My Current Data Backup System

My first computer hardware purchase of 2011 isn’t terribly exciting, but it is very useful: a batch of new hard drives. Let me put on my old-computer-user hat on for a moment and marvel that for about $450, I bought six and a half terabytes of storage. That was almost unfathomable just a few years ago.

It’s a good chunk of money, so I’m not spending it foolishly. Most of that spinning platterspace will be used for my backup system. As a self-employed author, editor, and photographer, $450 is a steal compared to what I’d face if I were to suffer a hard drive crash or otherwise lose my data if a backup system wasn’t in place.

A friend on Facebook asked what that system entailed, so I thought I’d elaborate here for anyone who’s curious.

As you’re reading the following, keep in mind that (a) I once had a bad hard drive crash that wiped out a lot of data, and (b) I’ve edited all versions of Joe Kissell’s excellent books Take Control of Mac OS X Backups and Take Control of Easy Mac Backups. So there’s a bit of overkill here. Or is there? Imagine losing all of your digital photos forever and then see if a good backup system is worth it.

The Computer

My main computer is a 15-inch MacBook Pro laptop that is where I do everything. I also own a Mac mini that’s currently acting as a media server and test computer.

When I’m working at home, the MacBook Pro is connected to a 1 TB (terabyte) external hard drive that has been set up as my Time Machine backup. Every hour, Time Machine makes an incremental backup of my laptop’s hard disk, copying only files that have been changed since the last backup and keeping copies of older versions of files. That 1 TB stores roughly a year’s worth of my data, so I can find a file that I deleted or edited from months ago if needed. (Time Machine does delete old files when the disk’s capacity reaches its full point, but so far I’ve never needed to go that far back in time.) Time Machine does not run when I’m at the office (a few times per week, depending on my schedule and workload) or traveling (infrequent).

The problem with a Time Machine backup is that it isn’t bootable—I can’t start up my computer from it in the event my MacBook Pro’s hard drive stops working. I could run the machine from it’s Mac OS X installer DVD and restore the contents of the hard disk from the Time Machine backup, but that would take hours.

So, I also maintain a bootable duplicate of my laptop’s disk, using SuperDuper!. Let’s say the internal drive pukes; if necessary, I can boot the Mac from the duplicate (via FireWire) and have most of my data available, and continue working. It gets me moving again as fast as possible in case disaster strikes on a deadline. It’s important to have both a duplicate and an incremental backup. A lot of people just keep a duplicate, updating every once in a while, but a duplicate represents your data only at the time it was last backed up.

That seems like a good strategy, but in fact I actually maintain two duplicates, which are rotated off-site. One stays at home, and the other goes to my office; if my computer and duplicate were both at home and the house burned to the ground, I’d be out of luck. As you’ll see, off-site backups are important.

For these duplicates, I buy bare internal 3.5-inch hard drives—just the drive, with no enclosure, cables, or power source. I already have too many outdated drive cases and power bricks, I don’t want more. To connect the drives to my computer, I use a “toaster,” in my case a NewerTech Voyager Q Hard Drive Dock. (My model has FireWire 800 ports for faster data transfer compared to a less expensive USB-only one.) When it’s time to run a backup, I insert the drive into the toaster (from the top, like a kitchen toaster) and power it on. For storing the drives, I bought a set of WiebeTech anti-static DriveBoxes. They’re easier to keep on a shelf, and I got tired of handling the slippery plastic bags the drives originally shipped in.

Lastly, although I don’t travel much, I bought a 500 GB portable USB 2.0 drive to use as a bootable duplicate when I go to Macworld Expo at the end of the month (and in case other trips come up).

External Data

Sticking with a laptop as my main computer has a limitation: laptop hard disks (2.5 inch) aren’t as capacious as full size (3.5-inch) desktop disks. This is especially true now that SSDs (solid-state drives) are finally becoming affordable, but in much smaller capacities. I generate a lot of data writing books and taking photos, so it all can’t live on my main hard disk.

A few years ago—before it was even realistic to have a 1 TB drive—I purchased a Drobo, a device that holds four bare internal hard drives and operates them as if they were one large disk (known as a RAID, redundant array of independent disks). What’s nice about a Drobo is that as my data needs expand, I can replace the drives with higher-capacity ones. The Drobo currently gives me 2 TB of storage, of which I’m using about 1 TB.

The Drobo contains my digital photo libraries (separate ones for iPhoto, Aperture, Lightroom, and Photoshop Elements…yeah, it’s kind of a mess, but I use them all for various tasks and assignments), my archive of past projects, and nearly all of my archived video and video editing projects. When I’m at home, the Drobo connects to the MacBook Pro via FireWire 800 and shows up as an external disk.

