🧠 Oracle Data Management Strategy Summary

🔁 Slogan: "Simply Complete and Completely Simple"

Oracle’s strategy focuses on delivering a comprehensive and easy-to-use data platform that
supports modern application development and analytics.

🔧 Key Pillars of the Strategy

1. Modern App Development Made Simple

 Oracle embraces modern development paradigms with a converged database model.

 Supports multiple data types and workloads (e.g., JSON, graph, spatial, blockchain,
unstructured data).
 Eliminates data fragmentation by enabling applications to use all data types in one
platform.
 Facilitates low-code development, APIs, geo-distribution, and event-driven
architecture.

2. Converged Database Architecture

 A single database engine handles:

o Transactional workloads
o Analytical workloads
o Machine Learning, IoT, and data lakes
 Enables unique cross-data-type queries and value creation.

3. Autonomous Database (ADB)

 Fully managed, automated cloud database.

 Features:
o Auto-scaling
o Auto-indexing
o Auto-patching and tuning
 New: MongoDB API support
 Simplifies data engineering and analytics with self-service tools.

4. Operational Strength for Mission-Critical Apps

 Oracle delivers high availability, built-in disaster recovery, backup, and security.
 Reduces need for complex sharding or manual data protection.
 Transparent and automated management of complex enterprise applications.
🚀 Big Picture: Oracle’s Commitment

Oracle offers:

 A scalable, secure, and highly available platform

 Unified support for all development styles, workloads, and data types
 A strategy built on years of proven technology, now accessible with modern cloud
simplicity

📌 Tagline Recap

“Simply complete” – Covers all use cases and technologies

“Completely simple” – Easy to adopt, deploy, and manage

☁️Oracle Database Offerings Overview

Oracle provides multiple deployment models:

 On-Premise: You manage both infrastructure and database

 DBCS (Database Cloud Service) / Exadata Cloud Service: Oracle manages
infrastructure; you manage the database (co-managed)
 Autonomous Database (ADB): Oracle fully manages both infrastructure and database
(fully managed)
 Third-Party Cloud: E.g., Amazon RDS for Oracle—typically more expensive and less
optimized

🔑 Why Autonomous Database (ADB)?

Fully Managed Cloud Database with:

 Automated provisioning, patching, tuning

 Built-in best practices for security, performance, and availability
 Elastic scaling and high availability
 Significantly lower total cost of ownership (TCO)

🧾 Wikibon Research:

 ADB = ~50% cheaper than on-prem

 Amazon RDS = ~50% more expensive than on-prem (with DR and multi-zone)
🔄 Migration to ADB Made Easy

🛠 Migration Path:

1. ADB Compatibility Tool: Scans your current database for unsupported features
2. Data Pump Export/Import: Move schema/data from on-prem to ADB via Oracle Cloud
Object Store
3. Preserve tuning/indexing if already optimized

🚀 New: Oracle Database Migration Service

 Automates the above steps

 Supports migrations from other clouds too
 Certified support for packaged apps: JD Edwards, PeopleSoft, Siebel, EBS (coming
soon)

👩‍💻 Low-Code Development with APEX

💡 Oracle APEX (Application Express):

 Oracle’s low-code platform for building data-driven applications

 10x–50x faster than traditional development
 Automatically handles:
o Middle-tier logic
o State management
o Connection pooling
 Highly productive for forms, reports, faceted search, and UI customization

✅ Real Customer Example:

Wilson Truck Lines rebuilt a 3-month Amazon-coded app in 2 days using APEX + ADB

📊 More Than a Data Warehouse

🔍 Autonomous Data Warehouse (ADW) supports:

 Massively parallel querying

 Self-service data tools: Drag-and-drop ETL, transformation, data loading
 Advanced analytics:
o Graph
o Spatial
o Machine Learning (AutoML)
 Data lake integration with Oracle/Object storage
  Third-party BI tool support (e.g., Tableau, BusinessObjects)

⚽ Seattle Sounders FC + English Premier League use ADW for advanced sports analytics

💸 Business Incentives for Moving to ADB

1. BYOL (Bring Your Own License)

Reuse existing Oracle licenses in OCI
2. Cloud Lift Services
Free expert help to move on-prem Oracle DB to ADB
3. Support Rewards Program
o Earn $0.25–$0.33 OCI credits per $1 spent
o Apply credits to reduce Oracle support costs

🎁 Try It Out for Free

 Always Free Autonomous Database

Get hands-on experience without commitment
 LiveLabs (developer.oracle.com/livelabs)
Guided labs and exercises to explore ADB, APEX, AutoML, and more

✅ Final Takeaways
Key Value Description

Lower TCO Up to 50% cost savings over on-prem and Amazon RDS

Fully Managed Oracle automates everything from patching to tuning

High Productivity Tools like APEX and AutoML accelerate app and analytics

Migration Help Free tools, services, and compatibility checks

Business Incentives BYOL, Support Rewards, and Cloud Lift make it affordable

☁️Oracle’s Approach to Multicloud and Hybrid Cloud Strategy

🧩 The Cloud Adoption Landscape

 Only <20% of mission-critical workloads have migrated to the cloud—but that’s

changing fast.
  Nearly 70% of enterprises are actively migrating or planning to migrate on-prem
workloads to the cloud.
 Challenge: Many orgs are locked into on-prem (e.g., Oracle Engineered Systems) despite
having a cloud strategy.

🌐 Multicloud Strategy Defined

 Multicloud = Using multiple cloud providers for different workloads.

 Common reasons to go multicloud:
o ✅ Best-of-breed solutions for specific workloads
o ✅ Redundancy and risk mitigation
o ✅ Reduced latency via regional clouds
o ✅ Compliance with data sovereignty regulations
o ✅ Avoid vendor lock-in
o ✅ Multicloud economics via virtual cloud infrastructure—no physical hardware
required

🔗 Secure Multicloud Connectivity

 Oracle enables multicloud connections without exposing data to the public internet.
 Private, predictable, and secure connections using:
o Megaport Cloud Routers (MCR)
o Oracle Cloud Infrastructure FastConnect
o Virtual Cross Connect (VXC) = Private Ethernet link across clouds
 Example: OCI ↔ AWS VPC, OCI ↔ Azure, enabling hybrid deployments

🏗️Oracle’s Hybrid Cloud Innovations

Oracle offers cloud-controlled services deployed on-premises to solve sovereignty, latency,

and control needs:

1. Dedicated Region

 Public cloud deployed inside customer’s data center

 Full OCI services + Sovereignty + Locality
 Managed and upgraded by Oracle
2. Exadata Cloud@Customer & ADB@Customer

 Oracle's Exadata and Autonomous DB delivered on-prem as a service

 Combines cloud benefits with local deployment

3. Global Cloud Reach

 30+ OCI public regions

 Automated region deployment, supporting:
o High availability
o Regional disaster recovery
o Resilient micro-deployments

🤝 Oracle + Microsoft Azure: Interconnected Multicloud

 Jointly engineered solution with Azure + Oracle Cloud:

o FastConnect + ExpressRoute: <2ms latency, no intermediaries
o Unified Identity (SSO) with automated provisioning
o Collaborative workload support: e.g., Oracle Apps on Azure, Oracle DB on
OCI
o Integrated support model across both clouds

💡 Strategic Takeaways

Benefit Description

🏆 Best-of-breed cloud flexibility Use the best service for each workload

🔒 Data sovereignty & security Run cloud in your own data center

🧳 Lower risk and vendor lock-in Avoid relying solely on one provider

⚡ Low latency & performance Deploy where your data/users are

💰 Economic efficiency Virtual infrastructure with cloud scaling

🔗 Seamless interconnectivity Private links, unified access, hybrid integration

🎓 Explore More

 ✅ Oracle Cloud Infrastructure (OCI) Foundation Specialist Exam – Free

  ✅ Try OCI Always Free Tier
 ✅ Check out Oracle + Microsoft partnership resources for real multicloud examples

The question asks which two statements are true about Autonomous Database (ADB).

Let's analyze each option:

1. ✅ Auto-scaling is based on workloads.

✔️True – ADB automatically scales CPU and storage up or down based on actual
workload demand.
2. ❌ Running of complex applications, such as Siebel, is not available.
❌ False – Siebel and other Oracle applications (e.g., PeopleSoft, JD Edwards) are
supported on ADB (or are in the process of being certified).
3. ❌ Indexes are added manually only after recommendations.
❌ False – ADB uses automatic indexing, which doesn't require manual intervention
unless chosen. It can implement and test indexes automatically.
4. ✅ It has self-service tools for analytics and data access.
✔️True – ADB includes built-in tools for analytics, data loading, and querying via a self-
service interface.

✅ Correct Answers:

 Auto-scaling is based on workloads.

 It has self-service tools for analytics and data access.
The question asks which two self-service tools are available with Oracle Autonomous
Database (ADB).

Let's examine the options:

1. ❌ Business Objects
✖ Not included as a native tool. Business Objects is a third-party BI tool (from SAP),
not a built-in Oracle self-service tool.
2. ✅ Low-Code App Dev, APEX
✔ True – Oracle APEX (Application Express) is a built-in, self-service low-code
development tool included with ADB.
3. ❌ Encryption only in Object Storage
✖ Misleading – Encryption is available in ADB and not limited to Object Storage. Also,
encryption itself is not a self-service tool.
4. ✅ Oracle Machine Learning
✔ True – Oracle Machine Learning is a built-in self-service tool in ADB that enables
data scientists to create, train, and deploy ML models directly in the database.

✅ Correct Answers:

 Low-Code App Dev, APEX

 Oracle Machine Learning
The question asks which two interconnect solutions are available to connect Oracle Cloud to
other cloud providers.

Let's analyze the options:

1. ❌ Virtual Cloud Network (VCN)

✖ This is Oracle’s internal networking construct within OCI, not a cross-cloud
interconnect solution.
2. ✅ Cloud Router
✔ Correct – "Cloud Router" typically refers to Megaport Cloud Router (MCR), which
is used to enable multicloud interconnectivity, including between OCI and other
clouds like AWS.
3. ✅ FastConnect
✔ Correct – Oracle FastConnect provides dedicated, private network connectivity
from OCI to other clouds, including Microsoft Azure in the Oracle-Azure Interconnect.
4. ❌ Virtual Private Cloud (VPC)
✖ This is an AWS-specific term for isolated network environments and not an
interconnect solution across clouds.

✅ Correct Answers:

 Cloud Router
 FastConnect

The question asks which two statements are true about a converged database.

Let’s evaluate the options:

1. ❌ It requires a dedicated hardware.

