Repository navigation
LAB 10 ‐ Instance Creation
An instance in Oracle Cloud Infrastructure is a virtual server (compute resource) that runs applications in the cloud.
It provides the necessary computing power such as CPU, memory, storage, and networking—just like a physical computer.
🧠 Simple Definition
An instance is a virtual machine in OCI used to run applications and workloads in the cloud.
⚙️ Key Components of an Instance
🧠 CPU (Processor) – Executes tasks
💾 Memory (RAM) – Runs applications
💽 Storage – Stores data (boot volume, block storage)
🌐 Network – Connects using public/private IP
Virtual Machine (VM)
Runs on shared hardware
Cost-effective and flexible
Bare Metal Instance Dedicated physical server High performance and full control 🧾 Example
You create an instance in Oracle Cloud Infrastructure and:
Install Linux or Windows Deploy a web server (Apache/Nginx) Access it via SSH or RDP
👉 It works like your own server hosted in the cloud.
🧠 Real-Life Analogy
An instance is like renting a computer online instead of buying one physically.
🚀 Why Use Instances? No need for physical hardware Easy to scale (increase CPU/RAM) Pay-as-you-go pricing High availability
✅ Short Definition (Exam Ready)
In OCI, an instance is a virtual server that provides compute resources to run applications in the cloud.
- Standard default option
- Always available when you launch
👉 Example:
Production web server (must not stop)
- Low-cost but can be terminated anytime
👉 Example:
Batch jobs, background processing that can restart if interrupted.
- You reserve capacity in advance
- Ensures resources are available when needed
👉 Example:
Planning a big app launch → reserve instances beforehand
- Physical server reserved for your use only
- No sharing with other tenants
👉 Example:
Compliance or security-sensitive applications
| Type | Availability | Cost | Use Case |
|---|---|---|---|
| On-Demand | High | Medium | General workloads |
| Preemptible | Low | Low | Fault-tolerant jobs |
| Reserved Capacity | Guaranteed | Planned | Predictable workloads |
| Dedicated Host | Exclusive | High | Security/compliance |
🧠 Simple Understanding
On-Demand → Use anytime
Preemptible → Cheap but risky
Reserved → Pre-book resources
Dedicated → Full hardware control
In exams or docs:
Sometimes only On-Demand + Preemptible are called “capacity types” Others (Reserved & Dedicated) are considered capacity options/features
✅ Short Definition
In OCI, capacity types define how compute resources are allocated, including On-Demand, Preemptible, Reserved Capacity, and Dedicated Hosts, each offering different levels of availability, cost, and control.
NOTE:-you cannot select multiple Availability Domains (ADs) for a single instance in Oracle Cloud Infrastructure.
🧠 Simple Explanation AD = Data center
Instance = Server
A single server cannot physically exist in multiple data centers at the same time.
🔄 What if you need high availability?
Instead of one instance in multiple ADs, you should:
✅ Recommended Approach Create multiple instances Place them in different ADs Use: Load Balancer Auto Scaling
👉 This ensures:
High availability Fault tolerance 🧾 Example Instance-1 → AD-1 Instance-2 → AD-2 Load Balancer distributes traffic
If AD-1 fails → AD-2 still works ✅
✅ Short Definition
In OCI, a single instance can be deployed in only one Availability Domain. To achieve high availability, multiple instances must be deployed across different ADs.
Vendor Lock-in is the risk where a customer becomes dependent on a vendor for products and services, unable to switch to another vendor without substantial switching costs or Vendor Lock-in occurs when an organization’s infrastructure, applications, or data become so tightly integrated with a specific cloud provider's proprietary APIs, services, and architecture that switching to another provider (like AWS or Azure) becomes economically or technically prohibitive.
- Proprietary Services: Using OCI-exclusive tools (like OCI Autonomous Database) that lack direct, compatible equivalents in other clouds.
- API Coupling: Writing application code that directly calls OCI-specific APIs for identity, storage, or messaging.
- Data Gravity: When the sheer volume of data stored in OCI makes the cost and time of moving that data to another provider unsustainable.
- Use Open Standards: Use open-source databases (like PostgreSQL) instead of proprietary ones when possible.
- Containerization: Use Kubernetes (OKE). Since Kubernetes is a standard, a containerized app can be moved from OCI to any other cloud provider with minimal changes.
- Infrastructure as Code (IaC): Use Terraform. Even if you change clouds, your infrastructure deployment logic stays consistent in the same language.
What is T-Shirt Sizing?
T-shirt sizing is a simple way to classify resources (like compute instances) into sizes such as:
Small (S), Medium (M), Large (L), Extra Large (XL)
Instead of using exact CPU/RAM numbers, it groups configurations into easy-to-understand categories.
T-Shirt Sizing simplifies the complex process of selecting OCI VM Shapes by creating standard 'tiers' of performance.
- Speed: Faster decision-making for dev teams (e.g., "Just give me a Medium VM").
- Budgeting: Financial teams can assign standard costs to 'Small', 'Medium', and 'Large' sizes rather than tracking thousands of unique configurations.
- Standardization: Reduces configuration drift by ensuring all production web servers use the same 'Medium' specification.
