Hypervisors play an essential role in enabling server virtualization, which is itself essential to enabling cloud computing. Broadly speaking, virtualization refers to the use of software to simulate or emulate physical resources. In the case of server virtualization, a hypervisor is a software process that creates and runs virtual machines (VMs) using the resources of physical hardware. The hypervisor abstracts and isolates the VMs and their programs from the underlying server hardware, enabling a more efficient use of physical resources, simpler maintenance and operations, and reduced costs.
What problems do hypervisors fix?
The primary technological problem that hypervisors solved was that most physical hardware could run only one operating system at a time. This constraint often led to wasted resources, as a single OS seldom fully utilized the hardware’s capacity.
Hypervisors address the above constraint by aggregating the resources of virtualized physical servers (such as memory, network bandwidth and CPU cycles) and then allocating those resources to virtual environments, called virtual machines. Hypervisors are also known as virtual machine monitors (VMM). A VM is essentially a software-based computer, with access to the same resources as a physical computer, including an OS and apps. However, a hypervisor lets you run multiple VMs as guests, thereby using the physical resources of the underlying host machine much more efficiently. Each VM can act as a dedicated machine for every service, app or operating system, allowing you, for example, to run multiple different OSs on a single server. The hypervisor also separates the VMs logically, which protects each individual VM against the effects of problems with other individual VMs on the same hypervisor, such as crashing, errors, or security attacks.
How hypervisors enable the benefits of virtualization
As software, hypervisors decouple the OS and apps from the physical host. This decoupling provides an array of benefits, including the ability to easily and quickly migrate the VM from one host to another without disruption. This capacity, called live migration, is essential for workload balancing. Live migration also occurs automatically in the case of node failure, providing high availability and increased uptime.
Virtualization enables cost savings through reducing physical footprint, which in turn reduces costs for electricity, cooling, and maintenance. Virtualization also greatly improves agility and speed in delivering IT services. For example, it is far easier to spin up a VM than to provision new environments to satisfy customer requests.
Type 1
Type 1 hypervisors are installed directly on the physical server, which is why they are also called “bare metal” hypervisors. Direct access to the resources of the physical server makes Type 1 hypervisors highly efficient. This design also makes Type 1 hypervisors more secure, as it limits the attack surface and potential for compromise. Type 1 hypervisors are by far the most common choice within enterprise IT contexts, primarily due to their strong security, scalability, stability, and performance. Examples of the most widely used hypervisors include Nutanix AHV, VMware ESXi, Microsoft Hyper-V, and Citrix Hypervisor.
Type 2
Type 2 hypervisors differ in that they run as applications on a physical server’s preexisting OS. Because they run on the host OS, which sits between the physical server and the hypervisor, they are also known as “hosted” hypervisors. Type 2 hypervisors are not ideal for server-based environments, given that they have a higher latency and risk exposure than Type 1. They are, however, relatively easy to install, and can work well in specific use cases, such as individual PC users who need to run more than one operating system, and where performance and security are not principle concerns.
Complexity
Is it easy to deploy and manage? Is it a separate product, with a separate console, that requires full-time specialists to maintain, operate, and troubleshoot? Is it something that an IT generalist could master relatively quickly?
Performance
Does it deliver enough performance to support your mission-critical applications? Check out the benchmarks for performance in production (as close to real-world conditions as possible).
Cost
Does it come with licensing fees, or is it built-in to the larger solution?
Ecosystem
Does it support a rich ecosystem? For example, does it support the most widely used guest operating systems? Microsoft, Suse, RedHat, Ubuntu, CentOS. Does it support leading enterprise apps and technologies? Microsoft SQL Server, Exchange, and Sharepoint; SAP, Oracle, Citrix, Hadoop, Splunk, MongoDB, SaS, OpenStack, Avaya, Docker
“With all of the management savings we’ve obtained by moving to the Nutanix Enterprise Cloud and AHV, we can now focus on supporting our Citrix end users, rather than spending all of our time on the underlying infrastructure.” Paul Farrell, Black Hills Energy
“AHV was a major selling point. We considered alternative virtualisation platforms, but as well as the licensing costs – always an issue for publicly funded organisations like us – we wanted a virtualisation platform that could take full advantage of other core infrastructure services, as well as one that didn’t require yet another management interface or the skillset to go with it. Nutanix AHV has given us all that, and more.” Amardeep Rai, SMG
“It took us less than a week to become comfortable with AHV. AHV is a simple solution with everything we need displayed in one console, and we love the one-click non-disruptive upgrades. We also don’t miss the nighttime upgrades or the inefficient silos of a separate NAS.” Armando Muniz, JetBlue
Are You Still Paying for Virtualization?
AHV Best Practices Guide
Citrix XenDesktop on Nutanix AHV
Acropolis Datasheet
