TU BCA Cloud Computing Master Guide: AWS, IaaS/PaaS/SaaS Models, Hypervisors & Docker Virtualization

Author: Bhuban Subedi | Subject: Cloud Computing (CACS402) | Semester: Seventh Semester


From serverless functions to multi-region kubernetes clusters, modern enterprise applications deploy almost exclusively to cloud infrastructure. In the Tribhuvan University BCA seventh semester, Cloud Computing (CACS402) equips students with the technical architecture of cloud resource provisioning, NIST service models, hypervisor virtualization, containerization with Docker, and cloud security frameworks.

In the final 60-mark TU board examination, examiners consistently formulate questions on the NIST 5 Essential Characteristics of Cloud, IaaS vs. PaaS vs. SaaS comparison, Type-1 vs. Type-2 Hypervisors, and Containerization vs. Virtual Machines.

In this guide, I will break down these cloud computing pillars with high-scoring comparative tables and architectural diagrams.


1. The NIST 5 Essential Characteristics of Cloud Computing

According to the National Institute of Standards and Technology (NIST), true cloud systems must satisfy 5 criteria:

                  +-----------------------------------------------+
                  |       NIST 5 CLOUD CHARACTERISTICS            |
                  +-----------------------------------------------+
                   ├── 1. On-Demand Self-Service (Provision without humans)
                   ├── 2. Broad Network Access (Standard HTTP/APIs)
                   ├── 3. Resource Pooling (Multi-tenant shared hardware)
                   ├── 4. Rapid Elasticity (Auto-scaling up and down)
                   └── 5. Measured Service (Pay-as-you-go metering)

2. Cloud Service Models: IaaS vs. PaaS vs. SaaS

+---------------+-----------------------------------+---------------------------------------+
| Model         | What You Manage                   | Provider Manages & Real Examples      |
+---------------+-----------------------------------+---------------------------------------+
| **IaaS**      | OS, Middleware, Runtime, Data,    | Virtualization, Servers, Storage,     |
| *(Infrastruct)*| Applications.                     | Networking (e.g., **AWS EC2, Azure VM**)|
+---------------+-----------------------------------+---------------------------------------+
| **PaaS**      | Application Code & Data only.     | OS, Runtime, DB Engine, Patching      |
| *(Platform)*  |                                   | (e.g., **AWS Elastic Beanstalk, Heroku**)|
+---------------+-----------------------------------+---------------------------------------+
| **SaaS**      | Nothing (You just consume the     | Entire Stack: Code, DB, Servers, UI   |
| *(Software)*  | software via browser or app).     | (e.g., **Google Workspace, Microsoft 365**)|
+---------------+-----------------------------------+---------------------------------------+

3. Virtualization: Hypervisors vs. Docker Containers

+---------------------------------------------------+---------------------------------------------------+
|               VIRTUAL MACHINES (VMs)              |              DOCKER CONTAINERS                    |
+---------------------------------------------------+---------------------------------------------------+
| * Heavyweight: Each VM runs a complete guest OS.  | * Lightweight: Shares host OS Linux Kernel.       |
| * Slow boot time (minutes).                       | * Instant boot time (milliseconds).               |
| * High memory and CPU overhead.                   | * High density; minimal resource overhead.        |
| * Managed by Hypervisor (ESXi, Hyper-V, KVM).     | * Managed by Docker Daemon / Containerd Engine.   |
+---------------------------------------------------+---------------------------------------------------+

Architecture Comparison:

  [ Virtual Machines ]                           [ Docker Containers ]
  ┌─────────────────────────────────┐            ┌─────────────────────────────────┐
  │ App 1   │ App 2   │ App 3       │            │ App 1   │ App 2   │ App 3       │
  ├─────────┼─────────┼─────────────┤            ├─────────┼─────────┼─────────────┤
  │ Guest OS│ Guest OS│ Guest OS    │            │ Libs    │ Libs    │ Libs        │
  ├─────────────────────────────────┤            ├─────────────────────────────────┤
  │ Hypervisor (Type-1 / Type-2)    │            │ Docker Container Engine         │
  ├─────────────────────────────────┤            ├─────────────────────────────────┤
  │ Host Hardware (CPU/RAM/Disk)    │            │ Host Operating System & Hardware│
  └─────────────────────────────────┘            └─────────────────────────────────┘

4. Cloud Deployment Models

+-------------------+-----------------------------------+---------------------------------------+
| Deployment Model  | Ownership & Accessibility         | Primary Use Case                      |
+-------------------+-----------------------------------+---------------------------------------+
| **Public Cloud**  | Owned by third-party CSP (AWS)    | Startups, consumer web portals,       |
|                   | over public internet.             | high-scalability e-commerce.          |
+-------------------+-----------------------------------+---------------------------------------+
| **Private Cloud** | Dedicated exclusively to one      | National banks, defense agencies,     |
|                   | enterprise on-premise or hosted.  | strict compliance requirements.       |
+-------------------+-----------------------------------+---------------------------------------+
| **Hybrid Cloud**  | Combines Private & Public clouds  | Sensitive data in private DB,         |
|                   | with encrypted interconnects.     | web traffic bursts on public cloud.   |
+-------------------+-----------------------------------+---------------------------------------+

Frequently Asked Questions (FAQ)

Q1: What is a Service Level Agreement (SLA) in Cloud Computing?

An SLA is a legally binding contract between a Cloud Service Provider and customer defining guaranteed performance metrics, uptime availability percentages (e.g., 99.99% “four nines”), and financial compensation credits for unplanned outages.

Q2: What is Serverless Computing (FaaS)?

Serverless computing (e.g., AWS Lambda) is a cloud execution model where developers execute code in response to events without provisioning, scaling, or managing any underlying virtual servers, billed strictly per execution duration in milliseconds.

LEAVE A REPLY

Please enter your comment!
Please enter your name here