Over the course of this tutorial, I’ll walk you through setting up a backup plan that works for you, I’ll tell you about the surprising number of options available (and why you shouldn’t pick just one!) and I’ll also point out some of the subtler benefits to backing up that you may not have considered.
You’ve heard it from a friend. You’ve seen the threads on countless forums, across all disciplines of work, in all corners of the Internet, written by all manner of otherwise perfectly savvy computer users — users who, up until that day, believed that they were safe. Users who believed, like you do, that data loss is something that happens to other people.
Ultimately, no amount of reading about the woes of others can prepare you for the sinking, panic-filled feeling of realizing that you’ve just lost something crucial.
Luckily, we can.
To Backup, or Not to Backup?
That is not the question. These days, the integrity of our digital lives has taken on an importance that no longer allows for an attitude where we trust in the power of technology to save us from accidents, negligence, or laziness.
No matter how advanced it becomes, it is not technology’s responsibility to maintain our data. We call it a backup plan because it’s not a question of whether it will happen, but of when and of how prepared we are to deal with it when it does.
No matter how advanced it becomes, it is not technology’s responsibility to maintain our data.
But when the burden falls to us, it is easy to be overwhelmed by the panoply of choice that confronts us even for the simple task of choosing a backup method, let alone actually implementing it. Part of the problem is that many backup advice articles you may have encountered describe solutions that are not applicable to your situation. They may be needlessly complicated, disorganized, or challenging to manage on an ongoing basis.
That is unhelpful. Instead of taking that approach, we’re going to discuss the available options and help equip you to make the choice of backup arrangement yourself, so that you are in control.
The first step is familiarizing ourselves with the options.
The Trouble With Acorns
Consider, if you would, the common squirrel (if you’ve not been outside in a while, imagine a tiny furry child with an affinity for nuts and climbing). When winter approaches, the squirrel prepares by storing away gathered acorns in a variety of places; he will store some near his burrow (for easy access when the mid-February munchies roll around), and then scatter some across his territory in secret burrows and caches (so he can snack no matter where he ventures).
This is because squirrels understand backup plans very well. They know that a proper backup is only helpful if it covers a variety of possible disaster scenarios and allows for access from a variety of locations. Rodent risk management at its finest, people.
What we need to take from this analogy is the notion of multiple backup locations. The basic principle is that you have a backup that is local, and you have one that is remote. Notice that I said “and” rather than “or” — by the end of this article, you too will scoff at friends who boast about having “a backup”. Just one, you will marvel, think of the squirrel!
The most obvious, common, and fragile backup solution is the one you set up in your own home/studio/office/bathroom/etc. Its advantages are obvious: easy access, convenient restores, direct management. For most people, a local backup consists of simply hooking up an external drive to their machine and letting Apple’s built-in Time Machine utility handle the rest.
Start with the basics
Truthfully, that may be sufficient as a local backup solution for many people. It is not particularly robust, but it is about as uncomplicated as it gets and will offer you a peace of mind rating of about two out of ten squirrels, which is better than nothing.
Since it is beyond the scope of this article to give a tutorial for using every single service we’ll touch upon (they all have excellent documentation), let’s discuss some of the major players in the local backup arena…
Apps & Strategies
- Time Machine: Setting Time Machine up is as simple as connecting your external disk and turning TM on from within OSX’s Settings. Once you’ve selected the disk, TM will start its regular backup schedule: performing a full backup and then backing up changes every hour, while maintaining hourly backups for the past day, daily backups for the past month, and weekly backups for as far back as it can. While it lacks control over backup scheduling, you can exclude files from the backup to avoid redundancy and performance issues (things like Windows VMs from Parallels will really slow things down). For a simple, integrated solution, Time Machine is sufficient.
- SuperDuper: One of the most popular third-party solutions is SuperDuper, a utility that’s specifically designed to create bootable clone copies of your system drive, so that if it fails you’re able to restore a 1:1 copy to new hardware. The SmartUpdate feature keeps your clone up to date without having to redo the entire filesystem each time you, and the detailed expert features give you powerful control over how files are handled.
- Carbon Copy Cloner: A personal favorite, this powerful application offers the precision cloning and incremental updating of SuperDuper, but also includes customizable scheduling, an open backup format (so Migration Assistant can read it), and a very well designed user experience that takes the difficulty out of setting up your backup plan. The ability to set up email alerts and simple rules for backups make this among the most powerful and reliable local backup solutions.
- ChronoSync: Another powerful option, ChronoSync features many of the benefits already mentioned and includes file versioning, enterprise-level multi-mac backup options, backup analysis tools, and more. It is probably overkill for most home uses, but for a small business owner it might be worth considering.
- Hazel: While not strictly a backup utility, Hazel’s powerful rule system can be used to set up precise backup tasks that work alongside your more complete solutions, offering you the ability to, for instance, organize everything that enters your Downloads folder into categories, and then back it up to a USB key whenever that is plugged into the computer. I will discuss my usage of Hazel when I reveal my own backup solution later on in the article.
This is by no means a comprehensive list of options, but selecting one of the above utilities will cover almost all possible local backup scenarios for the majority of users and (set up properly) offer you a peace of mind rating of about four out of ten squirrels.
You may have heard this acronym before and wondered if you’d accidentally wandered into the bug spray aisle of the Internet, but in fact RAID is a data redundancy method that you should be at least generally aware of when constructing your backup plan.
