Benefits of Server Virtualization
The 5 major benefits of server virtualization are:
Improved host utilization:
Legacy Physical hosts are most often provisioned with extra capacity since it is difficult to upgrade the host in service. Typically as much as 50% or more disk space, CPU, and/or memory is wasted with over-provisioning. Since CPU, memory and disk can most often be increased 'on-the-fly' on virtual servers without even rebooting the host, they can be set to the actual requirements leading to 2X or more improved utilization.
Of course this also means a drastic reduction in number of hosts and associated power and cooling costs. For example: a client with 50 to 75 hosts might consolidate down to only 4 hypervisor hosts and two shared storage units.
A virtual cluster continuously monitors for failures of any host (hypervisor). In the event that a hypervisor were to fail, all the servers on the failed host automatically restart on a spare host within a few seconds without operator intervention. You then have ample time to shutdown the failed hypervisor, replace components and return it to service. ( All without affecting any of the running virtual servers)
Expanded Maintenance windows
Since virtualized server workloads can be moved to other hypervisors while running, maintenance work on on a physical hypervisor host can be accomplished during normal working hours by simply moving workloads temporarily to another hypervisor.
Physical servers contain OS specific hardware drivers for that host. This means that bare-metal restores of the host need identical hardware ( down the the exact video card and disk ), or complicated restore processes. With a virtual environment ALL hardware drivers are identical. Not only can the virtual server be restored anywhere, but it can be moved while running.
Because many server 'blue-screens' can be attributed to driver issues, having a single set of 'virtual' drivers across all your servers can actually make the OS more reliable by eliminating those bad drivers.
Snapshots, Clones and Templates
Since a virtual server is just a few files inside a directory on the hypervisor, it is very easy to backup and duplicate the virtual server
A snapshot captures entire state of the VM in seconds. The space used starts small and grows as change are made.
A clone captures the entire state of the VM in minutes. Typically requires as much space as original VM, but does not grow over time.
A template is a pre-configured OS, with patches installed, typically used a a source for cloning new virtual machines.
These benefits are are real and are used every day by companies that have migrated to a virtual solution. We have a full VMware demonstration lab can demonstrate all of these features.
We generally recommend and support VMware the industry leading Virtualization platform.
Our experience ranges from installing small sites with only two hypervisors to supporting clients with up to 200 hypervisors and thousands of virtual machines
For an overview or demo Contact us today
We can help with all of the following:
- Health Checkup
Proper operation of VMware is very dependent on the health of servers and underlying storage and network. We check for improper configurations, use of best practices, and look for any evidence of performance issues.
- Architecture and Planning
We help you plan an affordable and expandable VMware environment according to your schedule. You could start with a small environment for internal development, or proceed directly to production. Production use generally requires investment in supporting infrastructure such as storage, networking, and backup/DR.
Upgrades to VMware can generally be done with little disruption to running virtualized servers. We have experience with upgrades from vSphere 3.5 to vSphere 4.0/4.1 to vSphere 5 in production environments. We have done upgrades in environments as small as 4 hypevisor hosts in-place with little disruption to production systems. If possible, upgrades to new hardware (new cluster) is somewhat easier and faster.
- Ongoing support
We can offer on-going support on a monthly or on-call basis. We typically review your system's configuration and performance on cause, or at least a monthly basis and provide input into your monthly maintenance planning schedule.