✖ False – A converged database does not require dedicated hardware. It can run on
various platforms including cloud infrastructure, on-prem, and engineered systems.
2. ✅ It supports both structured and unstructured data types.
✔ True – Oracle’s converged database supports structured (e.g., relational) and
unstructured data types (e.g., JSON, text, blockchain, etc.).
 3. ✅ It supports graph and spatial workloads.
✔ True – A key strength of a converged database is its native support for multiple data
models and workloads, including graph and spatial.
4. ❌ It does not support analytical workloads.
✖ False – A converged database does support analytical workloads, alongside
transactional, ML, and others.

✅ Correct Answers:

 It supports both structured and unstructured data types.

 It supports graph and spatial workloads.

🔑 Key Takeaways

🕰 Then vs. Now: Application Development Evolution

 Then: Simple architecture — single tool, single database, small team.

 Now: Complex requirements — apps must anticipate needs, support real-time insights,
and handle diverse data (spatial, JSON, sensor, transactional).

🧠 What Are Data-Driven Apps?

 Use multiple data types and data sources (in real time).
 Implement ML, graph analytics, spatial queries, and more.
 Built using modern paradigms like:
o Microservices
o Event-driven architecture
o API-first
o Low-code
o Continuous delivery (24/7, zero downtime)

⚠️The Problem with Single-Purpose Databases

 Quick to start but hard to scale or adapt.

 Require multiple specialized databases (e.g., for documents, graphs, spatial).
 Lead to:
o Data fragmentation
o Complex integration
o Custom code overhead
o Security redundancy
o Vendor/API lock-in
 o Higher operational cost and steeper learning curves

✅ The Oracle Converged Database Approach

A single database engine with native support for:

 JSON
 XML
 Spatial
 Graph
 Key-value
 Relational
 Machine learning
 REST & SQL APIs (uniform access)

🔗 Unified by:

 Pluggable databases (PDBs) for microservice isolation

 Engineered interoperability between data types and models
 Deployment flexibility (containers, on-prem, cloud)

🧩 Synergy with Modern Dev Paradigms

Dev Paradigm Oracle's Simplification

Microservices Use pluggable databases (PDBs) for modular data isolation

Low-code / API-first REST & SQL APIs across all data types

Event-driven Real-time streaming & in-database processing

CI/CD & 24/7 apps Zero-downtime patching, autonomous features

🎯 Final Summary

 Oracle Converged Database = one engine to handle all workloads, simplify operations,
and accelerate innovation.
 Develop faster, reduce complexity, and avoid vendor lock-in.
 Ideal for modern data-driven app development across microservices, analytics, IoT,
and more.
📘 Lesson Objectives

 Understand the graph feature in Oracle Autonomous Database

 Learn to create, query, analyze, and visualize graphs

🔗 What Is a Graph Database?

 Stores entities as vertices (e.g. accounts, customers)

 Stores relationships or interactions as edges (e.g. transactions, connections)
 Ideal for revealing hidden relationships, patterns, and structures

🧱 Graph Data Modeling Example

Tables:

 Accounts → Become vertices

 Transactions → Become edges (e.g. cash transfer from A → B)
 Edge properties: amount, date, type, etc.

Example:

 $1,000 transfer from account 1 to account 672 → becomes a directed edge

🔍 What You Can Do with Graphs

 Query direct and indirect relationships

 Visualize patterns (e.g. fraud indicators)
 Expose hidden connections, like:
o Distribution and accumulation patterns (used in fraud)
o Community detection (clusters of frequent interaction)
o Cycles (money leaving and returning to the same node)

💼 Use Cases of Graph Analytics

1. Financial Services

 Detect money laundering and fraud via cycles, unusual flows

  High in-degree/out-degree accounts are suspicious
 PageRank is used to measure importance beyond basic flow count

2. Manufacturing / Network Management

 What-if impact analysis

 Track downstream impact in case of design change or component failure

3. Customer Insight & Recommendation

 Cluster detection = finding similar users or products

 Handles flexible schemas easily (e.g. new product attributes)

🛠 Oracle Graph Features

 Free with all Oracle DB editions

 Core components:
o Graph data model (vertices + edges + properties)
o Property Graph Query Language (PGQL) – SQL-like, supports pattern
matching
o 60+ built-in algorithms (PageRank, shortest path, community detection, etc.)
o Visualization tools (interactive graph views)

🧮 Graph Algorithms

 Work on full or partial graphs

 Can be iterative, recursive, or computationally expensive
 Examples:
o Pathfinding: shortest path
o Community detection
o PageRank: ranks vertices by both number and importance of connections

💡 Real-World Insight (Fraud Detection Case)

 A customer reduced false positives by:

o Focusing on cycles (6–7 hops)
o Prioritizing accounts with high in/out degree
o Using PageRank instead of just raw transaction counts
✅ Summary

Oracle Graph on Autonomous DB helps:

 Model complex, connected data naturally

 Analyze hidden relationships with built-in algorithms
 Visualize interactions and patterns
 Apply graph analytics to real business challenges: fraud, logistics, recommendations

🌍 Lesson Overview

Goal: Learn how Oracle Autonomous Database (ADB) supports spatial data and how to:

 Develop spatial applications using developer tools

 Use Spatial Studio for no-code spatial analysis
 Leverage spatial data for location intelligence

🧭 Why Spatial Data Matters

 Spatial data captures real-world locations

 Helps answer location-based business questions:
o Are incidents duplicates?
o Where’s the nearest resource?
o What’s the coverage area of a service?
o Are assets in risk zones (e.g. flood zones)?
 Everything happens somewhere — location is a universal attribute of enterprise data

🛠️Oracle’s Spatial Capabilities (3 Components)

1. Core Spatial Functions in Oracle DB

 Hundreds of SQL-based functions and operators to:

o Store, query, and analyze spatial data
o Filter, aggregate, categorize, and enrich data
o Examples: distance calculations, area overlaps, path intersections

2. APIs and Developer Tools

 Use spatial services in your preferred languages:

 o Java, Python, JavaScript, REST APIs
 Build spatial apps with capabilities like:
o Map rendering
o Routing
o Geofencing
o Analytics integration

3. Spatial Studio (No-Code Tool)

 Intuitive, drag-and-drop web interface

 Enables non-technical users to:
o Perform spatial analysis
o Create maps and dashboards
o Conduct geocoding (convert addresses to coordinates)
o Visualize and publish insights

🗺️What Spatial Studio Offers

 Built-in geocoding via Oracle Maps Cloud

 Data indexing, preparation, and publishing
 Workflow creation for spatial analysis
 Integration with Oracle Analytics Cloud

✅ Spatial Studio is included with ADB, but requires compute resources

🔍 Deployable from Oracle Cloud Marketplace or Oracle LiveLabs

🏢 Industry Use Cases

 Retail: Identify optimal store locations, competitor analysis

 Telecom: Drop call analysis, cell tower proximity
 Utilities: Route planning, infrastructure management
 Public Sector & Defense: Emergency response, zoning
 Finance: Branch service area analysis, fraud detection

🧪 Sample Use Case Scenarios

 “Which warehouse is closest to customer X with items in stock?”

 “Do cell tower complaints cluster geographically?”
 “Can we close overlapping bank branches post-acquisition?”
All these are location questions answered with spatial queries or visualized via maps

🔁 Spatial + Multimodel Integration

 Use spatial data alongside:

o JSON
o Graph
o Oracle REST Data Services (ORDS)
 All within a single converged database

✅ Key Takeaways

 Location is a critical attribute of business data

 Oracle ADB includes robust spatial features natively
 Developers and non-developers can benefit via:
o APIs (for custom dev)
o Spatial Studio (for no-code analysis)
 Supports data-driven decision making through location analytics
☁️Oracle Base Database Service Overview

🔍 What It Is:

 A co-managed database service in Oracle Cloud Infrastructure (OCI)

 Combines the power of Oracle Database with the flexibility of the cloud
 You manage the VM and database contents; Oracle manages the underlying
infrastructure

🧰 Key Features and Capabilities

✅ Supported Database Versions:

 Oracle 12c, 19c, 21c, and 23ai

✅ Deployment Options:

 Single-instance VM
 2-node RAC (Real Application Clusters)
(Note: Cannot upgrade from single-node to 2-node later)

✅ Licensing Models:
  License Included (LI): Ideal for new deployments or users without licenses
Includes all Oracle EE options and Enterprise Manager packs
 Bring Your Own License (BYOL):
Allows use of existing on-prem licenses
Grants rights to key features like TDE, diagnostics, and tuning packs
Eligible for Oracle Support Rewards

💸 License Tiers and Features:

Tier Key Features

Standard Edition 2 Up to 3 PDBs, ML, Spatial, Graph

Enterprise Edition (EE) Data Guard, Masking, Diagnostics

EE High Performance Partitioning, Compression, Advanced Security

EE Extreme Performance Active Data Guard, RAC, In-Memory

🔐 All tiers include Transparent Data Encryption (TDE)

⚙️Compute Shapes and Scaling

💻 Shape Types:

Shape Type OCPUs Memory/OCPU Network Throughput RAC Support

Ampere A1 Flex 1–57 8 GB 1 Gbps/OCPU (max 40 Gbps) ❌

Intel Flex 1–32 16 GB 1 Gbps/OCPU (max 32 Gbps) ✅

AMD Flex 1–64 16 GB 1 Gbps/OCPU (max 40 Gbps) ✅

Fixed Shape 1–24 15 GB Varies ✅

🧠 Notes:

 Standard Edition is NOT supported on Ampere

 Scaling OCPUs adjusts memory and network throughput
 Scaling from single VM to 2-node RAC is not supported
🔄 Cloud Automation & Lifecycle Management

Base DB Service includes cloud automation for:

 Provisioning
 Patching
 Scaling
 Backup & recovery (via Object Storage or OCI Recovery Service)
 High availability and disaster recovery (with Data Guard)

Automation reduces DBA workload and deployment errors

🧱 Storage and Backup

 Uses OCI Block Volumes for database storage

 Up to 80 TB data + 20 TB RECO
 Storage scaling improves I/O performance
 Backup options:
o Object Storage
o OCI Recovery Service (recommended)

👤 Access and Control

 You have root access to the DB VM

 Full control over:
o OCPU and memory sizing
o Storage allocation
o Software version
o Backup strategy

🧾 Billing and Pricing

 Billed per second based on:

o OCPU usage
o License tier
 Elastic OCPU metering promotes cost efficiency
 Combined with Support Rewards, BYOL customers can reduce on-prem support bills
✅ Key Benefits Recap

 Full-featured Oracle DB on flexible cloud infrastructure

 Simplified lifecycle with automation
 Strong security model (e.g., TDE, built-in security layers)
 Cost-effective licensing (BYOL or License Included)
 Multiple shape and deployment choices
 Full VM-level control with cloud flexibility

☁️Oracle Base Database Service – Architecture, Availability, and

Security

🗄️Storage Architecture Options

✅ Storage Management Options (Single-node VM only):

 Logical Volume Manager (LVM):

o Supports fast provisioning
o Simplified for general usage
 Oracle ASM (Automatic Storage Management) (Default):
o Recommended for resilience and performance
o Uses 80% DATA and 20% RECO disk groups
o Required for 2-node RAC deployments
o Provides triple mirroring via OCI block storage
🛠 Additional Notes:

 Storage Scaling: Online scaling supported after provisioning

 Storage Allocation: Defined at creation; scalable post-deployment
 Cloning Support: Available for both LVM and ASM-based systems

🌍 OCI Concepts for Deployment

Concept Description

Region Geographical area; independent infrastructure

Independent data centers within a region; connected via low-latency

Availability Domain (AD)
networks

Fault Domain (FD) Hardware isolation within an AD; 3 per AD to minimize single-point failure

🛡️High Availability (HA) & Disaster Recovery (DR)

✅ Built-in Oracle Best Practices:

 RAC and Data Guard automatically spread across fault domains for redundancy
 DR and HA achieved using standby Data Guard instances across ADs or regions
 Rolling patching and online scaling supported

🧩 Oracle Maximum Availability Architecture (MAA) Components:

Feature License Tier Required

Flashback All tiers

Backup & Recovery All tiers (incl. Recovery Service & Object Storage)

Multitenant (3 PDBs) All tiers

Multitenant (up to 4,098 PDBs) High Performance or Extreme Performance

Data Guard All Enterprise Editions

Active Data Guard Extreme Performance only

Oracle RAC (2-node) Extreme Performance only

Application Continuity Extreme Performance only (requires ADG or RAC)

🔐 Security – Oracle’s “Defense in Depth” Strategy

🛡️Data-Level Security Features:

 TDE (Transparent Data Encryption): Encrypts data at rest

 Data Redaction, Masking, and Subsetting
 Oracle Key Vault: Separates encryption key control from infrastructure control
 Database Vault: Restricts DBA access to user data
 Database Firewall: Controls allowed SQL statements
 Oracle Data Safe: Identifies sensitive data & performs security assessments

🧱 Infrastructure & Network Security:

 Hardened Oracle Linux 7 OS image

 Only necessary packages and services included
 OCI VCN/VLAN isolation for network traffic
 Oracle Native Network Encryption for client-database traffic

🧪 Audit and Compliance:

 Industry certifications: PCI, HIPAA, ISO 27001

 Built-in features include:
o Antivirus scanning
o Alerting for unexpected changes
o Vulnerability scanning of Oracle-managed systems

✅ Key Takeaways

Area Highlight

Storage LVM (simple) or ASM (resilient, RAC-ready) options; block-based & scalable

HA & DR Built-in support for RAC, Data Guard, Flashback, automated backups

Security Multi-layered "Defense in Depth" from OS to SQL-level access control

Governance Customer has root access; Oracle enforces compliance & control boundaries

☁️Oracle Exadata Database Service – Summary Overview

🚀 What Is It?
A co-managed Oracle Database cloud service running on Exadata Cloud Infrastructure,
offering the full power of Oracle Exadata with cloud automation, flexibility, and pay-per-
use pricing.

🧩 Key Components and Capabilities

Area Highlights

Runs on Oracle Exadata, the highest-performing, most available platform for

Platform
Oracle DB

Provisioning Web-based UI & APIs with service isolation, no hardware overprovisioning

- License Included (LI)

Licensing Models
- Bring Your Own License (BYOL)

Database Features RAC, Multitenant, Active Data Guard, In-Memory, Partitioning, Flashback,
Included more

Automation Cloud automation for provisioning, patching, scaling, backup, lifecycle

Scaling Elastic OCPU scaling – pay only for what you use

Root Access Customers have full control of VM and can install 3rd-party agents

Zero Downtime Migration 100% compatible with on-premise Oracle DB deployments

🛠️Cloud Deployment Options

1. Exadata Database Service (Public Cloud)

 Fully cloud-managed in Oracle public cloud

 Oracle manages hardware, network, infrastructure
 You manage VMs and database instances
 Supports autonomous and co-managed databases
 Ideal for full cloud transition with elastic pricing

2. Exadata Cloud@Customer (Hybrid)

 Same Exadata platform, but deployed in your own data center

 Satisfies data residency, security, and latency requirements
 4-year minimum subscription for infrastructure
 Uses OCI control plane to manage as-a-service model
 Maintains compatibility with on-prem apps
🌐 Multicloud and Interconnect Options

 Supported in multicloud environments (e.g., with Azure)

 FastConnect & Oracle Database@Azure enable native connections
 Offers zero-downtime migration with consistent Exadata technology across clouds

💡 Comparison: Exadata Deployment Models

Deployment Model Location Who Manages Infra Elastic Scaling Data Residency

On-Prem Exadata Customer DC Customer Manual Customer-managed

Exadata Cloud Service Oracle Cloud Oracle Yes (OCPUs) Oracle Cloud

Exadata Cloud@Customer Customer DC Oracle Yes (OCPUs) Meets local laws

🔐 Security and Compliance

 Complete service isolation

 Exadata managed infrastructure includes RoCE network, servers, firmware
 Exadata Cloud@Customer retains on-prem compliance standards
 Supports all Oracle DB security features (TDE, redaction, masking)

💰 Cost Efficiency

 Pay-per-use with short 48-hour commitment (Exadata Cloud)

 BYOL + Oracle Support Rewards reduce total ownership cost
 No capex for infrastructure when using public cloud

✅ Why Choose Exadata Database Service?

Benefit Description

🔁 Full Compatibility Run any existing Oracle DB app without modification

💡 Cloud Automation Self-service lifecycle operations via GUI or API

 Benefit Description

📈 Performance In-memory, Smart Scan, storage indexes, Flash cache

🛡 Security & Isolation Full VM isolation and best-in-class data protection

🌍 Hybrid & Multicloud Deploy anywhere: Oracle Cloud, Azure, or on-prem

🔄 Easy Migration Zero downtime migration and full portability

☁️Exadata Database Service – Cloud Management Model

🔧 Responsibility Breakdown

Layer Responsibility Managed By

Infrastructure (servers, network, Ownership, patching, security scanning,

Oracle
hypervisor) updates

Guest VM (OS, DB software, grid infra, DB lifecycle, access control, patching,

Customer
data) provisioning

 Oracle manages everything up to the hypervisor

 Customer manages everything inside the VM
 Oracle cannot access guest VMs
 Customers can install third-party software inside the guest VMs
 Supported DB versions: up to 23ai

⚙️Cloud Automation Capabilities

Available via UI or REST APIs:

 Provisioning & Deleting VMs/DBs

 Patching
 Backup & Recovery
 Scaling resources (OCPUs, storage)
 High Availability (HA) and Disaster Recovery (DR) with Data Guard

🔄 Automation frees up DBA teams to focus on innovation and reduces risk of config errors.

🧱 Exadata Cloud Infrastructure Architecture

Key Components:
 Component Description

RoCE Network Fabric Ultra-fast internal networking (up to 100 Gbps)

Scale-out Architecture Independent scaling of DB and storage nodes

DB Virtual Machines (VMs) Run Grid Infrastructure, DB homes, automation tools

50 Gbps Client & Backup Network Supports app connectivity, backup, data loading, DR

 DB VMs communicate securely over RoCE

 External connections made via VCN subnets
 Oracle Cloud Ops manage infrastructure via isolated cloud management network

🏢 Exadata Cloud@Customer – Hybrid Architecture

📍 What It Is

 Full Exadata system deployed in customer’s data center

 Powered by Exadata Database Service, but data stays on-prem
 Ideal for customers with data residency, latency, or cloud readiness concerns

🔗 Architecture Flow

Component Role

Exadata system + RACs Same Exadata hardware as in cloud

Local control plane servers Tunnel telemetry and API calls securely to Oracle Cloud

Oracle OCI control plane Receives/manage REST API calls for lifecycle ops

End-users/apps Connect to DBs locally through internal network

Oracle Cloud Ops Manage physical hardware if needed via access-controlled tunnel

✅ Same user experience as Exadata in Oracle Cloud

✅ Secure tunneling to OCI control plane
✅ Local app compatibility without public internet exposure

🔐 Access Control and Compliance

 Oracle Cloud Ops has audited and controlled access to infrastructure

  Customer controls:
o VM content
o Encryption keys
o Access permissions
 Compliant with regulatory standards and data sovereignty requirements

🔁 Why Exadata Cloud@Customer?

Reason Benefit

🛡️Data Residency Keeps data within your data center

⌛ Latency-sensitive apps Avoids network delays of public cloud

💼 Legacy integration Continue using tightly coupled on-prem systems

💻 Public Cloud UI & APIs Same experience, tools, automation, and economics

✅ Key Takeaways
Exadata Database Service Exadata Cloud@Customer

Deployed in Oracle Cloud Deployed in customer’s data center

Oracle manages infra, customer manages VMs Same model

Cloud-scale economics, flexibility Data residency and legacy system compatibility

Automation via OCI Console & APIs Same UI/API experience

☁️Oracle Exadata Database Service – Final Key Concepts

🧾 Licensing Models

1. License Included (LI)

 Includes:
o Oracle Database Enterprise Edition
o All DB options and Enterprise Manager packs
 Ideal for:
o Customers without existing licenses
o Customers who want access to advanced DB features
 Simple subscription model with no upfront capex
2. Bring Your Own License (BYOL)

 Designed for:
o Customers with existing Oracle DB EE and options licenses
 Includes:
o Access to TDE, Data Safe, Diagnostics Pack, Tuning Pack, Data Masking &
Subsetting, Real Application Testing
o Exadata Storage Server software (no separate license needed)
 Standard Edition is not supported

📈 Elastic Scalability and Cost Efficiency

 Licensing cost is based on OCPUs in use

 OCPUs can be scaled online, with no database downtime
 Matches workload demand in real-time
 Contrasts with on-prem:
o On-prem systems must be sized for peak load
o Results in underutilization and wasted license cost

💡 Cloud = pay for what you use, not what you provision

🛡️Defense in Depth: Oracle's Security Strategy

Oracle applies layered security across the stack—from infrastructure to database.

