What is the Meaning of the term High Availability in a Virtualized Infrastructure?
You may have seen the term HA or High Availability used often around virtualized infrastructures. High Availability refers to having redundancy or several levels of redundancy in an infrastructure. Today I will discuss the meaning of HA in a virtualized infrastructure.
Some of the well known virtualized infrastructures are built with hypervisors like, Citrix Xen Server, Vmware, Hyper-V and there are several others. If HA is setup in a virtualized infrastructure it means that the virtual servers are configured on a system that has built in redundancy. So you might ask, how does High Availability work? Does it work with linux virtual servers as well as windows virtual servers? Yes it does. Well first off we will explain how server virtualization works.
A virtualized server environment is built with several servers that have lots of memory and lots of CPU cores and a SAN (storage area network) device. The servers are known as the hosts and the virtual servers or machines that run on them are known as guests. Virtual Servers are referred to as VMs or guests. So for instance, we have a virtualized infrastructure set up with 3 servers, each server has 8 processors and have 192 gigs of RAM each. The SAN has a capacity of 12 TB (tera bytes). All of these devices are interconnected to create a virtualized infrastructure. Once they are connected the hosts (physical servers) provide the memory and CPU to each of the guests (virtual servers) and the SAN provides the disks for each VM (virtual server or guest). So you can connect to the virtualized infrastructure from a management console and provision a virtual server with as much cpu, memory and disk space as required. Then from there you can select which operating system to install on the virtual server. This is basically how the virtualized infrastructure is set up and works. Once the virtualized infrastructure is set up you can now enable HA (high availability).
So to configure the virtualized environment in HA or make it highly available you will have to enable this feature within the hyper visor you are currently using. To enable HA it is different in each hypervisor but usually is not very difficult, you can find this information in the administrators manual. Ok so once HA is turned on and configured how does it work and what will it do? Well the main purpose for HA is to keep your linux virtual servers or windows virtual servers running even if one of the main servers (hosts) in the infrastructure was to fail, lock-up or die. So to run you through an example, if you have virtual servers running on a host (server) and the physical server stops functioning, all the virtual servers in the HA cluster will then automatically migrate to another working host(server) in the virtualized infrastructure. This is a really nice feature and helps administrators and clients sleep a little better at night. HA will keep your virtual servers running in case of failures and also adds to the level of redundancy of your running systems.
You can send us an email at email@example.com if you have any questions about this article.