01 - What is GCP and Cloud Computing

Welcome to Day 01 of Learn GCP in 30 Days! Today is your complete foundational guide to Cloud Computing and Google Cloud Platform (GCP).

The Core Problem: Why Cloud Computing Exists

What Existed Previously:

Decades ago, if a company wanted to launch a software application or website, they had to build an on-premise data center. This required buying physical server racks, routers, switches, backup generators, cooling systems, and hiring a dedicated team of system administrators.

Problems Faced:

• 💸 Huge Upfront Financial Risk: Buying physical servers required tens of thousands of dollars before getting a single user.
• ⏳ Slow Time-to-Market: Ordering, shipping, racking, and configuring hardware took months.
• 📉 Inflexible Capacity: If your website went viral on Friday, servers crashed under load. If traffic dropped on Monday, expensive hardware sat idle, consuming electricity without generating value.
• 🔧 Hardware Maintenance Burden: Component failures (hard drive crashes, RAM degradation, power unit failures) had to be manually diagnosed and fixed by on-site technicians.

How Present Technology Solves It:

Cloud computing converts physical computer hardware into an on-demand utility service delivered over the Internet. Just like plugging an appliance into a wall socket and paying only for the electricity consumed, cloud computing lets developers provision virtual servers, databases, and storage in seconds and pay only for the exact seconds of computing time used.

Financial Transformation: CapEx vs. OpEx

Cloud computing radically changes how companies spend money on technology:

Cloud Service Models: A Developer's Real-World Walkthrough

To understand cloud service models (IaaS, PaaS, SaaS, FaaS), imagine you are a developer building a web application (like an online store or mobile app backend).

When you build an app, your goal is simple: Write code, save data, and serve users.

However, running that code requires doing dozens of background tasks:

1. Setting up a computer / server.
2. Installing an Operating System (e.g. Ubuntu Linux).
3. Installing software runtimes (e.g. Node.js, Python, Java).
4. Configuring web servers (e.g. Nginx, Apache).
5. Opening network ports & firewall security rules.
6. Managing security patches and OS updates.
7. Restarting the server if it crashes or run out of RAM.
8. Adding more servers when 100,000 users visit at the same time.

GCP gives you 4 choices depending on how many of these tasks you want to manage yourself versus letting Google handle them automatically:

Figure: Shared Responsibility Breakdown across Cloud Service Models

Option 1: Infrastructure as a Service (IaaS)

Developer Experience: "Give me a blank computer in the cloud. I will set up everything myself."

• What YOU do: You rent a raw virtual machine (VM). You SSH into it, choose Linux or Windows, install Node.js/Python manually, configure Nginx, set up firewall ports, apply security patches, and configure backup scripts.
• What GCP does for you: GCP provides the raw virtual CPU, RAM, hard drive, and network wiring.
• GCP Product: Compute Engine (VMs), Persistent Disks, VPC Networks.
• When to choose it: When you need 100% full control over the Operating System or are migrating legacy software that requires custom server setups.

Option 2: Platform as a Service (PaaS)

Developer Experience: "Here is my application code. Google, run it and handle the servers for me!"

• What YOU do: You write your Python, Node.js, or Go code. You run 1 command (gcloud run deploy) to upload your app.
• What GCP does for you: GCP automatically installs the runtime, sets up Nginx/web server, patches OS vulnerabilities, scales up to 100 servers when traffic spikes, and scales down to 0 servers when traffic stops so you pay $0.
• GCP Product: Cloud Run, App Engine.
• When to choose it: When you want to focus purely on building your product without wasting time managing Linux servers.

Option 3: Function as a Service (FaaS / Serverless)

Developer Experience: "Run this single block of code only when an event happens, then delete the server."

• What YOU do: You write a small single-purpose function (e.g., "Whenever a user uploads a photo, resize it to a thumbnail").
• What GCP does for you: GCP listens for the photo upload event, instantly creates a micro-container, executes your 10 lines of code in milliseconds, and immediately destroys the container.
• GCP Product: Cloud Functions.
• When to choose it: For quick event-driven tasks (sending welcome emails, processing uploaded files, webhook notifications).

Option 4: Software as a Service (SaaS)

Developer Experience: "I don't want to code or manage anything. I just want to use a finished application."

• What YOU do: You log in via your browser, configure your settings, and use the software.
• What GCP does for you: Manages all code, servers, updates, storage, security, and scaling.
• GCP Product / Examples: Google Workspace (Gmail, Google Drive, Google Docs), Looker.

Cloud Deployment Models: Which One Should You Choose?

Where does your application actually live, and who shares the physical servers? Let me break down the 4 deployment models using real-world business scenarios:

1. Public Cloud (e.g., Google Cloud Platform)

🚀 Real-World Scenario: Spotify, Twitter (X), or a new startup launching an app tomorrow.

• How it Works: Google builds and manages massive data centers. Millions of companies rent virtual space on these physical servers over the internet.
• Is it Private?: Yes! Hypervisors and hardware-level encryption guarantee that your data is locked away inside your virtual project. Other companies sharing the same server rack can never see or touch your files.
• Why Businesses Love It: Zero server maintenance, instant scaling to millions of users, and pay-as-you-go billing ($0 upfront).

2. Private Cloud

🔒 Real-World Scenario: Central Banks, Military Networks, or High-Security Healthcare Systems.