🔐 Data-Level Security Features:

 TDE (Transparent Data Encryption) – Encrypts data at rest

 Data redaction, masking, subsetting – Granular access control
 Oracle Key Vault – Separates encryption key management
 Database Vault – Restricts even DBA-level access to sensitive data
 Database Firewall – Controls allowed SQL commands
 Data Safe – Risk assessment & sensitive data discovery

🧱 Infrastructure and Network Security

 OCI Virtual Cloud Networks (VCNs) or VLANs for network isolation

 Native Network Encryption for client-database traffic
 Dedicated client & backup networks for performance and separation
 Token-based SSH access, hardened OS with minimum packages

✅ Compliance Certifications
  Meets FedRAMP, HIPAA, PCI, and ISO standards
 Trusted by heavily regulated industries (finance, healthcare, etc.)

👀 Operator Access Control (OAC)

A critical feature for customers needing strict infrastructure access control.

🔐 Key Capabilities:

 Oracle staff must request time-limited access

 Customers:
o Specify components, duration, privileges
o Monitor commands and keystrokes live
o Terminate sessions or processes in real time
 Enables full visibility and control over Oracle support operations

🚨 Particularly valued in regulated industries (finance, healthcare, government)

✅ Final Takeaways
Area Key Benefit/Capability

Licensing Flexibility BYOL or LI; pay-per-use OCPUs

Elastic Compute Scaling Scale OCPUs online without downtime

Enterprise Features Full access to RAC, ADG, Partitioning, etc.

Zero Downtime Migration 100% Oracle DB compatibility

Advanced Security Multi-layered Defense in Depth with full compliance

Access Control Operator Access Control for regulated environments

☁️Exadata Database Service – Lifecycle Management Summary

🔧 1. Custom Software Images

 Create custom DB and Grid Infrastructure images to standardize deployments

 Can include:
o Oracle version
o Release updates
 o Proactive bundle patches / one-off patches
o Oracle Home inventory content
 Created and stored in Oracle-managed object storage
 Used for:
o Provisioning
o Patching
o Gold image standardization
 Managed via Console, API, CLI, or SDKs

🏠 2. Database Home Creation

 Navigate to VM Cluster → Database Homes

 Choose:
o Oracle-provided image or
o Custom software image
 Define:
o Display name
o Compartment
o Software version/image
 Used as a base for creating one or more databases

🗄️3. Database Creation

 Navigate to VM Cluster → Create Database

 Configure:
o DB name and version
o Optional PDB name
o Admin credentials (CIS user)
o Backup destination & schedule
o Encryption type (Oracle-managed or Customer-managed via OCI Vault or OKV)
 Fully integrated with Key Vault and data protection features

🧩 4. Pluggable Database (PDB) Management

✅ Key Actions:

 Create PDB inside existing CDB

 Clone PDB with three options:
o Local clone (same CDB)
 o Remote clone (different CDB, same AD)
o Refreshable clone (remote, periodically synced)

🔄 Other Lifecycle Tasks:

 Start/Stop PDB
 Relocate PDB to another DB system
 Restore PDB from backup or point in time
 Delete PDB

🔗 Connectivity:

 Easy Connect & Long Connect Strings

 Performance Hub monitoring

🔄 5. Data Guard Integration (Disaster Recovery)

✅ Key Capabilities:

 Simple setup via UI or API

 Choose:
o Standby location (region or AD)
o DB Home (existing or new)
o Admin password (must match primary)
o Data Guard type (including Active Data Guard)
 Supports:
o Switchover – planned role reversal
o Failover – unplanned role reversal
o Fast Start Failover – automatic failover
o Reinstate – bring failed DB back as standby

🧱 Requirements:

 Same compartment
 Same VCN (if in same region)
 VCN Peering (if across regions)
 Same Oracle DB version
 Unique DB_UNIQUE_NAME
 Firewall rules for TCP port 1521 (SCAN listener)
 Stateful egress traffic rules

🔐 6. Key Security Integrations

  Oracle Key Vault (OKV) for customer-managed keys
 Full support for:
o TDE
o Data masking
o Redaction
o Database Firewall
o Vault-based key encryption

💡 Use Cases & Benefits

Feature Benefit

Custom Software Images Standardize and automate DB deployments

Database Home Reuse Reduce redundancy, simplify upgrades

PDB Cloning Fast dev/test provisioning, DR readiness

Active Data Guard High availability, read-only workload offloading

Data Guard Automation Resilient, low-downtime disaster recovery

OCI Vault Integration Centralized key control for compliance

🧾 Recap – Lifecycle Operations with Exadata Database Service

Operation Tool Options

Create custom software image Console / API

Provision Database Home Console / API

Create & manage databases Console / API

Manage PDBs (clone, relocate, restore, delete) Console / API

Enable & manage Data Guard Console / API

Lifecycle monitoring & performance Performance Hub / REST

☁️Exadata Database Service – Maintenance Responsibility Summary

🛠️Maintenance Responsibility Matrix

 Component Managed By Oracle (OCI) Managed By Customer

Physical compute/storage servers ✅ ❌

RDMA network fabric & switches ✅ ❌

ILOMs, PDUs, Firmware ✅ ❌

Guest VM OS ❌ ✅

Oracle Grid Infrastructure ❌ ✅

Oracle Database Home ❌ ✅

Database schema/data/key management ❌ ✅

 Oracle handles Infrastructure Maintenance (quarterly)

 Customer is responsible for User-Managed Maintenance:
o Grid Infrastructure patches
o Database Home patches
o Guest VM OS updates

📋 User-Managed Maintenance: Key Tasks

✅ Patching Components:

 Guest VM OS (Exadata Image)

 Grid Infrastructure (GI)
 Database Home

Oracle recommends patching at least once every 180 days.

✅ Best Practices Before Patching

1. 🔒 Backup your databases before any update.

⬆️Update GI first, then create/update DB Homes.

2. 🆕 Prefer creating a new DB Home over updating the existing one.
3.
4. 📊 Run Precheck operations well ahead of the maintenance window.
5. 💾 Ensure at least:
o 15 GB free on /u01 (Grid Infrastructure Home)
o 15 GB free on /u02 (Database Home)
6. 🔐 Use OCI IAM policies for permission to apply patches via UI, CLI, or API.
🔍 Checking Current Patch Levels

 Navigate to VM Cluster Details

 View:
o Grid Infrastructure version
o Exadata image (OS) version
o Database Home versions

⬇️Finding Available Updates

Use OCI Console → Updates Page, or REST API.

Organized by:

 Exadata Image (Guest VM OS)

 Grid Infrastructure
 Database Homes

Each update lists:

 Patch Description
 Version
 Release Date
 Last Precheck Timestamp

🔁 Patch History Tracking

 Use Update History View to:

o Track applied and prechecked patches
o View success/failure status
o Retry failed patches (creates new entry)

⚠️Note: Patches via CLI (dbaascli) are not shown in Console history.

💡 How to Apply Updates

🖥️1. Guest VM OS (Exadata Image)

  Use OCI Console or API
 Precheck → Validate readiness
 Apply → Includes precheck + patching
 Rollback available if update fails
 Only latest 4 minor versions are shown

🧱 2. Grid Infrastructure

 Use OCI Console → Updates Page

 Rolling patching (one node at a time)
 Instance on updating node is temporarily unavailable
 Same precheck & apply workflow as OS image

🧩 3. Database Home

 Use OCI Console to:

o View patchable DB Homes
o Select patch version
 OCI supports:
o Oracle-provided and
o Custom software images created by your organization

✅ Summary Table – Patching Workflow

Component Tool Actions Notes

Guest VM OS Console / API Precheck, Apply Includes rollback

Grid Infrastructure Console / API Precheck, Apply (Rolling) Instance downtime during patch

Database Home Console / API Use custom or Oracle image Recommended to create new home

🛡️Security & Access Requirements

 OCI IAM policies needed for:

o Viewing updates
o Running prechecks
o Applying patches
 Error like “Unauthorized” means missing IAM role or policy
🚦 Key Considerations

 Always run precheck in advance

 Use OCI automation over manual patching tools
 Avoid modifying root OS manually
 Audit patch success/failure through Update History

🧱 1. Database Home Update Overview

✅ Two Methods for Patching:

Option Method Impact

1 Update Existing DB Home All databases in the home are patched.

Create New DB Home & Only selected DB is affected — preferred approach for minimal
2
Move DB downtime.

📌 Update Flow Using OCI Console:

1. Go to: OCI Console → Updates Page

2. Select Oracle or custom software image
3. Run Precheck: validate system readiness
4. Apply Patch: updates DB Home in rolling manner across RAC instances

Oracle recommends prechecking in advance of maintenance windows.

🧭 2. Move Database to Another DB Home (Preferred Patch Path)

🪜 Steps:

1. Navigate to: Database Details

2. Click: Move to Another Database Home
3. Select: Target DB Home (with desired patch level)
4. Click: Move Database

✅ Faster, less disruptive, and suitable for individual database patching.

⚙️3. Grid Infrastructure (GI) Upgrade

🔄 GI Upgrade Process (Rolling):

Phase Detail

🔍 Precheck Detect upgrade blockers

📶 Rolling Patch Node-by-node upgrade

🔒 Restrictions No Data Guard ops, provisioning, scaling, or backups during GI upgrade

⚠️Cannot upgrade GI if infrastructure maintenance is scheduled within 24 hours.

🆙 4. Oracle Database Upgrade

✅ Before Upgrade Checklist:

 🔐 Backup the database

 🧪 Test new version on dev/test

⬆️Create target DB Home with new version

 ✅ Run Precheck for upgrade

 🛑 Disable auto-backups temporarily
 📂 Ensure all PDBs can open (to avoid failure)

🔁 You cannot restore from backups made before upgrade using automatic backups.

📌 Upgrade Workflow via OCI Console:

1. Navigate to: Database Details

2. Click: More Actions → Upgrade
3. Input:
o 🎯 Target DB version
o 🏠 Target DB Home
4. Run Precheck
5. If successful, click: Upgrade Database

Upgrade Safeguards:

 Precheck runs automatically

 Guaranteed restore point set (for flashback)
  Uses DBUA (Database Upgrade Assistant) internally

🛡️5. Data Guard Consideration in Upgrade

Action Recommendation

Upgrade standby or primary ✅ Standby first (safer)

Redo Apply 🔕 Disabled during upgrade

Post-upgrade ✅ Confirm redo apply and open mode

✅ Key Takeaways from This Section

You’ve now learned how to:

 ✅ Create custom software images

 ✅ Create Database Homes and Databases
 ✅ Manage Pluggable Databases (PDBs)
 ✅ Enable Data Guard
 ✅ Apply User-managed maintenance
 ✅ Upgrade Grid Infrastructure and Databases
✅ Key Benefits of Oracle Autonomous Database (ADB):

 Massive reduction in operational complexity and cost:

o Automates provisioning, patching, tuning, securing, scaling, and recovery.
o Cuts administration costs by up to 80% and operational costs by up to 90%.
 Accelerates application development:
o Instant provisioning avoids delays.
o Enables faster innovation and saves tens of thousands of dollars per
application.
 Cloud-native by design:
o ADB is designed for autonomous management in the cloud, unlike traditional
lift-and-shift cloud models.

🧱 Core Architectural Components:

  Exadata Infrastructure Cloud:
o Optimized for Oracle workloads.
o Supports high availability and scale-out deployments.
 Triple-mirrored storage:
o Ensures data redundancy and fault tolerance.
 Autonomous Data Guard:
o Automatically provisions a standby system in the same or another region for
disaster recovery.

🔐 Security Features:

 End-to-end encryption:
o Data is encrypted at rest and in transit by default.
 Automated patching:
o Applies security patches immediately to eliminate vulnerabilities.

⚙️Automation Features:

 Self-managing:
o Handles OS, VM, storage, and DB patching.
o Uses machine learning to detect and diagnose problems.
 Self-securing:
o Continuously applies security measures with no manual intervention.
 Self-repairing:
o Detects system or user errors and recovers automatically, ensuring zero data
loss.
 Self-tuning:
o Auto-configures for OLTP or analytic workloads.
o Uses adaptive indexing, statistics, and plan optimization to ensure best
performance.

🌐 Deployment Flexibility:

 Available in public cloud and on-premises (e.g., Oracle Cloud@Customer).

 Provides a smooth transition for enterprise workloads to the cloud.

🔁 High Availability and Reliability:

  Clustered architecture: Supports failover in case of node or regional failure.
 Always-on availability: Designed for mission-critical workloads with zero downtime.

📌 Summary Statement:

Oracle Autonomous Database revolutionizes data management by automating every layer of

database operations, delivering unmatched reliability, security, and performance with minimal
human intervention—ideal for modern, cloud-native workloads.

🔑 Oracle Autonomous Database Licensing – Key Concepts

🧮 ECPU-Based Pricing Model

 ECPU = Elastic CPU: A performance-based metric for ADB compute usage.

 Not tied to physical cores, threads, or processor models — ensures durable pricing
across hardware generations.
 Granular scaling: Minimum configuration is 2 ECPUs, and scaling occurs in single-
step increments for cost efficiency.
 ECPU-based pricing applies to:
o Autonomous Data Warehouse (ADW)
o Autonomous Transaction Processing (ATP)

💳 Licensing Options

1. Universal Credit Model (UCM)

 License-Included: Standard pricing that includes all database options.

 Pay-As-You-Go (PAYG) or Annual Universal Credits:
o PAYG: Pay only for usage, no upfront commitment.
o Annual UCs: Prepaid commitment (like a gift card), allows use across OCI
services.
o May qualify for discounts based on commitment level.

2. Bring Your Own License (BYOL)

 For customers who already own Oracle Database licenses.

 Lower cost than UCM.
 Restrictions apply:
o Up to 64 ECPUs: Requires Enterprise Edition + RAC license.
o Up to 32 ECPUs: If using Standard Edition, with or without autoscaling.
📈 SLA (Service Level Agreement) Consideration

 99.995% SLA available only with:

o Oracle Enterprise Edition
o Real Application Clusters (RAC)
o Active Data Guard

🔁 Switching License Types

 You can dynamically toggle between:

o BYOL ↔ License-Included
 No downtime is incurred.
 Accessible via:
o OCI Console > More Actions > Update License and Edition

📉 Cost Optimization with ECPUs

 Simplified pricing model not tied to hardware specs.

 Autoscaling supported.
 Reduced storage cost with ADW ECPU pricing.

✅ Summary Table: BYOL vs License-Included

Feature BYOL License-Included (UCM)

Oracle license required Yes No

Cost Lower (if licenses owned) Higher, includes all licenses

Max ECPUs (Standard Edition) 32 N/A

Max ECPUs (Enterprise + RAC) 64+ Unlimited

SLA 99.995% Eligibility With EE + RAC + Active Data Guard Included by default

Toggle Supported ✅ Yes ✅ Yes

🔄 Auto Scaling with Oracle Autonomous Database

🧠 Key Capabilities

 Autonomous Database supports independent scaling of:

o 🧮 ECPUs (Compute)
o 💾 Storage
 Compute auto scaling:
o Automatically increases CPU up to 3x the base provisioned ECPUs
o Example: With 2 ECPUs, it can auto-scale up to 6; if manually increased to 6,
auto-scale can reach 18

⚙️Auto Scaling Behavior

Feature Details

Auto scaling availability Enabled by default (not available for Always Free)

Manual scaling No downtime required

Change ECPU / storage Can be done independently and on the fly

Auto scaling toggle Can enable/disable anytime via Console or API

Scaling visibility Lifecycle shows “scaling in progress” while active

🖥️Demonstration Walkthrough

1. Access ATP Database (e.g., ATP demo) from OCI Console.

2. Current settings show:
o ECPU Count: 2
o Storage: 1 TB
o Compute Auto Scaling: Enabled
3. Click Manage Resource Allocation.
4. Manually adjust ECPU (e.g., from 2 to 6) or storage.
5. Click Apply.
6. Status shows “scaling in progress” (database remains available).
7. Once done, Lifecycle state changes to “available”, and the new ECPU count is shown.

✅ Summary
  You can scale compute and storage independently with zero downtime.
 Auto scaling allows the database to dynamically adjust compute up to 3× provisioned
CPUs.
 Configuration changes are quick, easy, and non-disruptive.
 Best suited for dynamic workloads needing elastic compute.

☁️Provisioning an Oracle Autonomous Database – Key Summary

🧩 Overview

 Oracle automates provisioning of a highly available, secure, and scalable database.

 Built on:
o Oracle Exadata Infrastructure
o Oracle RAC cluster (scalable)
o Fully encrypted
o 19c pluggable database (PDB) within a container database (CDB)
o Pre-configured backups

⚙️Provisioning Steps (via OCI Console)

1. Navigation

 From OCI Console home, select:

o “Autonomous Data Warehouse” or “Autonomous Transaction Processing”
o OR navigate: ☰ Hamburger Menu → Oracle Database → Select desired ADB
type

2. Create Autonomous Database

 Click "Create Autonomous Database"

 Select/confirm:
o Compartment: logical grouping for your resources
o Display Name: user-friendly identifier
o Database Name: unique system identifier (can match display name)

3. Workload Type

 Choose one of:

o Transaction Processing
o Data Warehouse
o JSON
o APEX
4. Deployment Type

 Choose Serverless
 Select:
o Database version (e.g., 19c)
o ECPUs (Elastic CPU)
o Storage (in TB)
o Auto scaling: Enabled by default (except Always Free)

5. Security Settings

 Define Admin password

 Set network access:
o Public: Accessible over public IP
o Private: VCN-subnet-based internal access
 Define Access control rules: IP or CIDR-based restrictions (can be edited later)

6. License Type

 Choose between:
o BYOL (Bring Your Own License)
o License Included (pay-as-you-go)

✅ Final Steps

 Click "Create Autonomous Database"

 The database will appear in a “Provisioning” state, and then transition to “Available”
 You’re now ready to connect, load data, and begin using it

🔐 Behind-the-Scenes Features

 Runs on secure, software-defined private network

 Data and connections fully encrypted
 No manual setup for:
o Tablespaces
o OS or file systems
o Initialization parameters
 Backups are automatic (retention configurable)
📝 Summary

Provisioning an Autonomous Database:

 Fast – done in minutes

 Simple – minimal input required
 Secure & Scalable – out of the box
 Cloud-native – requires no manual infrastructure configuration

🛑🔄 Starting and Stopping an Oracle Autonomous Database – Summary

🎯 Purpose

 Start and stop your Autonomous Database on demand to:

o Conserve CPU resources
o Pause billing (CPU only – storage still billed)

🚦 Key Concepts

Action Result

Start DB Instance becomes Available – billing resumes

Stop DB Instance becomes Stopped – CPU billing pauses, data remains persistent

🖥️Steps via OCI Console

✅ To Start an ADB Instance:

1. Navigate to the Autonomous Database Console

2. Ensure instance is in "Stopped" state
3. Click More Actions → Start
4. Confirm start in the pop-up
5. Status transitions: Stopped → Starting → Available (Green tile)

🔴 To Stop an ADB Instance:

1. While running, click More Actions → Stop

2. Confirm stop in the pop-up
3. Status transitions: Available → Stopping → Stopped (Orange tile)
⏰ Automated Start/Stop Scheduling

 Found under More Actions → AutoStart/Stop Schedule

 Select:
o Start time (optional)
o Stop time (optional)
 Times are based on UTC
 Applied per day of the week
 Useful for:
o Business hours
o Non-peak downtime
o Cost optimization

⚙️Automation / Scripting Options

 Operations (start/stop) also available via:

o OCI REST APIs
o CLI/SDKs
o Integration with automation workflows

💡 Important Notes

 Stopping = Paused CPU billing, but:

o Storage continues to incur costs
 No data loss when stopping
 Ideal for dev/test environments or scheduled usage patterns

✅ Wrap-up

 Simple GUI process: point-and-click

 No downtime required for scheduling changes
 Adds flexibility and cost-efficiency to your database operations

🧰 Oracle Autonomous Database Tools – Summary

🚀 Key Characteristics of Autonomous Database

 Feature Description

Self-Driving Automates provisioning, management, monitoring, and tuning

Self-Securing Auto-patches, encrypts, and protects from internal/external threats

Self-Repairing Auto-recovers with < 2.5 minutes of downtime per month

🔧 Core Tool Interface: Database Actions

 Centralized GUI for development and administration

 Modules include:
o SQL Worksheet
o Data Modeler
o APEX
o Machine Learning
o REST & JSON
o Data Studio
o Monitoring & Administration Tabs

💻 Built-in Developer Tools

Tool Use Case

SQL Developer
Run SQL, load data, manage objects
Web

APEX Low-code app development platform

REST Services Create APIs over SQL/PLSQL objects

JSON Toolset Work with JSON natively in the database

Command-line tool with modern features like autocomplete, history, and format
SQLcl
support (CSV, HTML, XML)

Liquibase Support Schema lifecycle versioning & deployment

📊 Data Studio

 Visual interface for:

o Data loading
 o Catalog management
o Data insights & entity discovery
o Data sharing configuration
 Supports searchable metadata and data lineage

🤖 Analytics & Insights Tools

Tool Purpose

AutoML/ML Notebooks Build and run machine learning models directly on DB

Graph Studio Build and visualize graph data (nodes/edges)

Spatial Studio Visualize and analyze geospatial (location-based) data

🧩 Additional Capabilities

 Command-line integration: via SQLcl & OCI

 APEX Workspace Access: manage or build low-code apps directly
 RESTful Management APIs: Admin and manage database components

✅ Conclusion

Oracle Autonomous Database tools offer:

 A comprehensive toolset for developers, analysts, and DBAs

 Seamless integration with REST, JSON, spatial, and machine learning features
 A cloud-native experience with centralized access via Database Actions
🏛️Oracle Data Lakehouse on OCI – Key Concepts

🔹 What is a Data Lakehouse?

