A hypervisor is software that virtualizes and allocates a server’s computer processing unit (CPU), memory and network resources among multiple virtual machines (VMs).
This allows multiple, logically isolated instances of applications like operating systems (OSs) to run independently even though they rely on physically integrated hardware.
Hypervisor Definition
A hypervisor is software that allows many virtual machines to run on a single physical machine. It accomplishes this by allocating a computer’s hardware components to support multiple VMs, which possess their own operating system and applications.
Hypervisors are foundational to virtualization, enabling modern cloud computing models like Infrastructure-as-a-Service (IaaS). It allows providers such as AWS, Google Cloud and IBM to deliver scalable, partitioned computing resources across shared physical infrastructure.
What Is a Hypervisor?
A hypervisor is a program that takes a computer’s physical resources and allocates them accordingly to form multiple virtual machines. In this setup, each VM has an independent operating system and applications while sharing a computer’s memory, CPU and other hardware components. This makes it possible to run many VMs on a single physical machine.
How Does a Hypervisor Work?
Imagine you have a high-performance server with multiple CPUs, large memory capacity, and ample network bandwidth. These components make up your integrated physical hardware.
Say four different people want to use your physical hardware’s computing resources. Instead of allowing each person to use your hardware one at a time, you install a hypervisor, which allows you to orchestrate and manage the division of the hardware resources into four different VMs.
At the same time, you maintain control over the infrastructure and server security. VMs are logically isolated, meaning each runs independently with no direct access to the memory or data of other VMs. This means it’s typically not possible for one user to access data from the other virtual servers or host operating system.
Types of Hypervisors
There are two main types of hypervisors — type 1 and type 2 hypervisors — that deliver different levels of performance.
Type 1 Hypervisor
Type 1 hypervisors act as a lightweight operating system running on the server itself. We often refer to type 1 hypervisors as bare-metal hypervisors. Type 1 hypervisors run directly on hardware and do not require a separate general-purpose operating system like Windows or Linux, resulting in lower overhead.
Type 2 Hypervisor
Type 2 hypervisors are software that runs on an operating system, like many other programs. Since the requests for hosted resources must pass through an extra software layer compared to type 1 hypervisors, type 2 hypervisors tend to be less performant.
If you’re looking to experiment with type 2 hypervisors, KVM is a popular hypervisor that runs on the Ubuntu distribution of the Linux operating system and VirtualBox is a popular hypervisor for Windows.
Containers vs. Hypervisors
A container is a software package that stores all the components needed to run an application on any operating system.
Both containers and hypervisors allow computing resources to be used more efficiently, but they support virtualization at different levels:
- Hypervisors abstract hardware resources so multiple operating systems can run independently on one machine, while containers abstract the operating system to let applications run in isolated environments.
- Hypervisors share a computer’s CPU, memory and other resources; containers rely on a container engine to run.
- Hypervisors can support many OSs; containers share the host OS kernel, which means all containers on a host use the same underlying operating system (but can have isolated user-space environments).
Containers are better suited for tasks that require limited resources while hypervisors can handle more demanding projects.
Benefits of Using a Hypervisor
Ultimately, hypervisors reduce energy use, overhead costs and overall complexity in the cloud computing industry. Here are a few more reasons why hypervisors have become such a pivotal technology:
Resource Management
Being able to split a host machine into many VMs can be a cost-effective solution to using powerful hardware. Hypervisors can make it much easier to run simultaneous tasks by splitting them across many VMs instead of processing directly on the host, which potentially leaves resources unused.
Operating System (OS) Flexibility
Some software requires a specific operating system. It would be expensive and inefficient to run both a Windows machine and Linux machine, for example. By creating multiple VMs, the hypervisor allows multiple operating systems to run on the single host machine. This flexibility makes it possible to run applications that depend on a particular OS all within the host machine, thereby improving efficiency.
Scalability
Since the hypervisor creates VMs that are independent of the physical hardware, it’s possible to transfer hypervisors across servers to reallocate resources where needed. Hypervisors allow vertical scaling by reallocating host resources to VMs. Horizontal scaling typically involves orchestration tools that manage multiple hypervisor hosts.
Risks of Using a Hypervisor
Hypervisors are software, and software can have bugs and vulnerabilities cybercriminals can exploit. Here are some of the risks to consider when using hypervisors:
Software Vulnerabilities
Since hypervisors allow a system administrator to create and monitor VMs on their hardware, the entire system could be at risk if an attacker compromises the hypervisor. Threat actors do this by exploiting outdated hypervisor software or by infecting a machine with malware. When an attacker gains control of a hypervisor that manages many VMs, they can use those VMs for malicious activities like denial-of-service (Dos) or distributed denial-of-service (DDoS) attacks in addition to having privileged access to the machines themselves. As a best practice, update hypervisor software regularly to patch found vulnerabilities.
Network Vulnerabilities
Without the proper network security in place, the VMs themselves might be vulnerable to DoS or DDoS attacks because the hypervisor shares network bandwidth across the VMs. Additionally, the VMs are typically distributed on a network. Cloud companies often keep their VM networks separate from their management networks to help prevent a compromised VM from impacting day-to-day network usage.
VM Sprawl
It is imperative that a system administrator manages their VMs appropriately and doesn’t leave them unchecked, thereby taking up system resources after work on them has been completed. Old VMs might be running outdated operating systems or applications that are easier to exploit. Also, there could be sensitive files left on VMs that could damage a company or user if compromised. As a best practice, VMs should be shut down and removed after serving their purpose instead of being left online.
Frequently Asked Questions
What does a hypervisor do?
A hypervisor allocates a computer’s physical resources, such as CPU and RAM, across multiple virtual machines so they can run independently on a single server. Hypervisors allow multiple VMs to share hardware, reducing underutilization and enabling cost-effective, flexible resource allocation.
What are the differences between type 1 and type 2 hypervisors?
Type 1 hypervisors run directly on the hardware (bare-metal), offering better performance. Type 2 hypervisors run on top of an existing operating system, which can make them less efficient.
What is an example of a hypervisor?
Oracle VM VirtualBox is an example of a hypervisor. With this software, teams can run multiple operating systems — including Linux, Mac OS X and Microsoft Windows — on the same computer simultaneously.
