Containerization_and_DevOpsLab

Lab Report: Comparison of Virtual Machines (VMs) and Containers

---

1. Objective

The primary goals of this experiment are:

• To understand the conceptual and practical differences between Virtual Machines (VMs) and Containers.
• To install and configure an Ubuntu-based Nginx web server using both VirtualBox/Vagrant and Docker inside WSL.
• To compare resource utilization, performance, and operational characteristics of both environments.

---

2. Hardware and Software Requirements

Hardware

• 64-bit system with virtualization support enabled in BIOS.
• Minimum 8 GB RAM (4 GB acceptable).
• Internet connection.

Software

• Oracle VirtualBox and Vagrant.
• Windows Subsystem for Linux (WSL 2) with Ubuntu distribution.
• Docker Engine (docker.io).

---

3. Theory

| Feature | Virtual Machine (VM) | Container | | :— | :— | :— | | Virtualization Level | Emulates complete hardware and kernel. | Virtualizes at the OS level, sharing the host kernel. | | Isolation | Strong (Full OS isolation). | Moderate (Process-level isolation). | | Resource Usage | Higher (Requires dedicated RAM/CPU for guest OS). | Lightweight and efficient. | | Startup Time | Slower (Minutes/Seconds). | Fast (Milliseconds/Seconds). |

---

4. Experiment Setup - Part A: Virtual Machine

Step 1: Initialization and Deployment

Using Vagrant, an Ubuntu VM was initialized and started.

• Command: vagrant init ubuntu/jammy64 followed by vagrant up.

Observation: The system downloads the base box (Ubuntu Jammy) and configures the VirtualBox provider. Port forwarding (2222 -> 22) is established.

Step 2: Accessing the VM (SSH)

Once the VM was up, we established a connection to the guest OS.

• Command: vagrant ssh

Observation: Successful login to the Ubuntu 22.04.5 LTS environment.

Step 3: Installing Nginx

Inside the VM terminal, the package lists were updated, and the Nginx web server was installed.

• Commands: sudo apt update, sudo apt install -y nginx

Observation: The apt package manager retrieves necessary archives. This process is slower than Docker as it installs dependencies for a full OS environment.

Observation: Installation is complete. Triggers for man-db and ufw are processed.

Step 4: Verification Inside VM

We verified the server was running locally within the guest OS.

• Command: curl localhost

Observation: The curl command inside the VM returns the full HTML source of the “Welcome to nginx!” page.

---

5. Experiment Setup - Part B: Containers (Docker)

Step 1: Running the Container

The Docker engine was used to pull the Ubuntu image and deploy a containerized Nginx instance.

• Command: docker run -dp 8080:80 --name nginx-container nginx

Observation: Docker pulls the image layers and starts the container nearly instantaneously.

Step 2: Verification

The Nginx server was verified by accessing the mapped port on the localhost.

• Command: curl localhost:8080

Observation: The curl command confirms the Nginx “Welcome” page is active on port 8080.

---

6. Resource Utilization & Comparison

This section uses specific metrics captured during the experiment to contrast the two technologies.

A. Boot Time Analysis

• Metric: Time taken to reach a usable state.
• VM Command: systemd-analyze

Observation (VM): The VM took 36.819 seconds to finish startup (6.9s kernel + 29.8s userspace). Observation (Container): The container started in less than 1 second (refer to Docker output in Sec 5).

B. Process Overhead & Isolation

• Metric: Number of background processes required to run the application.
• VM Command: htop

Observation (VM): htop reveals a heavy process tree. Even though we only want Nginx, the VM is running systemd, snapd, rsyslogd, polkitd, and sshd. There are dozens of tasks running to support the OS.

Observation (Container): docker stats shows the container uses minimal resources because it only runs the application process (Nginx) and its direct dependencies.

C. Memory Usage

• Metric: RAM consumption.

Observation (VM): The free -h command inside the VM shows it has allocated 957Mi total, with 196Mi used immediately by the OS kernel and services.

Observation (Container): Referring to the Docker Stats image above, the container consumes only 13.22MiB of RAM.

Comparison Summary Table

Parameter	Virtual Machine (VM)	Container (Docker)
Boot Time	~36.8 seconds (Measured)	< 1 second
RAM Usage	~196 MiB (Base OS overhead)	~13.22 MiB (App only)
Background Processes	High (init, systemd, ssh, logs)	Low (only Nginx entrypoint)
Disk Usage	Larger (Full OS libraries)	Smaller (Layered Image)

---

7. Conclusion

The experiment validates that Containers are significantly more lightweight.

1. Speed: The Docker container started instantly, whereas the VM required nearly 37 seconds to boot.
2. Efficiency: The VM required ~200MB of RAM just to exist, while the containerized application ran comfortably on ~13MB.
3. Complexity: The htop analysis clearly shows that VMs run a full operating system stack, creating unnecessary overhead for single-application deployments.

Therefore, Containers are ideal for microservices and rapid scaling, while VMs are better suited for scenarios requiring full hardware simulation or complete OS isolation.

---