Being the repository for my digital photo library, the Drobo represents a significant point of failure in my system. So, of course, it needs to be backed up. In this sense I’ve been lucky: my amount of media has crept up slightly behind the capacity of hard drives, so until recently I was making duplicates of the Drobo’s data to two bare drives (a 750 GB one and a 1 TB one). Now that the Drobo contains more than 1 TB of my data, those drives don’t work as duplicates anymore, which is why I bought two 2 TB drives last week. (Remember, I’m using one for local backup and rotating the other off-site.) That 1 TB drive will end up in the Drobo, replacing an old 250 GB drive, once the first full backup completes (it’s running now, which is what prompted me to make a comment on Twitter about how unexciting buying new hard drives can be).

Internet Backups

The scheme so far sounds pretty thorough, but there are a couple of holes. Time Machine only backs up once per hour, so I could lose an hour’s worth of work—which, on some days, could be significant. Also, even though I’m rotating duplicates between two locations, there’s the possibility that both sets could be lost. Suppose Seattle gets hit by a giant earthquake or something? Sure, I’ll have more important things to think about than my data, like keeping my family safe, but at some point I’ll need to come back to my information.

So, I also use two methods of online backups.

The first is Dropbox, a service that’s proven to be almost essential in my digital life. Dropbox creates a folder on your computer (Mac or PC), and anything stored in that folder is automatically copied to the Dropbox servers on the Internet. Whenever you make a change to a file, it’s immediately copied in the background. What’s better—and the reason I started using Dropbox in the first place—when you install the software on another computer, that data is also copied there. I use it for quickly taking screenshots under Windows (running in VMware Fusion on the Mac mini) and having them appear almost instantly on my MacBook Pro. I also use the app Plaintext on my iPad and iPhone, which reads and writes files to my Dropbox folder, which means I don’t have to engage in the awkward sync-through-iTunes dance Apple has set up for synchronizing data files with iOS devices.

Another bonus of Dropbox is that the service maintains incremental backups of the data you store. If you accidentally delete a file from a Dropbox folder (something I’ve done a handful of times), you can go to the Dropbox Web site, locate the archived file, and restore it. For this reason, I now store my active projects in a Dropbox folder.

Dropbox offers 2 GB of storage for free, or you can pay for more capacity. (If you want to sign up for Dropbox, please consider using this link, which earns me free extra storage capacity.)

Last year I also started using CrashPlan, which offers online backups using its CrashPlan Central feature. That enables me to backup much more data than what Dropbox offers; I don’t back up my entire hard disk, but I do include my home directory and my Never Enough Coffee projects folder, totaling about 86 GB.

What’s great about CrashPlan is that you can use it to back up other computers on your network (the software is free for this purpose. I have it running on the Mac mini, where it backs up my wife’s laptop to an external hard drive; her MacBook Pro lives in the living room, which she doesn’t want cluttered with external hard drives and other computery things if possible. A novel CrashPlan feature, which I don’t currently use, is the capability to back up your computer to a friend’s computer over the Internet. If you’re both running CrashPlan on your machines, you can set up an off-site backup (and likely he or she would back up data to your computer), all of which is encrypted and protected.

The Rest of the Scheme

I use the same principles to back up the other machines at home, too. The Mac mini has a Time Machine backup and a duplicate—although in that case, both live on separate partitions on a single 2 TB external disk, without offsite duplicates because that machine doesn’t hold any critical data. For my wife’s laptop, in addition to the CrashPlan backup, I have rotating external bare hard drives as duplicates.

You’ll notice I don’t archive stuff to CDs or DVDs—that’s just too much trouble, and I’m not a big fan of optical media anyway. iPads, iPhones, and iPods are all backed up via iTunes to their respective synching computers (which are getting backed up), so I don’t worry about losing their data.

So that’s the mad backup scheme. There’s some work involved, mostly in updating the duplicates and rotating them off-site, but otherwise everything is automatic and operates in the background. No doubt overkill, but I don’t want to risk it. And as I mentioned earlier, I have lost data, important data, in my past, and it’s no fun.

Lastly, a backup system is only as good as the integrity of its data, so it’s important to make sure the things you intend to back up are really getting backed up. My friend Adam Engst has suggested that every Friday the 13th be “International Verify Your Backups Day” (read the article for details), which is a great idea.

Maintaining backups and buying hard drives isn’t sexy or exciting. But you don’t ever want to find yourself on the losing end of a hard drive disaster. Hard drives fail, computers die, and often they do so suddenly and inexplicably. When it happens, having a robust set of backups will save you an enormous amount of time, stress, and sleep.

  1. Backing up files can be such a labor-intensive task. But doing it can give us assurance that our files our safe. I envy you for having a 2TB disk! I can imagine the number of files you can store there – you’ll virtually have limitless storage space!

    Reply

  2. Ugh, indeed. There’s nothing worse than losing all that important data – so much time and effort spent gathering data, all went to waste due to some unforeseen tech difficulties. It’s a nightmare when you lose things related to work or family. There’s nothing wrong with being gung-ho about data backups.

    Reply

  3. Back-ups for data are important. When things go awry, it would be unbelievably problematic if the data in your computer goes kaput. It’s a good investment to have a safety net to save your data in case of problems.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: