Like it or not, cloud-based storage -- that is, data stored virtually on servers that are typically owned and managed by a hosting company -- is the future of data storage. in fact, one could say the future is already here, and what comes next is just the minor evolution of a major (but relatively quiet) revolution. As of two years ago, over an exabyte (that’s 1,073,741,824 GB) was stored in the cloud (see the cool infographic here) and a 2012 survey predicted that a third of people’s personal data would become cloud storage fodder by 2016.
We’re not here to poo-poo cloud storage; if we did, we’d be hypocrites. It’s not only impossible to circumvent cloud storage at this point, if we could, it would be a huge mistake. Both of us use consumer-based products (chiefly Google Drive and Dropbox) in both our personal and professional lives, and can attest to how useful it is. Other labels we’ve talked to use subscription-based services that sync their files in the background of their computers, and it seems to work for them (to a point).
As information professionals concerned with preservation of data, we believe that cloud storage has a role to play in a modern preservation plan. If you’re familiar with the 3-2-1 Rule (make three copies, store two on different types of media, and keep one in a different location than where the other is), you’ll know that cloud storage can handily be one of the supporting legs, especially apt for the different location copy.
That being said, there are risks associated with cloud storage, just as there are risks associated with storing your label's data on a server, or a series of external hard drives. And just because a major corporation is at the helm of protecting your data doesn’t mean those risks are essentially nullified. With those caveats in mind, we’d present a series of cloud storage caveats to bear in mind as you sync your label’s files to the cloud.
Make it, Dump it, Forget it
Probably the easiest mistake to make is to use the cloud as your only preservation plan. We’ve been guilty of this at times as well -- create a word doc in Google Drive export your PDF copy, and forget about it. If you create material in the cloud, you’re doing yourself a disservice by leaving your only copy there, whether or not a local copy is synced to your hard drive. You’re also opening yourself to potential data loss which, despite what cloud companies say, is a genuine possibility. Roger Grimes at InfoWorld.com writes: “The cloud vendor usually states that they do awesome, triple-protected data backups. But even in cases where vendors said that data backups were guaranteed, they've lost data -- permanently. If possible, your company should always backup the data it's sharing with the cloud or at least insist on legalese that has the right amount of damages built in if that data is lost forever.”
Last month, we showed you how to easily backup your Gmail using Google’s Take-Out export service. The method can (and should) be applied to your cloud-stored or cloud-created data: select only your Google Drive files, export as a zipped package, and put it with your other local backups. Realistically, this should be a regular task, so that you aren’t leaving one copy in one location. Your local Dropbox folder should also be a part of your local backups -- don’t fall into the trap of assuming a folder can be ignored since it’s backed up online.
The Process of Getting Your Data Back
Let’s say the worst happens and you lose your data through theft or disaster. How long is it going to take to get it back? Pete Lamson, senior vice president at Carbonite, told CIO.com: "Before you commit your data to a vendor, find out how quickly you'll be able to get it back in the event of data loss or disruption, what the restore process looks like and what kind of support you can expect to receive if you run into any issues."
Carbonite, like other cloud storage companies, offer customers a restore process to move their files back onto computers, but a necessary question to ask is how that process works. Do they provider offer bulk transfers? Are you subject to bandwidth limitations on your recovery process? If you synced a hundred gigabytes of data, how long will it take them to transfer it back to you? Paul McClure, chief technologist at Cloud Solutions Group, told CIO: “Cloud storage requires moving data outside of the enterprise's local area networks into a wide area network, often resulting in a higher cost and bandwidth requirement for cloud storage.”
Probably the most pressing concern about modern cloud storage is ownership. Who owns your data when you sign up to drop things into the cloud? Often, the answer is incredibly complicated, since cloud managers need to retain certain rights to transfer and manage data. The wording that is used is ready-made vague, and open to interpretation. Grimes writes: “Cloud vendors like owning the data because it gives them more legal protection if something goes wrong. Plus, they can search and mine customer data to create additional revenue opportunities for themselves.”
In a high-visibility example, U.S. patent attorney Michael S. Neustel told CIO last year that Google’s policies related to Google Drive are “inherently unfair to users, since it requires Google Drive users to provide Google with several unnecessary rights to their copyrighted works.” Neustel is referring to Google’s terms of service, which gives Google the right to use consumer data to create "derivative works" from content stored in Google Drive, and to "publically display and distribute such content" even with their partners.
The end result is a reminder that we’ve made before, and is absolutely worth repeating: use cloud storage, and use it well. But in the back of your mind, just remember: