Virtualization

Life Before Virtualization

In traditional data centers:

• One application typically ran on one physical server
• Hardware were statically allocated
• Servers were heavily underutilized

Typical utilizations:

• CPU usage often below 10-15%
• Memory mostly idle

Despite this, hardware costs were high, and scaling required buying new machines.

The core problem was tight coupling between applications and hardware.

What Is Virtualization?

Virtualization allows multiple isolated operating systems to run on a single physical machine.

Each virtual machine (VM):

• Has its own OS
• Has allocated CPU, memory, and disk
• Believes it is running on real hardware.

At the center of this is the hypervisor.

The Hypervisor: The Key Enabler

A hypervisor is a thin layer of software that:

• Sits between hardware and operating systems
• Manages CPU, memory, storage, and networking
• Provides isolation between VMs

Types of Hypervisors

Type 1 (Bare-Metal Hypervisor)

• Runs directly on hardware
• Examples: VMware ESXi, Xen, Hyper-V
• High performance and strong isolation

Type 2 (Hosted Hypervisor)

• Runs on top of a host OS
• Examples: VirtualBox, VMware Workstation
• Common for development and testing

Cloud providers use Type 1 hypervisors.

Why Virtualization Was Revolutionary

Virtualization introduced several game-changing capabilities:

1 Hardware Consolidation

• Multiple VMs on one server
• Better CPU and memory utilization
• Lower hardware costs

2 Faster Provisioning

• VMs could be created in minutes
• No need to procure new hardware

3 Isolation

• Failures in one VM didn't affect others
• Security boundaries at OS level

4 Snapshot and Migration

• VM snapshots for backup
• Live migration between physical hosts

These features fundamentally changed data center operations.

The Limitations of Virtual Machines

Despite its benefits, virtualization introduced new problems.

1 Heavyweight Abstraction

Each VM includes:

• Full operating system
• System libraries
• Kernel

Result:

• Large images (GBs)
• Slow boot times (minutes)

2 Operational Complexity

• OS patching per VM
• Configuration drift
• Manual scaling

Infrastructure improved, but deployment was still manual.

3 Environment Inconsistency

VMs reduced hardware issue, but:

• Dev, staging, prod VMs still differed
• OS-level differences persisted

“It works on my machine” still applied.

Virtualization and the Dev-Ops Divide

Virtual machines strengthened the separation between:

• Developers (application code)
• Operations (VMs and OS management)

Developers shipped binaries. Ops managed machines.

This slowed down deployment and innovation.

Why Virtualization Was Not Enough

Virtualization solved:

• Hardware inefficiency
• Server provisioning speed

But it did not solve:

• Application portability
• Environment consistency
• Deployment automation

The industry needed a lighter, faster abstraction.