• How it Works: A single organization builds or leases physical servers that are 100% dedicated to them. No external company ever shares or touches the hardware.
• Why Businesses Choose It: Strict legal regulations mandate that sensitive financial or medical records never leave physical, air-gapped company hardware.
• The Catch: Extremely expensive, requires buying servers upfront (CapEx), and your team must manually replace blown power supplies and failed hard drives.

3. Hybrid Cloud

🔌 Real-World Scenario: A 50-year-old Bank or Hospital modernizing its systems.

• How it Works: The company keeps its core customer database on legacy mainframe servers inside its office building, but connects it directly to GCP using a secure, private high-speed fiber link (Google Cloud Interconnect).
• The Best of Both Worlds: The bank keeps sensitive account records on-premise for compliance, while using GCP Cloud Run to power a lightning-fast mobile banking app for millions of smartphones.

4. Multi-Cloud

🌐 Real-World Scenario: Global E-Commerce Giants (like Target or Home Depot).

• How it Works: A company runs their primary application on GCP (leveraging BigQuery for analytics & Gemini for AI) while running other workloads on AWS or Microsoft Azure.
• Why Businesses Choose It:
  1. No Vendor Lock-In: Avoids total dependence on a single cloud company.
  2. Maximum Reliability: If one cloud provider suffers a rare global outage, the business switches traffic to the second cloud provider instantly.
  3. Best-of-Breed Services: Uses GCP for AI/Data Analytics and AWS/Azure for regional legacy tool compatibility.

What is Google Cloud Platform (GCP)?

When you build any application (a web app, mobile app, API, or AI bot), it cannot run in thin air. Every application requires 5 core physical building blocks:

1. 🧠 CPU & RAM: Compute power to run your code and calculate logic.
2. 💾 Storage & Databases: Hard drive space and databases to save user accounts, images, and files.
3. 🌐 Internet Bandwidth & Connectivity: High-speed networks so users worldwide can open your app.
4. 🛡️ Security & Protection: Firewalls, encryption chips, and access controls to block hackers.
5. ⚡ Auto-Scaling Systems: Extra capacity to keep your app from crashing when 100,000 users log in at once.

GCP is Google's Digital Supermarket

Instead of buying physical RAM sticks, hard drives, fiber network cables, and firewall boxes yourself, Google Cloud Platform (GCP) is Google's all-in-one digital supermarket that provides all 5 building blocks on demand.

Google built the world's most powerful infrastructure to run Google Search, YouTube, Gmail, and Google Maps. GCP opens up that exact same infrastructure for your application:

• Need CPU & RAM? Rent virtual servers with Compute Engine or run containerized apps with Cloud Run.
• Need Storage & Databases? Save files in Cloud Storage or databases in Cloud SQL / Firestore.
• Need Global Internet Speed? Route your users' traffic over Google's private subsea fiber network.
• Need Hardened Security? Benefit from Google's hardware-level Titan Security Chips and automatic DDoS protection.

Global Physical Infrastructure: Regions & Zones

Google operates data center buildings all around the world. To deploy your application on GCP, you only need to understand 2 simple terms:

1. Region (The Target City)

A Region is a major geographic city where Google operates cloud data centers.

• Examples: asia-south1 (Mumbai), us-central1 (Iowa, USA), europe-west1 (Belgium).
• Selection Rule: Always pick the region physically closest to your end users to keep your app fast (low latency).

2. Zone (The Physical Data Center Building)

A Zone is an actual physical data center building located inside that region city. Every GCP region has at least 3 separate zones, each built with independent power grids, cooling, and backup generators.

• Example in Mumbai (asia-south1):
  • asia-south1-a (Data Center in Navi Mumbai)
  • asia-south1-b (Data Center in Chandivali)
  • asia-south1-c (Data Center in Thane)

Signal vs. Noise: What to Remember vs. What to Ignore

Common Doubts & Interview Traps

Q1: How does GCP compare to AWS and Microsoft Azure?

• Answer: All three offer equivalent core IaaS and PaaS services (VMs, Databases, Networking). GCP stands out for its superior global private network speed, BigQuery data analytics, Kubernetes management (GKE), and Vertex AI / Gemini models.

Q2: If I deploy my application into a single zone, is it fully high-available?

• Answer: No! Single-zone deployment leaves you vulnerable to data center failure. High Availability (HA) requires deploying your app across at least two or more zones within a region behind a load balancer.

Q3: What is the difference between Public Internet routing and Google Global Network routing?

• Answer: Public Internet routes traffic through multiple third-party ISPs, causing unpredictable hops and latency. GCP uses Standard Tier (Public Internet) and Premium Tier (Google's private subsea network, entering at the nearest Edge PoP to the user).

Daily Practice Drill & Self-Check

Test your understanding of today's comprehensive lesson:

// Try answering these:
1. A developer wants to deploy a React + Node.js web app as quickly as possible without configuring Linux OS security patches or Nginx web servers. Which cloud service choice should they pick?
2. Why does GCP build 3 separate data center zones in Mumbai (asia-south1-a, asia-south1-b, asia-south1-c) instead of putting all servers in 1 big building?

🎉 Awesome job! You have completed Day 01.

You now have a complete, zero-gap mental model of cloud computing, CapEx vs. OpEx, service models (IaaS/PaaS/SaaS) from a developer's perspective, deployment models, and Google's physical infrastructure (Regions & Zones). Take a break, let today's concepts sink in, and come back tomorrow fresh!

Tomorrow on Day 02, we will log into the GCP website together, tour the Web Console interface, and learn how to navigate every key menu item with confidence.

← 00 - Learn GCP in 30 Days: Course Syllabus | Next Topic → 02 - Navigating the GCP Console Website