T-Shirt Sizing is not static. As your application scales, OCI’s Flexible Shapes (Flex Shapes) allow you to adjust the OCPU and RAM within that "T-shirt" category without needing to change the underlying infrastructure instance, making it a perfect match for modern OCI architecture.
Architects use T-Shirt sizing to standardize resource requests, making capacity planning, budgeting, and deployment significantly faster.
| Size | Use Case | OCI VM Shape (Example) | Resources |
|---|---|---|---|
| Small (S) | Dev/Test, Low-traffic Web App | VM.Standard.E5.Micro |
1 OCPU, 1GB RAM |
| Medium (M) | Mid-tier App, Production Web | VM.Standard.E5.Flex |
2 OCPU, 16GB RAM |
| Large (L) | Heavy Database, Analytics | VM.Standard.E5.Flex |
8 OCPU, 64GB RAM |
| XL | High-performance Clusters | BM.Standard.E5.128 |
Bare Metal (128 OCPU) |
Note: In OCI, we prefer using Flexible Shapes (Flex) for Medium and Large sizes, allowing us to 'Right-Size' the exact OCPU and RAM without needing to migrate to a different instance type.
A Shape defines the compute capacity of your instance. It is the architectural foundation that dictates the performance ceiling of your workload.
- OCPU Count: Determines the processing power.
- Memory (RAM): Determines the instance's ability to handle concurrent data processes.
- Network Bandwidth: The throughput allocated to the instance.
Unlike static cloud environments, OCI Flexible Shapes allow you to dynamically adjust your OCPU and memory allocation even after the instance is running. This enables Right-Sizing, where you pay only for the exact hardware your workload requires, optimizing both performance and cost.
📌 What is OCPU?
In Oracle Cloud Infrastructure, an OCPU (Oracle CPU) represents:
1 physical CPU core = 2 hardware threads
📌 What is vCPU?
In most cloud providers (like AWS, Azure), a vCPU (virtual CPU) represents:
1 hardware thread of a CPU core
| Feature | OCPU | vCPU |
|---|---|---|
| Definition | 1 physical core | 1 thread |
| Threads | 2 threads | 1 thread |
| Performance | Higher | Lower (per unit) |
🧠 Simple Understanding
1 OCPU = 2 vCPUs
🧾 Example If you choose 2 OCPUs in OCI → 👉 You get performance similar to 4 vCPUs
🎯 Why This Matters?
Helps in comparing pricing across cloud providers Better capacity planning
✅ Short Definition
OCPU in OCI represents a full physical CPU core (2 threads), while vCPU represents a single thread of a CPU core used in other cloud platforms.
📌 Overview
Oracle Linux → Traditional operating system managed by users
Oracle Autonomous Linux → Self-managing OS with automation using Oracle services
| Feature | Oracle Linux | Oracle Autonomous Linux |
|---|---|---|
| Management | Manual (user-managed) | Automated (self-managing) |
| Patching | Manual updates | Automatic patching (no downtime) |
| Security | User responsibility | Built-in automated security |
| Monitoring | Manual setup | Integrated monitoring & alerts |
| Downtime | Possible during updates | Near zero downtime |
| Automation | Limited | High (uses OCI automation) |
| Use Case | Full control environments | Cloud-native, low-maintenance apps |
🧠 Simple Difference
Oracle Linux = You manage everything
Oracle Autonomous Linux = Oracle manages it for you automatically
🧾 Example Oracle Linux: You manually install updates, monitor system, fix issues
Oracle Autonomous Linux: System automatically patches, monitors, and secures itself
🎯 When to Use Use Oracle Linux → when you need full control
Use Autonomous Linux → when you want automation and less maintenance
✅ Short Definition
Oracle Linux is a user-managed operating system, while Oracle Autonomous Linux is a self-managing OS that automates patching, security, and monitoring in OCI.
Step 1: Navigate to burger menu, select Compute and click on Instances.
Step 2: Click on create instance. Parent compartment should be "Hub_Compute_and_Storage_Compartment"
Step 3- In the Image section, click Change Image. Select Oracle Linux as the image (it is free in OCI, you can select the image as per your requirement).
Step -4 Click on Change Shape, then select Virtual Machine. Choose the required Shape Series based on your workload. Click on the shape name to customize the configuration by selecting the appropriate number of OCPUs and Memory as per your requirement. Once the desired configuration is selected, click on Select Shape to proceed.
Step -5 Click on Next. Now page moves to Networking. In Primary Network, Select "select existing virtual cloud network". (Always select public subnet in the VCN).
Step -6 In "Private IPv4 assignments, select automatically assign private IPv4 address. In Public IPv4 address assignments, enable "Automatically assign public IPv4 address"
Step -7 Select 'Generate a key pair for me' in the Add SSH Key section, then download and save both the (private and public) keys.
Step -8 Enable "Specify a custom boot volume size and performance setting" (In real time, select custom Boot and select 100 GB minimum to upgrades)
Step -9 Check the Estimated Cost and review all the details
Step 10- Now the instance is created.
Step 11-Next, download the software PuTTY key Generator and convert the private key into .ppk format
Step 12- Now we login to the server that we have created just now (Bastion-server) by using putty application and in the Host Name (IP Address) we put our public IP address
