

id_rsa.pub

OCI injects this key into the server during provisioning.

Step 14 – Create the DB System

Click:

Create DB System

Provisioning typically takes:

20 to 40 minutes

Step 15 – Verify Deployment

After deployment verify:

Database Status

Available

Listener Status

lsnrctl status

Database Status

sqlplus / as sysdba

Then:

select name,open_mode from v$database;

Expected:

READ WRITE

Step 16 – SSH Access to Oracle Server

SSH Command

ssh -i id_rsa opc@<PUBLIC-IP>

Switch to Oracle user:

sudo su - oracle

Step 17 – Verify Oracle 23ai Features

Connect database:

sqlplus / as sysdba

Check version:

select banner_full from v$version;

Expected output:

Oracle Database 23ai Enterprise Edition

Step 18 – Configure Automatic Backups

Navigate:

DB System → Backup Configuration

Enable:

• Automatic backups

• Recovery window

• Object storage backup

Recommended:

Setting	Recommended Value
Backup Window	Night Time
Retention	30 Days
Incremental Backup	Enabled

Step 19 – Enable Monitoring and Alerts

OCI monitoring helps DBAs proactively identify issues.

Navigate:

Observability & Management → Monitoring

Create alerts for:

• CPU utilization

• Tablespace usage

• Memory pressure

• Backup failures

• Storage growth

Step 20 – Security Best Practices

Recommended Security Standards

1. Use Private Subnet for Databases

Never expose database servers directly to internet.

2. Restrict SSH Access

Allow SSH only from office or VPN IP.

3. Enable TDE

Transparent Data Encryption should remain enabled.

4. Rotate SSH Keys

Rotate keys periodically.

5. Enable Audit Policies

Monitor database login activities.

DBA Operational Validation Checklist

Validation	Status
Listener Running	Verified
Database Open	Verified
SSH Access	Verified
NSG Rules Applied	Verified
Backup Enabled	Verified
Monitoring Enabled	Verified

Common Deployment Issues

Issue 1 – SSH Timeout

Cause:

• NSG rule missing

• Port 22 blocked

Solution:

• Verify ingress rules

• Verify public IP

Issue 2 – Listener Not Reachable

Cause:

• Port 1521 blocked

Solution:

• Add NSG ingress rule

Issue 3 – Database Creation Failed

Cause:

• Insufficient quota

• Wrong shape

Solution:

• Verify tenancy limits

Real DBA Recommendations

Recommended Shapes

Environment	Shape Recommendation
Dev	VM.Standard.E4.Flex
Test	VM.Standard.E5.Flex
Production	Exadata / DenseIO

Why Oracle 23ai on OCI?

Key Benefits

• AI-ready database engine

• Integrated vector search

• High scalability

• Enterprise-grade HA

• Built-in security

• Automated patching

• Cloud-native architecture

OCI Cost Optimization Guide for Database Workloads

Cloud adoption is growing rapidly, but many organizations migrating Oracle databases to Oracle Cloud Infrastructure (OCI) often face an unexpected challenge: rising monthly cloud bills.

In most environments, database workloads consume a major share of cloud resources through:

• Compute instances
• Block volumes
• Backup storage
• Data transfer
• Monitoring and logging services

This guide explains practical cost optimization methods specifically for Oracle database workloads on OCI.

Why OCI Database Costs Increase

The most common reasons for high OCI bills are:

1. Oversized Compute Instances

Many teams migrate on-premises workloads to OCI using the same sizing assumptions.

Example:

• Production database requires 8 OCPUs
• Team provisions 32 OCPUs “for safety”

Result:

• 4x unnecessary compute cost

Recommendation:
Monitor:

• CPU utilization
• Memory usage
• Load trends

Target:

• CPU average utilization between 40–70%

2. Idle Non-Production Databases

Development, UAT, and testing databases often run 24/7 unnecessarily.

Typical issue:

• Dev DB active only during office hours
• Still billed for full month

Cost Optimization Strategy

Schedule automatic shutdown/startup.

