Möchten Sie diesen Beitrag in Deutsch zu lesen? Lesen Sie die Deutsch-Version hier.
The benefits of server virtualization are many. Today, companies virtualize large portions of their computing environment from file shares, to databases to web servers. Just a few years ago, it wasn’t hard to find articles advising against virtualization for certain server tasks that require extreme performance or host mission critical applications.
But with increased CPU and memory performance, the number of tasks you shouldn’t virtualize continues to shrink. As more of your company’s critical infrastructure moves to a virtual machine (VM), it’s wise to be aware of the increased complexity that comes with backing up and recovering within that environment.
Your IT staff may be entirely comfortable managing updates, patches and backups for your on-premise, physical servers. The same can be said for managing VMs in house, but more companies are moving to a hosted VM or hybrid solution, blurring the lines between which party is responsible for backup and recovery tasks. Never assume someone else has it covered.
This week I’d like to take a look at how virtualization can enhance your backup and recovery plans.
Optimization With Virtualization
When I’m talking about optimization in this context, I’m referring to the physical server. You don’t want the server to pause applications during the backup process. This can happen for a number of reasons. Sometimes your server’s hardware isn’t up the task in terms of processor performance. Or you may need to add more memory. It can also happen is the backup tool you’re using isn’t virtualization-aware. Proper optimization of your server will allow backup to continue while users access the applications running on that server.
Shared storage is also important when optimizing your server. The storage is often shared across a Fibre channel or SAN. This is critical because it allows fast recovery to almost any other connected server. The downside? You’ll need to allocate significant storage capacity to hold each VM, but that cost will be offset by the high availability of those applications.
It’s critical that you’re using the right software for the task. Too many companies assume the backup software they’ve been using for years will work in a virtualized environment, and that’s usually not the case. The right software will be able to “see” into the complexity of a multi-VM physical server. The software should also be able to perform backups from the images and not only the physical server itself. If you’re ensure about the licensing or features of your software, please consult with an MSP who can assess your needs and recommend the right product.
In short, use virtualization-aware backup tools, make sure VMs are stored on a SAN and give your server plenty of storage and memory. It’s wise to invest resources upfront to optimize your server before it goes into production. You don’t want to be making hardware changes once you’ve moved the server into production.
Recovery in a virtualized environment provides a number of substantial advantages. Because you’re able to decouple the applications and OS from the physical server, you can easily move them to new hardware if needed. This model helps facilitate a high availability and disaster recovery for your company’s most mission critical applications. Some companies choose to setup a continuous replication of their environment. If you go the continuous backup route, take note of the server’s performance because this choice can potentially utilize significantly more system resources. Of course, you can also perform a more traditional, interval backup depending how quickly need to recover.
These are the most popular backup and recovery options for virtualized environments:
Full backup: Every VM is backed up as if it were on a physical server. This method can increase I/O load.
Dedupe backup: Data deduplication is performed on the VMs with only the changes being backed up. This decreases I/O load, but increases recovery time.
Snapshot backup: Periodic VM snapshots are transferred to a proxy server for offsite backup. This requires more IT involvement to manage the proxy server host.
The benefits of running VMs are so great that it’s hard to find a company that’s not virtualizing at least part of their computing environment. Using VMs can help to:
- lower hardware and licensing costs;
- save on power and cooling costs;
- simplify your infrastructure and staffing needs.
It can also streamline your backup and recovery needs, but don’t assume that’s going to happen by accident. Optimizing your company’s backup and recovery takes a lot of planning. Servers that ran a process or two just fine may buckle under the load of a running VMs. Assess your hardware needs before you layer on the backup and recovery software.
You’ll be glad you took the time to optimize your environment before you’re in a situation where you’re attempting to recover a VM running a mission-critical application. If you’re uncertain where you stand today in terms of backup and recovery, now would be the time to speak with an MSP.