Last year, I wrote an article on World Backup Day that was not much unlike what other enterprise marketing folks put out there. Truthfully, I do not understand the real motives behind this “day”, but I have been passionate about the protection of data for more than two decades. As a proof point, my first job out of college was with the company who produced the backup software that I used religiously.

Since Google recognizes World Backup Day, it must be real. 😃 So here’s what I did over the weekend for my personal data, but this also applies to enterprise data just as well.

Analyze your backup strategy

For many of us, protecting data is not the highest of priorities because we gravitate to focusing on higher-value activities. For me, setting aside a bit of time every six months to optimize my backups and automate it goes a long way.

Start by picking some data sets that are important to you or your business and walk through how they’re stored and protected. There’s a good chance you’ll find holes in your strategy or execution, or you’ll think of ways to better protect your data moving forward. As an example, here’s how I have been protecting my photos:

  • Primarily they exist on our phones (where they’re captured)
  • These photos are backed up to a computer when it physically connects (once every few weeks)
  • A percentage of photos are copied onto a computer when they need to be edited/utilized
  • Those photos on a computer are backed up nightly to disk onsite
  • Backup data is copied weekly to a disk drive to be taken offsite (photos are included in this data set)
  • Drive is physically taken offsite to protect from a major disaster

Immediately, I was able to identify flaws in my backup strategy for photos:

  • They exist unprotected on our phones for too long because we do not often connect our phones to our computers to sync them with iTunes. I do not want my photos going to iCloud (for a few reasons), so I enabled Wi-Fi sync to increase the frequency of these backups.
  • Our photos are all over the place. On different devices and in different backup archives. If we need to recover photos for a specific event or just utilize them more, it would take a lot of time to discover them. Once per month, I now set 30 minutes aside to set my favorite photos for that month into a folder structure on my home server, which is backed up daily. This now makes is easy for us to find past memories. Sometimes your primary data is not not organized properly.
  • While I have a semi-automated system for getting my data physically offsite, it does require me to pick something up and move it, which I’m often not good at. Eventually I get around to it, but it can leave my data only in a single site for weeks. With the cost of public cloud storage continuing becoming more affordable, like Google Cloud Storage Nearline or Backblaze B2, I plan to minimize or eliminate the onsite/offsite rotation of media. If I had multiple sites, I would utilize multi-region private cloud storage like SwiftStack or replicate the data using SwiftStack’s Cloud Sync. 😃

As you walk through how your data is protected, write ways you can improve your practice and work on implementing those changes in the near term.

Only protect the data you need to

Making copies of your data is easy, but with the amount of data growth that has occurred (and will continue to occur), it’s most often does not fit within budget to have copies of all data in all locations. Managing data you do not need also takes focus away from the data that really matters. For some data sets, you do not need multiple copies onsite and another offsite at all times, or you do not need to retain that data for years.

For example, I record surveillance data from a few cameras mounted on the outside of my home. This data has proven to be useful on more than a few occasions, but if for whatever reason this data was lost, it would not have an affect my life. Most people do not even record events that happen at their front door, so just having this data in a single location is better than most. Since this is a significant amount of data that changes every day, I choose not to protect this data aside from primarily storing it on a redundant system onsite.

Another example is that I realized I did not need to protect operating system and application data every day. I changed my daily backups to be data-only and now backup my entire systems every other week. This significantly reduced the amount of backup data I was storing without significantly complicating my disaster recovery scenario.

Take a look at what backup data you’re storing onsite and offsite and ensure it’s the right balance. You’ll find that you do not have enough copies of some data, while you’re storing way to much of other types of data.

Ensure you can recover your data

Even if we have a well designed backup strategy and we successfully execute that strategy, often we forget about the restore part of the greater solution. Few events are more stressful than right after primary data is lost. Even if you have backup copies, blood pressure shoots up until the data is safely recovered. This being the case, it makes sense to simulate recovery events once in a while to increase the success rate when it’s really needed.

When I took a look at my sets of backup data, first I realized I had data on certain types of media that are challenging to recover from. I actually had to recently recover a folder of files from LTO tape, which is not easy in my personal setup. The LTO-2 tape drive worked, but I did not have a computer anymore with a SCSI connection. I successfully used a SCSI-to-FireWire converter and connected it to an old Mac that still had a FireWire port. I got very lucky and it made me realize I had some work to do to get off these tapes. Once I had my setup working, transferring the data I still needed from tape to disk was fairly simple. I just needed to do it. Like many enterprises, I also still relied on tape much more than I should.

Aside from proving you can read data from your backup media, you also need to ensure you can get that far. Often many restores fail because you do not have the password or encryption keys. Others fail because the catalog of that media is missing and cannot be recreated. Practicing recoveries will uncover all of these issues that you can easily work out, as you do not want to have to work through them during a real-life recovery event. When doing it for real, you just want to quickly get the data back so your blood pressure can return back to normal levels.

A few times a year, just set aside a few moments to focus on your backup and recovery strategy and you will thank yourself the next time you need to leverage it for real.

About Author

Erik Pounds

Erik Pounds

Erik is an avid technology geek, attacks opportunities by building things, and currently leads the marketing function at SwiftStack. Prior to SwiftStack, he led the Sync team at BitTorrent, ran product management at Drobo, and held various product and marketing roles at Brocade and EMC. He proudly graduated from the University of San Francisco, where he captained their Division 1 Golf Team.