Example schedule:

• Start: 8 AM
• Stop: 8 PM
• Weekends off

Potential savings:

• 50–65% on non-prod compute costs

3. Storage Overprovisioning

Many OCI environments allocate excessive block storage.

Common pattern:

• 2 TB allocated
• 500 GB actually used

Best Practice

Review:

SELECT tablespace_name,
       ROUND(SUM(bytes)/1024/1024/1024,2) size_gb
FROM dba_data_files
GROUP BY tablespace_name;

Actions:

• Resize unused volumes
• Archive old data
• Move historical backups to cheaper storage tiers

4. Backup Storage Cost Explosion

RMAN backups accumulate quickly.

Typical issue:

• Daily full backups retained for 90+ days

This increases:

• Object storage usage
• Archive costs

Recommended Backup Policy

Production:

• Weekly full backup
• Daily incremental backup
• Archive log backup every 30 mins

Retention:

• 14–30 days online
• Older backups archived

Example RMAN:

CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 14 DAYS;
DELETE OBSOLETE;

5. Unused Block Volumes and Snapshots

After migrations or server rebuilds:

• Old block volumes remain attached
• Snapshots never deleted

Monthly hidden cost source.

Audit Checklist

Review:

• Unattached block volumes
• Old boot volumes
• Snapshot age > 30 days

Delete if unused.

6. High Logging and Monitoring Costs

OCI Logging and Monitoring can grow silently.

Common issue:

• Debug logs retained indefinitely

Best Practices

Reduce retention:

• Dev logs: 7 days
• Test logs: 14 days
• Prod logs: 30–60 days

Disable unnecessary verbose logging.

7. Wrong Database Deployment Model

Many organizations use expensive deployment types unnecessarily.

Compare:

Workload	Recommended Option
Small dev DB	Compute VM + Standard DB
Enterprise HA	Exadata / RAC
Variable workload	Autonomous DB
Archive/reporting	Lower compute shape

Choose based on actual workload.

8. Network Egress Charges

Cross-region traffic increases costs.

Examples:

• Backup replication
• Data Guard sync
• Application traffic

Reduce Cost By

• Keeping workloads in same region
• Reviewing outbound traffic
• Compressing backup transfers

9. License Cost Optimization

For BYOL environments:

Review:

• Actual processor usage
• Edition requirements

Sometimes Enterprise Edition is used where Standard Edition is sufficient.

Potential savings can be significant.

10. Monthly OCI Cost Governance Framework

Implement monthly review.

Checklist:

Compute

• Idle instances
• CPU utilization
• Shape right-sizing

Storage

• Unused block volumes
• Snapshot cleanup
• Backup growth

Database

• License review
• Storage growth trend
• DR cost validation

Monitoring

• Log retention
• Alert efficiency

Sample Monthly Cost Review Script

Track storage growth:

SELECT owner,
       segment_type,
       ROUND(SUM(bytes)/1024/1024/1024,2) gb
FROM dba_segments
GROUP BY owner, segment_type
ORDER BY gb DESC;

Top space consumers can be archived or optimized.

Estimated Savings by Optimization Area

Optimization Area	Savings Potential
Auto shutdown non-prod	50–65%
Right sizing compute	20–40%
Backup retention cleanup	15–30%
Storage optimization	10–25%
Logging optimization	5–15%

Final Thoughts

OCI offers strong pricing flexibility, but cloud costs increase quickly without governance.

A DBA should monitor not only database health but also:

• Resource efficiency
• Backup growth
• Storage utilization
• Compute sizing
• DR cost impact

Cost optimization is now a critical DBA responsibility in cloud environments.

Keywords for SEO

• OCI cost optimization
• Oracle cloud cost reduction
• OCI database cost management
• Oracle DBA cloud optimization
• OCI storage optimization

Oracle Database@AWS – The Silent Shift to True Multicloud (A DBA’s Perspective)

 

Introduction

For years, “multicloud” has been more of a strategy slide than a reality. Moving databases between cloud providers often meant complex migrations, latency challenges, and operational overhead.

But with Oracle Database@AWS, Oracle is quietly redefining what multicloud actually means — not as integration, but as co-existence.

