Cloud Technology Essentials

By reading these notes, you will gain knowledge tailored to your role, whether you are a beginner or practitioner focusing on Cloud technology foundations or a manager looking to master Cloud leadership and strategy.

1. What is Cloud Computing?

Cloud computing means using someone else’s computer (servers) and other compute services over the internet to store data, run applications, and manage services without buying any hardware.

Cloud Computing is like “Renting a House” instead of buying one. You don’t have to build the walls or fix the roof; someone else (the Cloud Provider like AWS, Azure, or GCP) does that. You just move in with your furniture (your code and data) and pay only for the time you stay there.

It is the on-demand delivery of IT resources (servers, storage, databases) over the internet with pay-as-you-go pricing. Instead of buying and maintaining physical data centers, you access technology services on an as-needed basis.

1. No need to buy servers, you simply “rent” from Cloud service providers like AWS, Azure, GCP.
2. You pay only for what we use.
3. Access from anywhere via internet.
4. Cloud gives speed, scalability, and less operational headache.

Cloud = Someone else’s computer, available on demand, and you pay only for use.

Before cloud → company needs to buy servers, AC room, networking, admins, etc. long process.
With cloud → click a button and create a server in minutes.

1.1. Cloud computing providers list

1.2. Traditional IT vs Cloud Computing

This table highlights the differences between traditional on-premise IT and modern Cloud Computing.

1.2 The “Shared Responsibility Model”

One of the biggest mistakes in Cloud is assuming the provider handles all security. In a DevSecOps culture, we follow the Shared Responsibility Model.

1.3 Secure Cloud Practice

1. Immutability: In the cloud, we don’t fix servers; we replace them. If a server has a security vulnerability, kill it and launch a new, patched version from an automated script.
2. Infrastructure as Code (IaC): Every item in the “Cloud Computing” should be managed by code (Terraform/CloudFormation). Clicking buttons in the console is for beginners; writing code to build the cloud.
3. Observability: Because “Access is from anywhere,” you must log everything. Every login, every file change, and every network connection should be monitored using Cloud-native tools.

2. Why Cloud Became Popular?

In the old days, starting a software business was like starting a factory you needed a huge building, expensive machines, and months of setup. Cloud changed the game. It turned infrastructure into a utility, just like electricity. you just plug it in and pay for the units you consume. Cloud became popular because it removed the “fear of failing” by making it cheap and fast to try new ideas.

The “Big Three” Problems Cloud Solved: –

1. Speed (Agility): – Get your website live in minutes, not months.
2. Cost (Economy): – No need to block lakhs of rupees in hardware. Pay only for the minutes your server runs.
3. Flexibility (Elasticity): – If your website gets a sudden surge of traffic (like a Diwali sale), the cloud grows automatically. When the sale ends, it shrinks back to save money.

2.1. Business Benefits of Cloud

From a business leadership perspective, Cloud is a “Growth Accelerator.”

• Zero Upfront Investment: – Move from CapEx (Capital Expenditure) to OpEx (Operational Expenditure).
• Just-in-Time Infrastructure: – No more “Guessing Capacity.” You get what you need, exactly when you need it.
• Reduced Time-to-Market: – Spend time on your business logic, not on fixing server cables.
• Global Footprint: – Start in Mumbai and expand to London or New York with a few clicks.

2.2. Cloud Advantages

2.3. Real-World Use Cases

To make your KB more practical, here are common scenarios where Cloud is the only logical choice:

1. Content Streaming (Netflix/Hotstar): Using CDNs (Content Delivery Networks) to stream high-quality video to millions of users simultaneously with zero buffering.
2. E-commerce Websites: During “Big Billion Day” or “Great Indian Sale,” traffic can increase by 100x. Cloud handles this spike and then scales down at midnight to save costs.
3. Backup & Disaster Recovery: Small companies can store terabytes of data in AWS S3 for a very low cost, ensuring their data is safe even if their office computer is damaged.
4. Startups & MVPs: A founder can launch a “Minimum Viable Product” for almost zero cost using Free Tier services, only paying once they get real customers.

Guru Tip – “Cloud is not just a place to store data; it is a way to automate your business.” A DevSecOps engineer doesn’t just look at the cost of the cloud, also look at the security and stability it provides. If you use the cloud but still configure everything manually, you are not getting the full benefit. Always aim for Automation.

3. Key Characteristics of Cloud Computing

According to NIST SP 800-145, there are 5 essential characteristics. If even one is missing, it is not Cloud Computing.

3.1 Design for Failure: The “Chaos” Mindset

In the Cloud, hardware will fail. Cables will break. Power will go out. A Guru doesn’t try to prevent failure; they prepare for it.

The Pessimist’s Rule of Thumb: –

“Everything fails, all the time.” – Werner Vogels, CTO of Amazon

• Assume Hardware Failure: – Don’t put everything in one server. Use Auto-Scaling Groups.
• Assume Data Center Outage: – Don’t stay in one building. Deploy across Multiple Availability Zones (Multi-AZ).
• Assume Regional Disaster: – For critical apps, replicate data across Multiple Regions (e.g., Mumbai and Singapore).
• Assume Traffic Surges: – Prepare for “The Slashdot Effect” where thousands of users hit your site at once. Use Load Balancers.
• Assume Human Error: – People will make mistakes. Use Infrastructure as Code (IaC) so you can “Redeploy” your whole cloud in minutes if someone deletes a database.