 A Lakehouse combines the scalability and flexibility of data lakes with the reliability
and performance of data warehouses.
  Oracle Lakehouse enables use of machine learning (ML), open source services, and
the Autonomous Database in a unified architecture.

🔑 Core Components of Oracle Lakehouse

1. Data Warehouse – stores structured data for analytics (e.g., Autonomous Data
Warehouse).
2. Data Lake – stores raw/unstructured data (e.g., object storage).
3. Managed Open Source Services – includes tools like Spark, Hadoop, Redis.
4. Data Integration – supports tools like Oracle GoldenGate, Oracle Data Integration, and
third-party ETL.
5. Data Catalog – metadata management system that enables discovery, tagging,
governance, and consistent terminology.

🧠 AI/ML & Analytics Integration

 Built-in Oracle AI Services and OCI Data Science tools for advanced ML.
 Oracle Analytics Cloud and tools like Tableau or Looker supported.
 MySQL HeatWave boosts performance for analytics on MySQL with no app changes.

⚙️Key Features

 Write once, read anywhere: Store data once and use across different engines.
 Eliminates silos: Integrates data warehouses and lakes seamlessly.
 Serverless Spark via OCI Data Flow: For real-time big data analysis.
 Supports external lakes (AWS, Azure) using Oracle SQL.

📚 Metadata and Governance

 OCI Data Catalog is central for:

o Harvesting metadata from Object Store and databases.
o Managing business glossary and searchable datasets.

🛠️Developer and Integration Tools

  Open tool support: Python, Notebooks, REST, SQL, etc.
 Data stored in Object Storage (lake) or Autonomous DB (warehouse).
 Data Studio, Graph Studio, Spatial Studio for visualization and advanced analytics.

💡 Business Benefits

 Accelerates innovation by:

o Reducing time-to-insight.
o Using all data sources—structured and unstructured.
o Providing AI/ML for anomaly detection and prediction.
 Supports hybrid and multi-cloud use cases with full governance.

🧠 Oracle Machine Learning (OML) with Autonomous Database – Key

Concepts

🔍 What is OML?

 Oracle Machine Learning (OML) enables data scientists and developers to:
o Build, train, and deploy ML models directly inside the Autonomous
Database.
o Avoid data movement by working in-database with optimized algorithms.
o Use SQL, Python, R, and no-code UIs.
o Leverage AutoML to automate algorithm selection, tuning, and model training.

🔧 Tools and Interfaces

 OML for Python (OML4Py) – Write Python code directly in the database.
 OML AutoML UI – No-code interface for model building and deployment.
 OML Notebooks – Create, run, and schedule notebooks.
 Visualization tools – Built-in support for R, Python, and third-party libraries.
 SQL and REST APIs – Easily integrate with applications and Oracle stack.

📊 ML Process with OML

1. Define the problem

2. Prepare data (cleaning, transforming)
3. Build the model
4. Evaluate and tune
5. Deploy the model
 6. Monitor (with built-in model/data monitoring UIs)

💡 Key Features of OML

 In-database execution (no need for external engines)

 AutoML (for Python and UI): simplifies model training
 Built-in algorithms:
o Classification
o Regression
o Clustering
o Association rules
o Anomaly detection
o Time series
 Automatic data prep, text mining, and partitioning
 Scalable and secure architecture, spanning:
o On-premises
o Cloud
o Hybrid

💰 Licensing and Cost

 OML is included with Autonomous Database (no extra cost).

 For on-premises: SQL-based ML is included; R/Python require extra installation.

🧩 Real-World Use Cases

🔹 Industry Examples:

 Finance: Fraud detection, credit risk scoring, customer retention

 Healthcare: Diagnosis support, disease prediction
 Energy: Forecasting, predictive maintenance
 Transportation: Route optimization, failure prediction
 Retail: Personalized marketing, recommendation engines
 Government: Public safety analytics, fraud detection
 E-commerce (e.g. El Tronics): Targeted promotions, demand prediction

🔹 Technique Examples:

 Classification: Churn prediction, loan default

  Regression: Revenue forecasting, price prediction
 Clustering: Customer segmentation
 Association: Cross-selling analysis
 Anomaly Detection: Fraud, system failure alerts
 Time Series: Sales and demand forecasting

✅ Benefits

 Reduced cost and complexity.

 Increased accessibility to non-data scientists.
 No-code and low-code development support.
 Broad support for modern tools and languages.

📚 Oracle Data Mesh Architecture on OCI – Study Guide

🔍 What is Data Mesh?

 Data Mesh is a decentralized, domain-driven approach to data architecture.

 It promotes treating data as a product, owned and managed by domain teams, rather
than centralized IT.
 It enables better data agility, scalability, and innovation, especially in complex
enterprise environments.

🧠 Key Concepts & Principles

✅ Core Mindset Shifts:

 From centralized data lakes → decentralized data products

 From ETL pipelines → Continuous Transformation and Loading (CTO)
 From IT ownership → Domain ownership of data
 From passive storage → Active data in motion (event-driven)

✅ Four Core Principles of Data Mesh:

1. Data Liquidity – Data can move and be used freely and efficiently.
2. Data Productivity – Data must be easily consumed to produce business outcomes.
3. Data Security – Secure by design, covering production and consumption.
4. Data Governance – Clear policies, access control, and lineage.
🧩 Components of Data Mesh

Component Role

Data Products Purpose-driven, discoverable, reusable data assets

Event-driven Architecture Enables data-in-motion using Kafka, Pulsar, or similar streaming tools

Streaming Services Real-time processing via Oracle Stream Analytics, Kafka, GoldenGate

Decentralized IT Systems Aligns with SaaS, multi-cloud, and hybrid models

Data Ledgers Kafka, Pulsar, CDC, or blockchain for event tracking

Time-based Processing Process data by event time, useful for time series & alerting

🔧 Oracle Tools Supporting Data Mesh

Tool Function

Oracle IoT Cloud IoT event generation and processing

Oracle GoldenGate Change data capture (CDC), sync across stores

Oracle Integration Cloud Event-driven workflow, alerting

Oracle APEX Self-service low-code apps producing curated data

Oracle Data Lakehouse Store raw + curated data in various formats (JSON, Graph, Relational, etc.)

Oracle Analytics Cloud Dashboarding and visualizing domain data

Oracle Stream Analytics Real-time time series analysis

Oracle Data Science ML modeling and deployment

🏗️Data Mesh vs Other Concepts

Not a Data Mesh Why Not?

Single Data Lake with Catalogs Still a monolith; lacks decentralized ownership

Self-Service Analytics Alone UX layer only; doesn’t imply domain-based data ownership

Point Product / Tool Data Mesh is an architecture and mindset, not a single product

Data Fabric More about integration; Data Mesh is about decentralized ownership
 Not a Data Mesh Why Not?

📈 Business Benefits

 Unlock hidden data capital trapped in silos

 Accelerate innovation cycles with domain-driven data ownership
 Improve compliance by aligning data use with regulation and business ethics
 Enhance competitive strategy by turning proprietary data into strategic assets
 Enable multi-cloud and hybrid data strategies

🧠 Strategy Frameworks Referenced

 Michael Porter’s Competitive Strategy – Use unique data assets to enhance

competitive position.
 Design Thinking – Break silos and solve complex problems creatively.
 Jobs-to-be-Done Theory – Design data products for real end-user needs.
 Outcome-Driven Innovation – Focus on value delivery, not just tech implementation.
🔑 Key Takeaways: REST API for Autonomous Database

🧩 Purpose of REST APIs in Oracle Cloud

 Enables programmatic control over Oracle Autonomous Database without using the
web console.
 Ideal for automated scripts, CI/CD pipelines, infrastructure as code (IaC).
 Supports setting gold standards and repeatable deployments.

🔐 Security & Protocols

 REST APIs use:

o HTTPS (TLS 1.2)
o HTTP Signature-based Authentication
 o RSA-SHA256 for signing requests.
 Requires:
o An SSH key pair in .pem format for authentication.
o No need for usernames or passwords.

🧪 Languages Supported

 Scripts can be written in:

o curl, bash
o Python, Java, Node.js
o Perl, Ruby

🧾 Steps to Make a Secure API Request

1. Form HTTPS request

2. Create signing string using request parts
3. Generate signature using private key + RSA-SHA256
4. Add Authorization Header with signature

🛠️Examples of REST API Operations

✅ Create Database

 Example: create ADB in Phoenix region

 Parameters: DB name, password, CPU (e.g., 8), storage (e.g., 1TB)

⚙️Response Includes:

 Compartment/Tenancy
 Display name, DB name
 CPU/storage specs
 License model
 Creation timestamp
 Status (e.g., provisioning, available)
 Console link

🗑️Delete / 💤 Stop / ▶️Start ADB

 Simple API calls to:

o Delete an instance
o Start or stop a database
⬆️Scale ADB

 Example: scale up to 20 ECPUs

💾 Backup ADB

 Trigger a manual backup via REST API

✅ Use Case Summary:

Using REST APIs with Oracle Autonomous Database:

 Boosts automation and consistency

 Enables DevOps-style database operations
 Secures interaction via cryptographic signatures
 Supports diverse scripting environments

🧰 Oracle’s Data Toolset for Autonomous Database

🔧 Key Built-in Tools

1. Oracle Data Integrator (ODI)

o Enterprise-class ELT (Extract, Load, Transform) tool for data integration.
2. Enterprise Data Quality (EDQ)
o For profiling, cleaning, and preparing data.
3. Analytic Views
o Built into the Oracle DB to define semantic models for accessible analytics.
4. Oracle Analytics Cloud (OAC)
o For visualizing and analyzing the data beautifully and insightfully.

🚀 Integrated Platform Benefits

 No need to buy or install anything extra.

 Pre-configured & pre-deployed — ready to use.
 Shared metadata and components across all tools via a central catalog.
 Consistent user experience with best practices built-in.
 Collaboration by design, not as an afterthought.