This is not just another partnership.
This is a fundamental shift in how DBAs will design architectures going forward.

What is Oracle Database@AWS (Beyond the Marketing)

At surface level, it sounds simple:

“Run Oracle databases inside AWS”

But the real innovation is:

Oracle brings its database infrastructure (Exadata, Autonomous DB, tooling) directly into AWS data centers.

This means:

• No traditional migration
• No cross-cloud latency
• No re-platforming effort

Key Insight:
This is not “OCI connecting to AWS” — this is OCI running inside AWS.

Why This Matters (Real DBA Problems Solved)

Let’s break it from a practical DBA angle.

Traditional Challenges

• Data gravity → hard to move TB/PB data
• Network latency between AWS apps & OCI DB
• Licensing complexity
• Different monitoring & tooling

With Database@AWS

• Applications stay in AWS
• Database runs with native Oracle performance
• Same tools: RMAN, Data Guard, AWR
• Minimal architecture changes

Result: You bring the database to the application, not the other way around

Reference Architecture (Simplified)

6

Flow:

1. Application hosted in AWS (EC2 / EKS / Lambda)
2. Oracle Database deployed via Database@AWS
3. Internal high-speed network (no internet routing)
4. Unified identity & access control

Hidden Advantage:
No need for VPN / FastConnect between OCI and AWS

Key Components You Should Know

1. Exadata Database Service in AWS

• Same Exadata performance
• Smart scans, storage indexes
• Ideal for high-performance OLTP & DW

2. Autonomous Database

• Self-patching
• Self-tuning
• Minimal DBA intervention

But real talk: Still requires DBA governance for critical systems

3. Unified Security Model

• IAM integration across AWS + OCI
• Encryption by default
• Works well with compliance-heavy workloads

4. Native Tooling Continuity

No need to relearn:

• RMAN
• Data Guard
• OEM / Cloud Monitoring

This is huge for Oracle DBAs transitioning to cloud

Real-World Use Case (Unique Scenario)

Scenario: Financial Application Modernization

Before:

• App in AWS
• Oracle DB on-prem
• High latency + expensive network

After Database@AWS:

• App remains in AWS
• Oracle DB deployed via Database@AWS
• Zero migration downtime approach

Outcome:

• 40–60% latency reduction
• No code change required
• Licensing optimized

Final Thoughts

Oracle Database@AWS is not just a feature — it’s a strategy shift.

For DBAs, this means:

• Less migration work
• More architecture decisions
• More relevance in cloud strategy

If you ignore multicloud now, you’ll be catching up later.

OCI Generative AI Expansion (Cohere + AI Services) – 2026 Complete Guide

 

Introduction

Generative AI is rapidly transforming cloud operations, and Oracle Cloud Infrastructure (OCI) is aggressively expanding its AI ecosystem in 2026.

With the integration of Cohere foundation models, AI Agents, and multi-model support, OCI is positioning itself as a multi-model enterprise AI platform.

In this blog, we’ll cover:

• Latest OCI Generative AI updates (2026)
• Cohere model expansion
• AI services ecosystem
• Real DBA use cases

What is OCI Generative AI?

OCI Generative AI is a fully managed service that provides:

• Large Language Models (LLMs)
• Text generation, summarization, embeddings
• API-based integration

It allows enterprises to:

• Use pretrained models
• Fine-tune custom models
• Deploy AI securely in cloud

Key point:

OCI offers enterprise-grade AI with security, privacy, and scalability

2026 Major Expansion – Cohere Models

OCI recently expanded its AI ecosystem with new Cohere models, including:

 New Models Added:

• Command A Vision (multimodal AI)
• Command A Reasoning (advanced reasoning AI)

These models enable:

• Image + text understanding
• Complex decision-making
• Agentic AI workflows

 Update highlight:

OCI now provides multiple Cohere models via a unified API experience

Why Cohere Integration Matters

Cohere models bring:

 1. Enterprise Security

• Data is not used for training
• Full IAM integration

 2. High Accuracy & Reasoning

• Better contextual understanding
• Ideal for enterprise queries

 3. Multimodal Capabilities

• Text + image inputs
• Advanced analytics

 Multi-Model Strategy (OCI’s Biggest Strength)

OCI is not limited to one AI model.

