This post also appeared in The Tennessean, where Concept Technology has a bi-weekly feature in the Business section.
When it comes to data protection, most organizations still rely on a traditional file-based backup model for their servers. It’s a model that has been around for decades and works well for many environments. More recently, though, we’ve seen some businesses evolving towards versioned replication — a better backup option, in my opinion.
In any server backup model, you start by creating an identical copy of your server. With versioned replication, as new layers of data get added to your server, identical copies of these layers are also added to your backup server so that at any given time, your primary server and backup server are in sync with each other. This backup “server” is actually just a bunch of files that can be booted as a “virtual” server when needed.
If something happens to your primary server and you decide that you need to revert back to what the server looked like three days ago, versioned replication makes it possible for you to easily access your backup, strip multiple layers away and continue working with little to no downtime.
If you need to completely restore your server after a disaster, with versioned replication you can boot up the backup server, which will stand up and run on its own.
This compares to a traditional file-based backup model — which still gives you a good backup — but if you ever need to restore your server you have to reinstall your operating system and your backup software, all before you can even start the restoration process.
One of the primary benefits of versioned replication is the time it saves if your company has to recover a server after a disaster. Versioned backups can bring a server back online in a matter of minutes (and usually under 30 minutes); a file-based system can take hours or days, depending on how much data your organization stores.
Here are two additional benefits to versioned replication:
- Test restorations. It’s much easier to perform test restorations of your data when you use versioned replication. Test restorations are extremely important, and your system administrators should do them as part of their routine maintenance of your company’s servers.
- With versioned replication, your system administrators can restore an entire server in a test environment to make sure that backups are actually working, giving you the peace of mind that if your server goes down tomorrow, the backup will be there.
Handling big data. While it’s technically possible to replicate a large amount of data offsite with file-based backups, it’s a whole lot easier to do with versioned replication. As your organization’s data needs increase, traditional backup and recovery vehicles become much more unmanageable.
Ask yourself, if your server goes down and you can’t access your emails and files, how long will it be before your business is affected? This is your tolerance for downtime. For some businesses having any sort of downtime on their systems is just not an option.
Does your company have an application that you rely on and use every single day? Maybe it’s a business management software, or an accounting application or email platform. Calculate what it costs your business to have this system offline and an unproductive workforce for 24 hours. You’ll often find that this is far more expensive than investing in a versioned offsite backup system.
If your company has a lot of time and money invested in an existing file-based backup system that works well, you still might want to consider switching to versioned replication. While this decision should be made on a case-by-case basis, there are many businesses, like an architectural firm, for example, where the prospect of losing a whole day’s worth of data after a server crash would be a big deal.
It may just be time to make the switch to versioned replication. You won’t be disappointed with the change.