Example:
  Hierarchies recognized in data prep → instantly accessible for analysts → semantic
modeling → accessible to data scientists → predictive model → used by CRM
developers.

🔄 Open, Not Locked-In

 Supports open standards:

o SQL
o Python
o Data formats: CSV, JSON, etc.
 Tools are accessible via the OCI Console, under Database Actions.

🧭 Navigating the Tool Suite

 Accessible from:
o Card-based interface (Development, Data Tools, Admin, Monitoring)
o Mega Menu (hamburger icon, top-left)
 Includes:
o Quick links
o Context-sensitive help
o Preferences and About info

📊 Data Loading Capabilities

 Load from:
o Local files
o Remote DBs
o Object stores (OCI, AWS S3, Azure Blob, GCP)
 Setup:
o Cloud storage locations
o Monitor data load jobs
o Explore data after loading

🔍 Explore & Profile Loaded Data

 Example:
o Loaded Movie Sales 2020
o Identified issues:
 Extra months for Q2 analysis
  Invalid day entries (e.g., “Monday” vs “MONDAY” → case mismatch)

✅ What Makes This Special

 Autonomous Database offers:

o A complete suite of tools, not a “kit”.
o Business model and metadata integration across tools.
o Fast data loading and error spotting (e.g., inconsistent date values).
o Smooth onboarding for any data team member — from DBA to data scientist.

🧪 What is CI/CD?

✅ CI/CD = Continuous Integration / Continuous Delivery (or Deployment)

 Purpose: Add automation across the development lifecycle.

 Benefits:
o Consolidate code
o Prevent “cowboy coding”
o Improve security, quality, and consistency
o Automate testing and deployment
o Support frequent, reliable releases

🔁 Why CI/CD for APEX and Oracle Database?

🧱 Traditional Challenges:

 Developers overwrite each other’s changes in shared DB environments.

 Difficult to version APEX apps and database schema.
 Manual change tracking (e.g. scripts, ZIP files, emails).
 Risky deployments due to inconsistent practices.

🛠 CI/CD Benefits in Oracle Context:

 Better change visibility.

 Automated unit tests and deployments.
 Improved security and auditability.
 Early issue detection through automated testing on cloned environments.
🧰 Essential Tools for CI/CD Pipeline
Tool Purpose

GitHub/GitLab/Bitbucket Version control for code & APEX exports

SQLcl Lightweight CLI for Oracle DB; supports Liquibase

Liquibase Change tracking & migration tool for schemas and APEX

Jenkins CI pipeline and automation engine

VS Code / SQL Developer Code editing and deployment

Terraform Automate infrastructure setup in OCI

OCI (Oracle Cloud Infrastructure) Host ADBs and run cloud-native pipelines

🧩 SQLcl + Liquibase Integration

🔹 SQLcl:

 A modern, CLI version of SQL*Plus

 Easy connection to any Oracle DB (cloud or on-prem)

🔹 Liquibase in SQLcl:

 Change tracking for schema and APEX

 Smart diffs: Understands changes without destroying objects
 Rollback support
 Automatically builds change logs and executed SQLs

🧠 Smart Behavior Example:

 If a column is missing, Liquibase issues ALTER TABLE instead of dropping/recreating the

table.

🧩 Managing APEX Apps with Liquibase

Export APEX App for CI/CD:

bash
CopyEdit
liquibase genobject --type apex --applicationid 123

 Use --split to export app into multiple files.

  Use --skipExportDate and --exportOriginalIds to avoid false diffs due to metadata
noise.

Push to Git:

 Only actual changes are included.

 Enables clean code reviews and pull requests.

🗃 Moving Data with Liquibase

 Supports liquibase dataobject <table> for inserting small metadata tables.

 Not ideal for large data sets → use ORDS, Data Pump, or custom scripts instead.

🔧 Automating Deployment Pipelines

Example CI/CD Pipeline Steps:

1. Setup Environment (via Terraform)

2. Apply Schema Changes (Liquibase)
3. Deploy APEX App (via SQLcl + Liquibase)
4. Load Metadata Tables (if needed)

☁️OCI + Terraform Integration

 Use Terraform OCI provider to:

o Create Autonomous Databases (ATP)
o Setup networks / VMs / security lists
 Example: Create 1 TB ATP with 2 lines of code
 Pay-per-second model → great for ephemeral test environments

💡 Summary: Why This Matters

 Oracle APEX + DB environments are traditionally hard to version.

 CI/CD brings:
o Automation
o Standardization
o Security
o Cost-efficient testing
 Tools like SQLcl, Liquibase, Terraform, and Jenkins simplify the process.
  Oracle is investing in this space with prebuilt tools and DevTools GitHub repo.

🧱 Full CI/CD Pipeline for APEX + Oracle Database (Dev → Prod)

🔹 1. Developer Workflow

 Developers work on tickets → code locally → push to Git repo (e.g., GitHub)
 They create branches per feature/ticket and push code
 A pull request (PR) triggers the CI/CD pipeline

🔹 2. Jenkins CI/CD Pipeline Trigger

 Jenkins server is running (on OCI VM or on-prem)

 Triggered by webhook (GitHub/GitLab/etc.)
 Jenkins has:
o Git, Terraform, SQLcl, Liquibase, PL/SQL Unit Test (UTP)
 The VM can have OCI instance principals for secure automation

🔹 3. Clone Test Environment (via Terraform or other tools)

 Spin up:
o Autonomous DB / DB VM / PDB / Docker DB / RMAN clone / Flashback DB
 Load:
o Latest schema + data (via Data Pump or Liquibase)
o APEX app (via Liquibase or SQLcl)
 Environment is ephemeral (can be destroyed later)

🔹 4. Apply Changes & Run Tests

 SQLcl:
o Applies DB schema
o Installs APEX app
o Runs PL/SQL unit tests (UTP)
 Jenkins:
o Checks for errors in logs (e.g., ORA- errors)
o Reports status to developer
🔹 5. Post-PR Actions

 ✅ If test passes:
o Code is merged into main
o Development environment is destroyed
 ❌ If test fails:
o Developer is notified with logs
o Developer fixes code → new PR starts cycle again

🚀 Deployment to Production

🔄 Two options:

Option Description

🛠 Manual Download Jenkins-generated artifacts and apply manually

⚙️Automated Use pipeline to deploy to PROD or Pre-PROD environments

 Since the same code has already been tested multiple times on clones, deployment should
be safe and predictable.

🧯 Rollback Options
Method Description

🔄 Flashback DB If using pluggable DBs or guaranteed restore points

🔙 Liquibase rollback Revert DB schema & APEX app

↩ Restore
Flashback or import small tables
metadata/data

➡ Roll Forward Fix the issue, patch main, redeploy

🔧 Dev/Test Environment Creation Options

Tool/Method Description

Terraform Create DBs/VMs/load balancers in OCI (IaC)

Autonomous DB Clone Fast clone of ATP with APEX

Multitenant PDBs Up to 3 free in Oracle 19c

 Tool/Method Description

Flashback DB For non-CDBs or DBs w/o multitenancy

RMAN / Data Pump For on-prem or non-ATP clones

Docker / VMs Local reusable instances

REST APIs (ORDS) REST services to manage clones or schema

✅ Key Oracle Tools Used

Tool Role

SQLcl CLI tool for schema and APEX operations

Liquibase Schema and APEX change tracking/versioning

Jenkins CI/CD orchestration

Terraform Infra-as-code for provisioning

UTP (PL/SQL) Unit testing framework

ORDS REST APIs for DB operations

Data Pump Schema/data export/import

🧠 Final Thoughts

 You don’t have to be on OCI – this process works on-prem, too.

 CI/CD helps prevent:
o Overwrites
o Cowboy coding
o Manual deployment risks
 Enables:
o Repeatable, tested, and traceable database & APEX app deployments
o Better collaboration across devs
o Rollback or forward flexibility
🔐 Oracle Maximum Security Architecture – Summary

1. Security Starts with Configuration

 Implement secure configurations and monitor for drift.

 Validate against CIS benchmarks or DISA STIGs.
 Tools: DBSAT (Database Security Assessment Tool) and Data Safe.
2. Encrypt Data

 Protect data at rest and in motion.

 Protect encryption keys from loss or theft.
 Oracle supports:
o Native network encryption
o TLS-based encryption

3. Control Access

 Enforce least privilege (grant only necessary privileges).

 Remove excessive/dormant accounts.
 Use multi-factor authentication, PKI, or Kerberos.
 Use Privilege Analysis (Oracle Enterprise Edition only) to monitor and reduce unused
privileges.

4. Plan for Failure

 Audit all access, especially failed logins, DCL, DDL.

 Use native auditing and network-based SQL monitoring.
 Know where sensitive data resides — use DBSAT and Data Safe to identify and
quantify it.

5. Prevent, Detect, Assess

 Prevent unauthorized data access:

o Use context-aware access control.
o Block export/sniffing attacks.
 Detect anomalous access patterns.
 Assess configurations and risks continuously.

6. Attack Vectors & Weak Points

Attackers target:
  Admin accounts (for full control).
 End-user accounts (stepping stone).
 Apps (often more exposed).
 Backups & Exports (stealthy full theft).
 Unpatched DBs (exploit kits).
 Non-prod clones (often overlooked).

7. Centralized User Management

 Enterprise User Security: Uses Oracle LDAP.

 Centrally Managed Users: Uses Microsoft AD (Oracle 18c+).

8. Baseline Security Posture

All Oracle DBs should:

 Be properly configured and patched.

 Have auditing enabled.
 Apply least privilege.
 Use strong authentication.
 Encrypt data at rest/in transit.
 Know and monitor sensitive data.

Most of these baseline protections are available at no extra cost, especially with Oracle Cloud
services (includes Data Safe).