 It supports:

• Cohere models
• Meta Llama models
• Upcoming Google Gemini

 Big advantage:

OCI gives model choice + flexibility for enterprises

 OCI Generative AI Ecosystem (2026)

1. Generative AI Service

Core platform for:

• Chat
• Summarization
• Embeddings

2. Generative AI Agents

OCI introduced AI Agents (very important update )

 These agents:

• Use LLM + enterprise data
• Provide context-aware responses
• Automate workflows

Key capability:

AI agents combine LLMs with enterprise data for intelligent automation

3. AI Vector Search (Database Integration)

• Enables semantic search
• Integrated with databases

 Example:

• Search logs using meaning, not keywords

4. AI Services Suite

OCI also includes:

• Speech AI
• Vision AI
• Document Understanding
• Language AI

DBA Use Cases 

1. SQL Query Generation

Input:
“Show top 10 slow queries”

Output:
AI generates optimized SQL

2. Performance Analysis

• Analyze AWR reports
• Suggest tuning actions

3. Log Analysis Automation

• Detect errors in logs
• Generate root cause summary

4. Security Monitoring

• Identify suspicious DB activity
• AI-driven anomaly detection

5. Backup & Alert Automation

Using AI Agents:

• Trigger alerts
• Generate reports
• Suggest fixes

Real-Time Use Case

Scenario:

Your production DB is slow.

With OCI Generative AI:

1. Feed AWR report
2. AI analyzes performance
3. Suggests:
   • Index creation
   • Query rewrite
   • Resource scaling

Result:

• Faster troubleshooting
• Reduced manual effort

Architecture (Simple View)

User → OCI Generative AI API → Cohere Model
     → AI Agent → Enterprise Data (DB/Logs)
     → Response (Insights / SQL / Summary)

Key Benefits for Enterprises

Productivity Boost

• Automates repetitive DBA tasks

Enterprise Security

• Data remains private

Faster Decision Making

• Real-time AI insights

Cost Optimization

• Reduce manual effort & errors

OCI vs Other Clouds (Quick Insight)

Feature	OCI	Others
Multi-model support		Limited
Enterprise security	Strong	Moderate
AI + DB integration	Native	Partial
Pricing flexibility	High	Medium

Future Roadmap

OCI is moving towards:

• Fully autonomous AI operations
• AI-driven cloud management
• Self-healing databases

AI will become:

“Default layer in all OCI services”

Conclusion

The OCI Generative AI expansion with Cohere is a major step forward in enterprise AI.

It enables:

• Smarter automation
• Faster DBA operations
• Secure AI adoption

If you’re working on Oracle Cloud Infrastructure, this is the next big skill to learn in 2026.

Exploring OCI Resource Analytics — A Next-Gen Cloud Inventory & Analytics Service

In large cloud environments, visibility and governance are huge challenges. When your Oracle Cloud Infrastructure (OCI) footprint spans multiple regions, tenancies, and service types, tracking what’s deployed, how resources relate to each other, and whether everything is compliant becomes a full-time job.

This is where OCI Resource Analytics comes in — a relatively new OCI service that provides a centralized, near-real-time inventory of your cloud resources with rich analytics, SQL access, graph visualizations, and customizable dashboards. In 2026, this service has grown from early previews into a capable platform for cloud teams.

What Is OCI Resource Analytics?

OCI Resource Analytics (RA) is essentially a cloud inventory + analytics platform built natively on OCI using an Autonomous Data Warehouse (ADW) and optionally Oracle Analytics Cloud (OAC) for dashboarding. It continuously ingests resource metadata from across all your OCI tenancies and regions, structures it into a relational model, and lets you explore it using SQL, graphs, and visual analytics.

At its core, RA answers questions like:

• Which compute instances exist across all tenancies?

• What networking resources are tied to specific compute workloads?

• What resources don’t have tags and might be unmanaged?

• Are my databases backed up? What dependencies exist between services?

