Skip to content

Virtualization

So What is Virtualization?

Virtualization and Cluster Basics

1. Introduction of Virtualization

What is it?

Virtualization is the process of creating a virtual version of a physical resource such as a server, operating system, storage device, or network. It allows multiple virtual environments to run on the same physical machine.

Theoretical Definition

According to the Cloud Computing Black Book (2024), virtualization is the technology that abstracts computing resources from the underlying hardware and makes them available as software-defined instances.
For example, instead of running one operating system on one physical server, we can run multiple OS instances (called virtual machines) on the same hardware using a hypervisor.

Example

  • A single physical server in a data center can host multiple virtual machines, each running different operating systems (e.g., Windows, Linux).

WOW Tip

Virtualization is one of the key technologies that made cloud computing possible, because it enabled resource pooling and efficient hardware utilization.

2. Virtualization Types: Type 1 and Type 2

What is it?

There are two main types of virtualization based on how the hypervisor interacts with hardware.

Theoretical Definition

  • Type 1 Hypervisor (Bare-metal): Runs directly on the hardware without needing an operating system. It provides better performance and efficiency.
  • Type 2 Hypervisor (Hosted): Runs on top of a host operating system, making it easier to set up but with slightly less performance.

Example

  • Type 1: VMware ESXi, Microsoft Hyper-V, Citrix XenServer
  • Type 2: Oracle VirtualBox, VMware Workstation

WOW Tip

Type 1 hypervisors are commonly used in enterprise data centers where performance and reliability are critical.

5. Cluster Architecture

What is it?

A cluster is a group of interconnected computers (or nodes) that work together as a single system to provide high availability, performance, and scalability.

Theoretical Definition

Cluster architecture distributes workloads across multiple nodes. If one node fails, others continue to provide services, ensuring reliability.

Example

  • A web server cluster can have multiple servers hosting the same website. If one fails, traffic is redirected to another server.
  • Hadoop clusters are used for big data processing.

WOW Tip

Google built its early search engine on clusters of commodity servers, proving that clusters can outperform supercomputers at scale.

6. Cluster Requirements

What is it?

For a cluster to work effectively, certain hardware, software, and network requirements must be met.

Theoretical Definition

A cluster requires:
- Nodes (servers with adequate CPU, memory, and storage)
- Networking (high-speed interconnects like Gigabit or InfiniBand)
- Shared Storage (SAN, NAS, or distributed file systems)
- Cluster Management Software (e.g., Kubernetes, Hadoop YARN, or Windows Failover Clustering)

Example

  • A database cluster may need redundant storage and a reliable network to avoid single points of failure.