🏛️Oracle Maximum Availability Architecture (MAA) – Summary

🔧 The Problem

 Downtime is costly:
o $350K/hour average downtime cost
o 87 hours/year average downtime
o $10M per major outage
o 91% of companies had unplanned outages in 2 years
 Data loss = high business risk
💡 Key Concepts and Terminology
Term Description

Redundant components for uninterrupted service during hardware

High Availability (HA)
failure

Scalability Ensures consistent performance under increased load

Rolling Updates/Patching Minimize planned downtime

Disaster Recovery (DR) Fails over to alternate site in case of complete site failure

RPO (Recovery Point

Max allowable data loss
Objective)

RTO (Recovery Time Objective) Max allowable downtime

⚙️Resilience Engineering with Chaos Engineering

 Proactively break things to validate failover, resiliency

 Test failures in:
o Network, storage, servers, human error, power failure
 Use real-world simulations for design

🧱 Oracle MAA Blueprint

🔍 Purpose

Oracle's MAA is a best-practice framework to ensure:

 High availability
 Data protection
 Disaster recovery
 At low cost and complexity

📘 MAA Components

Goal Tools/Features

Data Protection Flashback, Zero Data Loss Recovery Appliance

Active Replication Active Data Guard, GoldenGate

Scale-Out RAC, ASM, Sharding

 Goal Tools/Features

Continuous Availability Application Continuity (AC), Global Data Services (GDS)

🧩 MAA Deployment Choices

Offering Description

Oracle DB Cloud Service RAC optional, ZDM, Active Data Guard

Exadata Cloud Service Exadata performance, HA/DR, ExaCC for on-prem

Autonomous DB (ADB) Built on RAC + Exadata, 2 types: shared & dedicated

Hybrid / Multicloud Use Oracle RAC/ADG with third-party clouds via GoldenGate

Partner Example Megaport provides redundant networking for hybrid setups

🔄 Technologies for HA/DR

Technology Role

RAC High availability, scale-out, session failover

Transparent App Continuity Protect in-flight transactions

Active Data Guard DR with read-only standby, fast failover

GoldenGate Logical replication, hybrid DR support

🧠 Summary Points

 MAA = Comprehensive blueprint for both planned and unplanned events

 RPO/RTO requirements define which tier of architecture is best
 Oracle provides tiered MAA architectures to match business needs
 Business continuity planning is essential, not optional
Oracle Database Cloud Migration

✅ Key Concepts from the Lesson

🔹 Reasons for Migration

 Upgrade to new database versions

 Move to new hardware
 Re-platform to Oracle Cloud

🔹 Migration Strategies

1. Downtime Tolerant
o Suitable when apps can be shut down during migration (e.g., business hours only).
o Simpler and more flexible options available.
2. Zero Downtime (Continuous Operations)
o Required when apps must run 24/7.
o Tools like Zero Downtime Migration (ZDM) and GoldenGate are used.

🔹 Cloud Migration Tools

Tool Purpose Notes

CPAT (Cloud Pre-Migration

Checks compatibility of source database Now integrated into DMS
Advisor Tool)

DMS (Database Migration

Automates migration (focus: ease of use) Less control, but simpler
Service)

ZDM (Zero Downtime Fine-grained, supports ExaCC and non-

Requires more setup
Migration) autonomous targets

Compares and validates source vs. target

GoldenGate Veridata Ensures data accuracy
data

Different from database

OCI Application Migration For moving apps and VMs to cloud
migration

🔹 Migration Flow

1. Evaluate → Use CPAT or Migration Advisor

2. Plan → Validate source/target compatibility
3. Migrate → Use DMS, ZDM, or GoldenGate depending on need
4. Validate → Use GoldenGate Veridata
🔹 Target Environments in OCI

 Oracle Database Cloud Services

 Exadata Cloud@Customer (ExaCC)
 Oracle Autonomous Database (Autonomous Shared/Dedicated)

✅ Oracle Database Upgrade Best Practices Summary

📌 1. Choose the Right Target Version

 Target version = Oracle 19c

o It is the current Long Term Support (LTS) version.
o Support through April 2027 (5 years Premier + 2 years Extended).
 Avoid upgrading to 18c or 21c for production:
o 18c is out of support.
o 21c is innovation-only (2 years support), not LTS.

📌 2. Use AutoUpgrade Tool

 Always use the latest version from My Oracle Support.

 Two simple commands:
o analyze → Checks for readiness
o deploy → Performs upgrade and post-tasks
 Compatible with 19c, 18c, 12.2

📌 3. Pre-Upgrade Preparation

 Apply latest RU (Release Update) after installing 19c.

 Gather statistics in advance:
o Dictionary and Fixed Object stats → ideally 7 days prior
 Use SQL Plan Management (SPM):
o Prevents SQL execution plan regressions

📌 4. Performance Stability Tools

 Use these tools before and after upgrade:

o SQL Plan Management
o SQL Tuning Advisor
o SQL Patch
 o Database Replay (for load simulation, requires license)

📌 5. Fallback Strategy

 Use Flashback Database + Guaranteed Restore Point

o Best option, but only available in Enterprise Edition
o Cannot be used after go-live
 Alternatives:
o Data Pump
o Database downgrade
 Always test fallback procedures in advance

📌 6. Compatible Parameter Handling

 Set compatible = 19.0 (after testing period: 7–10 days)

 Do not raise compatible immediately unless:
o Downtime window doesn’t allow later change
o You are aware of rollback limitations
 Once set, you cannot Flashback or downgrade

📌 7. Post-Upgrade Best Practices

 Do NOT regather all stats (keep existing ones unless stale)

 Avoid gathering system stats unless you’ve tested thoroughly
 Gather Fixed Object stats only when system is warmed up
 Review and reduce custom parameters, especially:
o Underscore parameters (e.g. _some_hidden_param)
o Event parameters (e.g. event="10949 trace name context forever")

📌 8. General Tips

 Run as close to default Oracle configurations as possible

 Follow vendor-specific settings (e.g., EBS recommendations)

🟨 Summary of Upgrade Flow:

1. 🎯 Download Oracle 19c

2. 🧰 Apply latest RU
3. 🔄 Get latest AutoUpgrade
4. 🔍 Analyze (pre-check)
5. 🚀 Deploy upgrade
6. 🔁 Monitor with AutoUpgrade logs
7. ⚠ Set fallback plan (Flashback or Downgrade)
8. 🔧 Adjust compatible after stability
9. 📊 Monitor post-upgrade performance
✅ MySQL HeatWave Overview

 MySQL is the world’s #1 open-source database, second overall to Oracle Database.

 Used by leading companies: Twitter, Facebook, Netflix, Uber, etc.
 Popular in LAMP stack and across industries (finance, manufacturing, telecom,
education).

🔥 What Is HeatWave?

 A massively parallel, high-performance in-memory query accelerator.

 Adds analytics and machine learning to MySQL without requiring ETL or a separate
analytics DB.
 No need for separate services—OLTP + OLAP + ML in one system.

🚀 Key Benefits

 Real-time analytics directly on MySQL data.

 Eliminates ETL complexity, latency, security risk, and costs.
 Faster and more cost-efficient than comparable services.
 Works seamlessly with Amazon Aurora-based applications.
 Fully managed on OCI (Oracle Cloud Infrastructure).
🧠 HeatWave AutoML

 Built-in machine learning lifecycle:

o Training, deployment, and explainability
o No additional cost
o Uses SQL commands (no need for new tools/languages)
 Integrated with Jupyter and Apache Zeppelin

⚙️MySQL AutoPilot

Automates performance tuning:

 Auto provisioning of HeatWave nodes

 Auto shape prediction for optimal compute
 Auto query plan improvement
 Auto thread pooling
 Auto error recovery
 Uses machine learning to optimize operations over time

💾 MySQL HeatWave Lakehouse

 Queries up to 500 TB of data in object storage (e.g., CSV, Parquet).

 Enables analytics across:
o MySQL data
o Object store data
o Combined queries using SQL
 No data duplication or movement needed

🔐 Security & Compliance

 Built on OCI Gen 2 Cloud Infrastructure

 Encrypted data at rest and in transit
 Complies with GDPR, PCI, HIPAA
 Security patches from MySQL core dev team
 Only service using MySQL Enterprise Edition (vs. others using Community Edition)
💼 Responsibility Model

 Customer: Schema design, query optimization, access policy

 Oracle/MySQL Team:
o Automated backups, patching, monitoring
o Handles security and OS maintenance

🧠 Key Takeaways

 One service for transactions, analytics, and machine learning.

 No ETL required
 In-database ML with AutoML and explainability
 Advanced automation with AutoPilot
 Secure and fully managed with enterprise-grade compliance
 Cost-effective and high performance

🔷 1. Challenges of Modern Applications

 High data volume and speed (e.g., IoT, real-time data)

 Demand for low latency and real-time responses
 Data is dynamic and diverse (structured, semi-structured, unstructured)
 Need for always-on availability
 Requirement for on-demand scalability
 Preference for low operating costs
🔷 2. Oracle NoSQL Database Cloud Service: Overview

 Fully managed, serverless cloud DB service

 Elastic scaling of throughput and storage
 Offers < 10 ms latency
 Supports multiple data models:
o Document (JSON)
o Key-value
o Columnar (fixed schema)

🔷 3. Key Features

 Developer-friendly APIs
 Integration with popular tools
 Enterprise-grade security (encryption at rest/in motion)
 Multi-cloud and hybrid deployments supported
 Cost-effective and predictable performance
 No infrastructure or server management required

🔷 4. Use Cases

 Social media: Store and query user content (tweets, chat, ratings)
 IoT: High-speed data ingestion from sensors
 Customer 360: Aggregating customer behavior data
 Catalogs: Products, users, devices, BOM systems
 Online gaming: Real-time stats, leaderboards
 Profile management: Personalized UI experiences
 Others: Anomaly detection, advertising, content management

🔷 5. Deployment Flexibility

 Run anywhere: On-prem, Oracle Cloud, or other clouds

 Hybrid support: Some data on-prem, some in cloud
 Uniform APIs and data model support across environments
🔷 6. Data Model Types

 Key-Value: Basic pairing of key and value

 Document (JSON): Flexible, schema-less structure
 Columnar: Relational style, fixed schema
 All models interoperate seamlessly with unified SQL/API access

🔷 7. Security and Governance

 Managed via Oracle Identity and Access Management (IAM)

 Compartment-based access control
 Policies with SQL-like syntax for access
 Encryption at rest and in motion

🔷 8. Elastic Scalability & Billing

 Dynamic provisioning of:

o Read Units (RU)
o Write Units (WU)
o Storage Capacity
 Charges based on actual usage:
o RU = 1 for eventual consistency
o RU = 2 for absolute consistency (ensures fresh data)

🔷 9. Architecture & Developer Tools

 Client-server model:
o App connects to NoSQL driver (Java, Python, Node.js, Go, Spring, etc.)
o Executes reads/writes via API
 Tables have:
o Table Definition (schema, data types, keys)
o Capacity Settings (throughput, storage)
 Shard keys used for locality and faster access

🔷 10. Performance & Infrastructure

 Built on Oracle Gen 2 Cloud:

 o Latest CPUs/GPUs, NVMe SSDs
o Non-oversubscribed infrastructure
 Predictable high throughput and low latency

✅ Bottom Line

Oracle NoSQL Database Cloud Service provides:

 Flexibility for modern app development

 Low-latency, highly scalable data handling
 Fully managed, cost-effective infrastructure
 Support for real-time and varied workloads
 Seamless integration with Oracle and third-party tools