How to Protect Data When Traditional Backup Isn't an Option

Data backups are undeniably important--especially as ransomware continues to rear its ugly head. As a general rule, any data that has value should be backed up. Having said that, some data sets may be impractical or even impossible to back up. Here’s how to protect data when traditional backup seems to be out of the question.

Of course, every data set is different, and every organization’s situation is unique. As such, organizations will generally need to look for creative solutions when strategizing how to protect data. This blog post lays out a number of questions that you might consider as you look for ways to overcome data protection challenges. Not every question will pertain to every situation, but that’s OK. The questions are designed to help you to think of some out-of-the-box approaches for how to protect data when traditional backup just doesn’t make sense.

What Is the Main Issue Preventing You from Backing Up a Data Set?

The first thing that you should do if you are having trouble backing up a particular data set is to identify why. Identifying the issue up front puts you in a better position to work the problem. If, for example, you can’t back up the data because doing so would be cost prohibitive, then there may be a way that you can shrink the data or back up a subset of the data while archiving the rest. Similarly, if your difficulties are tied to an issue in the backup software itself, there may be a way to work around the issue or perhaps even adopt a different backup application.

What Is the Data’s Change Rate?

One of the first big questions that you should consider is how rapidly the data in the data set changes. If you find that the data set has a relatively low change rate, then you may be able to find a way to protect the newest data. Although protecting only the newest data is not ideal, the most recently created data often holds the most business value. Once you put a solution into place for protecting the most current data, you can begin working out how to protect data that is older.

Can Any of the Data Be Expired?

In many cases, data needs to be retained only for a specific amount of time. For example, compliance mandates might require that an organization hold onto data for a certain number of years. Or, an organization might find that certain data has little business value once it reaches a particular age.

When situations like these apply, you may be able to purge some of the outdated business data, thereby reducing the volume of data that needs to be backed up. You could also configure your backup so that it ignores data beyond a certain age. 

Can the Data Be Compacted?

Another factor to consider is whether the data that you are trying to protect can be compacted. Many database platforms include a compacting tool that can shrink databases by removing empty database pages or optimizing the data in other ways. Oftentimes, compacting a database will significantly reduce the amount of space that the database consumes, and it may even improve the database’s performance.

Keep in mind that databases are not the only type of data that can be compacted. If you are trying to back up virtual machines, you may also be able to compact the underlying virtual hard disk files. Server virtualization platforms often create dynamically expanding virtual hard disks. These virtual hard disk files start out small and then grow as data is added to the virtual hard disk. Depending on the virtualization platform that is being used, however, these virtual hard disk files may not automatically shrink if data within them is deleted. You may be able to significantly reduce the size of your virtual hard disk files by compacting them.

Is Backup Seeding an Option?

Sometimes a data set is simply too large to back up, regardless of any steps that an organization might take to retire old data or compact the data that needs to be protected. In this type of situation, backup seeding may be an option.

The idea behind backup seeding (which goes by a variety of different names) is that it is likely going to be impractical or impossible to back up a huge data set to the cloud if the data must traverse a WAN link. The major cloud providers offer programs in which they physically ship an appliance that you can use to set up a temporary data center. You can write a backup to this appliance and then physically ship the appliance to the cloud provider. The cloud provider then ingests the data into the cloud data center. At that point, there is a relatively current backup of your data in the cloud, so all you have to do is back up newly created or modified data.

Can You Protect the Data Through Redundancy?

If traditional backups simply are not an option, then you might be able to offer at least some level of protection through redundancy. For example, you might be able to mirror the storage arrays that contain the data.

Another option might be to take advantage of snapshots. A snapshot is different from a backup because it does not create a true copy of the data, which means that it won’t protect you against data-loss events stemming from an array failure, a fire in your data center or a similar catastrophe. Even so, a snapshot will generally give you the ability to revert a system back to an earlier point in time. As such, you might find that you are able to use redundancy and snapshots together in a way that gives you an acceptable level of protection for your data.

Bottom Line: How to Protect Data

Not backing up data is not an option, but traditional data backup doesn’t always make sense. There are several techniques that organizations can leverage to ensure that any data that has value is protected in an efficient and cost-effective way.