Why Preventing Virtualization Sprawl Matters

Virtualization is not only for early-adopters or IT geeks anymore. The technology has found its place in almost every IT department and data center, but the advantages of the technology - fast and almost immediate server deployment, increased server uptime, and isolated applications - also has its disadvantages. One of them being virtualization sprawl (definition found here). Since creating and running a virtual machine (VM) or virtual servers is very easy and quickly achievable, users tend to over-strain their IT department with lots of VMs. At a certain point, the number of available virtual machines and virtual servers is so huge that the IT administrator cannot manage them effectively.

Users create lots of VMs over time – sometimes only for a single project – and forget about them afterwards. In many cases, there are virtual machines available with no use that still take up disk space. What often goes overlooked is that creating a virtual machine, or several VMs, in a virtual server has the same impact as if one would create a physical server. The same licensing, support, security and compliance issues must be in consideration.

Furthermore, too many virtual machines pose several technical challenges and risks to the IT infrastructure and its IT administrators.

Lack of technical control

• Users aren’t able to create or even delete virtual machines. One downside is disassociation of the VMs. This can happen when accidentally deleting or removing a VM from the hypervisor inventory.
• When removing VMs from one cluster and moving them to another host cluster, the risk is high that they will experience rearranging, especially in larger environments.
• Virtual machine fragmentation can occur. When VMs experience frequent movement and copying fails, large fragments will remain on the disk.
• Creating huge amounts of virtual snapshots will result in significant storage space use.
• Not having a complete overview of all VMs in the company can result in problems when determining which one is erasable or– in case of data loss – which one is critical to the production environment and requires recovery.

Lack of management – Additional tasks for IT administrators

• More work for VM administrators: Managing more VMs with physical resource constraints.
• More work for storage administrators: Provision more disk and storage space.

Higher costs due to unnecessary VMs

• Creating and running VMs is not free: They consume CPU time, memory, and disk space.
• Licensing costs for OS, backup agents, and databases.
• Purchasing additional storage space.

Reduced performance

• In some severe cases, too many VMs can have a crippling effect on the physical server performance. When there are too many VMs running side-by-side, the overall speed slows down. When the physical server resources become too low to handle the work load, the VMs can stop working or fail altogether.
• In rare instances, data loss can be the result that you don't want. Since virtualization creates another layer to the data structure, data recovery is somewhat more difficult as if one were recovering data from a simple Windows 10/NFTS formatted hard disk drive.

How to prevent virtualization sprawl

• Audit VMs. It may seem like a simple solution, but make it a policy that every VM and virtual server must be documented and registered.
• Optimize storage and implement data policies. To prevent the usage of unnecessary disk and storage space, use technologies like snapshots and thin provisioning. But beware: When you use these technologies and implement them into policy, make sure that they are setup correctly, otherwise, you not only risk losing data or failure, but you also risk an extremely challenging data recovery process. And in some case data recovery will not be able at all.
• Implement lifecycle management tools. To keep track of all virtual machines, as well as virtual servers, it’s a good idea to use a lifecycle management tool. With such a tool, it’s possible to provide virtual machines only for the users of one specific project or track all objects within a centralized database. When a project is finished, it’s far easier to identify unneeded or unused VMs for secure erasure.
• Implement VM archiving. Many users create virtual machines just for one project, use it a couple of times, and then leave it untouched for months. Instead of keeping these VMs inside the production environment, they should be archived. Many backup solutions provide the possibility to archive unused VMs on cheaper storage or tape.

Delete non-essential VMs whenever possible

An administrator should frequently review VMs and determine if they need to be permanently deleted, not just from a storage and cost perspective, as state before, but also because of data security reasons. In cases of a data breach, chances are high that restricted and business critical data could get into the hands of criminals. The consequences can be severe: Lack of trust by partners or customers, fines for not complying with data security laws, or becoming a victim of criminal activities such as blackmailing. To prevent this from happening, a valid data erasure plan should be implemented inside the data security and business continuity plan.

For the purpose of securely destroying virtual data there are several professional tools available that ensure the specific virtual machine(s) are securely erased, while the live environment is left intact and running. Such a solution is the Virtual Machine Eraser from Blancco, Ontrack´s partner in secure data erasure. It allows for a secure and automatic destruction of virtual machines and supports hypervisors across VMware ESXi, Microsoft Hyper-V, Oracle Virtual Box and virtual hard disk formatted for VMDK, VHDX, VDI and OVF.

What to do when struck by data loss due to virtualization sprawl

If data loss occurs, it’s important not to panic and to initiate the correct steps to avoid further damage. First, it’s absolutely necessary to stop working on the affected virtual machine as well as the physical server hosting the VMs or virtual server. Secondly, try to terminate the service normally. Do not shut down the server via brute force because it could cause further harm to the data. Lastly, contact a data recovery expert immediately. More information about how Ontrack can help individuals and businesses alike regarding virtualized data recovery.