3.2. CSA and NIST

You don’t have to invent security rules. Two global organizations have already done the hard work for you.

Cloud Security Alliance (CSA) and the National Institute of Standards and Technology (NIST) are two prominent organizations that play significant, complementary roles in defining and promoting cloud security best practices.

4. Types of Cloud Deployment Models

Deployment models describe who owns the hardware and who has access to it.

• Public Cloud is like a Public Bus: Many people share the same bus, it’s cheap, and follows a fixed route.
• Private Cloud is like a Personal Car: Only you use it, you have total control, but you pay for all maintenance.
• Hybrid Cloud is like having a Car and a Bus Pass: You use the car for private family trips (Sensitive Data) and the bus for daily office commutes (Web traffic).

Understanding how cloud resources are deployed and applied is critical for designing secure and scalable architectures.

1. Core Cloud Deployment Models

These are the standard models used by 99% of businesses today.

2. Specialized Deployment Models – Advanced

DevSecOps Note on Deployment

1. Data Residency: – Under India’s DPDP Act (Digital Personal Data Protection), certain sensitive data must stay within Indian borders. A Guru uses Distributed Cloud or Local Zones (like AWS Mumbai/Hyderabad) to ensure legal compliance.
2. The “Shared Secret” Risk: – In a Public Cloud, if you don’t configure your “Security Groups” correctly, you are essentially sitting in a room with a thin curtain. Always use Encryption at Rest and In Transit to make your data unreadable to others sharing the hardware.
3. Hybrid Connectivity: – When connecting Private and Public clouds, the “VPN Tunnel” is your weakest link. Always monitor this connection for “Man-in-the-Middle” attacks.

5. Cloud Service Models

1. A list of Cloud Service Models XaaS

As a DevSecOps Practitioner, you must understand how the “Responsibility” shifts as you move from IaaS to SaaS:

1. IaaS is the most “Risky” for beginners: – Because you own the OS. If you don’t patch the Linux/Windows kernel, your cloud is vulnerable.
2. SaaS is the “Easiest” but “Least Flexible”: – You can’t change how the software works; you can only change who can log in.
3. The “Serverless” Revolution (FaaS): – It is the most secure model for many because the “Server” exists only for milliseconds while your code runs. This makes it very hard for a hacker to “stay” inside your system.
4. Identity is King (IDaaS): – In modern cloud, we don’t care about IP addresses as much as we care about “Who is this user?” Always enforce MFA (Multi-Factor Authentication) across all service models.

6. Global Infrastructure – Region, Availability Zone, Edge Location

7. Major Cloud Services by category

Services from 1 to 6 is very important should learn.

7.1. Compute Services

7.2. Storage Services

7.3. Databases Services

7.4. Networking in Cloud

7.5. Identity and Access Management (IAM)

7.6. Monitoring & Observability

Good if you learn this below listed services

7.7. Cloud Security Essentials

7.7.1. Cloud Security Posture Management (CSPM) Checklist

Before you go live, use this checklist to ensure your cloud isn’t just “running,” but “hardened.”

1. Identity & Access (The “Who”)
   • [ ] Root Account Lockdown: Is the Root/Global Admin user secured with MFA and the access keys deleted?
   • [ ] Least Privilege: Do users and services have the minimum permissions required to do their job?
   • [ ] No Hardcoded Keys: Are you using OIDC or IAM Roles instead of storing long-lived Access Keys in your code/GitHub?
   • [ ] MFA Everywhere: Is Multi-Factor Authentication mandatory for every human user?
2. Network Security (The “Where”)
   • [ ] Public Access Blocked: Are your S3 buckets and Databases blocked from the public internet?
   • [ ] Security Group Audit: Are you only opening specific ports (e.g., 443 for HTTPS) rather than “All Traffic”?
   • [ ] Private Subnets: Are your backend servers and databases placed in Private Subnets with no direct public IP?
3. Data Protection (The “What”)
   • [ ] Encryption at Rest: Is every disk, database, and storage bucket encrypted using KMS (Key Management Service)?
   • [ ] Encryption in Transit: Is all traffic moving between your users and the cloud protected via SSL/TLS (HTTPS)?
   • [ ] Backup Validation: Do you have automated backups, and have you tested a “Restore” recently?
4. Governance & Logging (The “How”)
   • [ ] Audit Trails Enabled: Is AWS CloudTrail (or equivalent) turned on in all regions to log “Who did what”?
   • [ ] Billing Alarms: Have you set up a budget alert to notify you if costs spike unexpectedly?
   • [ ] Automated Scanning: Are you using a tool (like AWS Config or GuardDuty) to scan for real-time security threats?

7.8. Cloud Scalability and Performance Optimization

7.9. Cost Management

7.10. Cloud Disaster Recovery

7.11. Migration to Cloud

7.12. Cloud Best Practices

8. Compare AWS and Azure and Google Cloud Documentation

9. Pet vs Cattle Analogy (Concept)

A fundamental concept in Cloud computing and DevOps that describes two different approaches to managing servers and infrastructure.

1. Encourages automation, not manual work.
2. Makes infrastructure scalable and reliable.
3. Enables immutable infrastructure (no manual changes on servers).
4. Reduces downtime.
5. Supports CI/CD and modern deployment workflows.