When properly implemented, server virtualization can yield a number of benefits. It can reduce hardware costs, improve availability for mission-critical workloads and allow you to scale workloads up or down as needed. These benefits, however, can be realized only if the virtualized environment has been created in a way that works with you rather than against you. When the implementation is handled poorly, virtualization pitfalls have the potential to degrade performance and, consequently, increase support costs.
Effective server virtualization management means avoiding missteps. Otherwise, you'll be left with an inefficient virtual environment. VM sprawl is a common trouble spot during virtualization efforts. Server virtualization makes it easy to deploy new virtual machines, especially when VM deployment is automated. In fact, the deployment process can become so easy that an organization may well find itself drowning in an excessive collection of VMs. VM sprawl not only complicates management, it can also lead to other virtualization pitfalls, such as wasted host resources and difficulty managing software licenses.
Because VM sprawl is so difficult to rectify, the best strategy is prevention. Consider requiring formal business justification and documentation for each new VM.
One tactic used in the battle against sprawl is to assign each VM an expiration date. In this process, a virtual machine nearing its expiration will be flagged, and a message sent to the appropriate person requiring his or her approval to keep the VM active.
Another common virtualization problem is resource contention. In a virtualized environment, VMs running on a particular host compete with one another for a finite pool of physical resources. The best way to fend off resource contention problems is to be realistic about the resources that each VM and the hypervisor itself will need so as to avoid overloading a host.
It's also a good idea to invest in the resources that VMs are most likely to deplete. Resource contention is usually most problematic with storage I/O and network bandwidth. As such, you might consider provisioning your hardware with extra physical storage and network connectivity.
Another potential issue is high availability for mission-critical workloads. If a host server fails, all of the VMs running on the host will also fail. Thus, your hypervisor deployment needs to be clustered so that, if necessary, VMs can fail over to another host server.
Keep in mind also that simply making VMs highly available isn't enough. If the hosts within a cluster are operating near capacity, there might not be sufficient resources to support a VM failover. For that reason, you'll want to avoid placing too much of a load on your host servers. Set some priorities so that your hypervisor knows which VMs are most important in the event of a failover.
Server virtualization has the potential to streamline IT operations and to make an administrator's life significantly easier. However, the virtualization infrastructure must be correctly aligned with the hypervisor vendor's best practices; otherwise, virtualization can introduce more problems than it solves.
Solutions to five of the most common virtualization problems
Prevent virtualization problems with workload balancing
Explore data virtualization tools in this essential guide
Dig Deeper on Preventing virtual machine sprawl
Related Q&A from Brien Posey
Staying current with the latest Windows updates is a difficult task, but IT can accomplish this without harming the UX with Group Policy, disk ... Continue Reading
IT professionals should deploy Group Policies to their browsers to improve the user experience. Check out this step-by-step guide on applying Group ... Continue Reading
The ability to share GPUs among multiple virtual desktop instances has contributed to a rise in GPU use in many VDI deployments. Continue Reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.