Key Components of OCI Resource Analytics

OCI Resource Analytics is built around four major pillars:

1. Autonomous Data Warehouse (ADW)

RA provisions an ADW instance in your tenancy that serves as a centralized inventory database. All resource metadata, relationships, and configuration details are stored here in a structured, query-friendly schema.

• You can connect via SQL clients.

• Data is updated continuously to reflect near-real-time cloud state.

• Useful for automated inventory queries and custom analytics.

Think of this as your “cloud inventory lakehouse” — a data model designed for analytics, not just reporting.

2. Graph Visualization with Graph Studio

One of the standout features of RA is visual graphs that map how resources relate to one another.

• Compute ➜ network ➜ storage ➜ DB dependencies

• Visualize relationships across tenants/regions

• Drill into nodes for deeper insight

Graph Studio lets you visually connect the dots between resources — extremely helpful for troubleshooting and architecture reviews.

3. Prebuilt Dashboards via Oracle Analytics Cloud (OAC)

RA can optionally create an OAC instance with ready-made dashboards:

Common dashboards include:

Dashboard	Purpose
Resource Inventory	Complete overview of resources by type
Resources Without Tags	Identify untagged assets
Compute & Tag Insights	Correlates compute instances with tagging
Load Balancer & Networking	Zoom into networking resources
Database Insights	Database resource details & tags

These dashboards provide domain-specific views with filters and visual analytics — ideal for governance, auditing, and cloud cost reviews.

What’s New in 2026 (Why This Matters)

Oracle continues enhancing Resource Analytics throughout 2026 with new subject areas, dashboards, graph notebooks, and OAC improvements:

Expanded Data Coverage

The latest 2026 update includes ADW views and subject areas for:

• Kubernetes Engine (OKE)
• Object Storage usage
• Logging infrastructure
• OpenSearch clusters
• OCI Cache (Redis/DLM)
• Oracle Integration Cloud metadata
• Network Firewall details
• … and more.

This means Resource Analytics now covers more critical cloud services than ever before, letting you analyze:

• Storage usage patterns

• Logging and audit data infrastructure

• Cache performance and dependencies

• Kubernetes clusters and nodes

• Network security policies and firewall rules

Enhanced Dashboards & Filters

Dashboards in OAC now include filters like:

Tenancy Name
Compartment Name

…instead of just OCIDs, making the UI far more human-friendly.

This significantly improves the usability of analytics for teams managing large hierarchies of compartments and organizational units.

Why Resource Analytics Matters

Centralized Inventory Across Clouds & Regions

Gone are the days of manually gathering OCI resources from regions and tenancies. RA gives you one source of truth.

Accelerated Troubleshooting

Graph visualizations help you understand dependencies quickly — for example, how a compute instance is connected to networking and storage — speeding up root cause analysis.

Better Compliance & Governance

Unify your inventory for audits, compliance monitoring, and operational oversight. You can even enrich inventory data with custom datasets to build organization-specific reports.

Best Practices for Implementing RA

Here’s how you can get the most value from Resource Analytics:

1. Enable RA Across All Tenancies

Make sure every region and compartment you care about is included in the ingestion scope.

2. Integrate with Enterprise Dashboards

Connect OAC dashboards to internal governance systems — it’s perfect for compliance teams and cloud ops.

3. Automate Compliance Checks

Use SQL queries against the inventory lakehouse to automatically audit configurations, tagging compliance, and security hardening.

4. Blend with Internal Metadata

Join RA data with your org’s internal metadata (e.g., team ownership, project cost centers) for powerful cross-reference reporting.

Wrap-Up: A Cloud Inventory Game-Changer

OCI Resource Analytics is still a fresh innovation in 2026, evolving rapidly and filling a critical gap in cloud governance and inventory visibility. With its centralized data model, powerful dashboards, SQL access, and graph-based visualization, it’s much more than an inventory tool — it’s a solid foundation for enterprise-grade monitoring, compliance, and strategic cloud management.

If you manage OCI at scale — especially in multi-region or multi-tenancy environments — this service deserves your attention. Start exploring RA today and watch your cloud visibility skyrocket