RAID (Redundant Array of Independent Disks) is actually a series of methods for distributing your data across multiple hard-drives in order to keep it safe and, in some cases, increase performance. There are now many types of RAID array, and the theory behind them is quite complex, but it is worth taking a bird's-eye view of the major ones:
- RAID 0: The most basic RAID array is one that distributes data across two or more physical hard-drives. In essence, every file is written so that it is being accessed from multiple drives, allowing the bandwidth to be shared and the read/write times to increase significantly for each additional drive. Performance gains are significant; unfortunately, there is absolutely no redundancy and the risk of failure is huge since any single drive failing will destroy the entire array. Why even mention it then? Keep reading.
- RAID 1: The primary RAID method for safety is RAID 1, which requires two hard-drives and writes all data to both drives. Effectively, your computer uses two physical drives to represent each virtual one and keeps the data identical across both physical disks. The advantage is that there is no data loss even if one of the drives dies.
- RAID 1+0: Also known as RAID 10, this arrangement combines the best features of the two methods above into a single hybrid array. A minimum of 4 drives is required for a RAID 10 array; the easiest way to imagine it is to think of a RAID 0 system with each branch actually consisting of two drives in a RAID 1 array.
These diagrams are useful for understanding the structure of RAID arrays.
As you might be fearing, the cost of having so many drives and the necessary RAID controllers is prohibitive for your computer, so where does RAID actually come in handy for a backup solution?
Ideally, you would implement a RAID 1 or 10 system in your local external backup by purchasing an external drive bay and filling it with 2 or 4 physical disks (as necessary). This way, instead of your external backup being a single disk that’s as prone to failing as your internal drives, your data is being handled with redundant security.
Now we’re at a peace of mind rating of about six out of ten squirrels, so what’s left to do?
Today, cloud storage is the off-site backup solution of choice.
In years past, we would be discussing putting external drives (or DVDs, remember those?) into a storage locker somewhere away from your workplace as an addition to your local backup. This, by the way, is not an obsolete or bad idea by any stretch of the imagination (except for the DVD bit), and now exists as a sort of “third” storage location option.
The major issue with this kind of physical off-site backup is that it is extremely inconvenient to manage; you have to drive out to the storage location each time you have new data to add, and there’s no way to access the information or restore it without retrieving the physical disk.
Nowadays, this method is more suitable for files that you’re unlikely to need on a regular basis, as a sort of long-term archiving system for old project files, documents, and similar odds & ends.
Today, cloud storage is the off-site backup solution of choice.
The largest shift in computing over the past few years has been toward a cloud-based architecture for communication, data storage, and even applications. A lot of paranoia surrounds the notion of cloud storage, and while caution is required, the technology is no longer nascent and has evolved into a reliable and amazingly powerful affair that offers way more redundancy and security than any local system could accomplish.
For those a bit fuzzy on the details of what exactly this cloud thing is, the answer is actually very straight-forward: cloud storage means that data is held on massive distributed storage servers owned and managed by specialized companies whose job it is to ensure the integrity, availability, and redundancy of that data for you. They, for example, can afford to implement RAID arrays in each server so that your material is secured in each data centre — and they often have several distinct data centres that share information so that one entire facility can go dark without data being lost.
The major challenge with cloud storage, then, is choosing to align yourself with services that have a proven track record of stability, are open about their practises and security measures, have clear plans in place to mitigate disaster scenarios, and who offer you a convenient means of taking advantage of these services at a fair price.
Here are a few that fit the bill:
- CrashPlan: Among the most trusted names in online backup, CrashPlan (and their related Pro offering) embody the best in cloud storage solutions: security, convenience, customizability, and unlimited storage for a paltry $3/month. Their cross-platform applications make it dead simple to initiate, schedule, monitor, and restore from your backups on a file-by-file basis, and all transfers are secured with a powerful 448-bit encryption that keeps data private.
- Carbonite: Much like CrashPlan, Carbonite offers an alternative cloud backup solution that offers a similar feature set with a focus on automation and easy data recovery, including the option to have your data mailed to you on a physical drive (if you’re in the US) for quicker recovery.
- Mozy: Mozy’s comparable feature set makes it worth looking at, but its significantly higher prices and storage limits keep it from being the most competitive option.
- Arq (Amazon): While some companies have focused on setting up their own cloud network, others have taken a different approach. Arq makes use of Amazon’s existing cloud storage network, S3, giving you the same features you’d expect from online backup services, but with data stored on the servers of one of the most powerful cloud networks on the Internet, backed by Amazon’s incredible 99.999999999% durability guarantee and the peace of mind that comes from trusting a $100B company rather than a smaller one that may be more likely to go out of business.
- DropBox / SkyDrive: These and other similar offerings are not technically backup services, though some people use them as partial backups. Since their primary goal is file synchronization, they are not optimized for creating bootable backups, or for performing secure incremental backups of an entire computer filesystem. That said, they are incredibly useful for keeping your personal files in sync across multiple devices and giving you an extra layer of protection for those absolutely critical files.
Now we’re finally at nine out of ten squirrels of peace of mind. All that’s left is to discover how to attain that elusive last squirrel…
Backups as a Way of Life
Once you have established a system that keeps your files secure both locally and offsite, you have done your utmost to prepare for any eventuality — whether it’s a simple drive failure or a catastrophe like an earthquake, your data at least will remain safe. But there’s more to a good backup plan than just setting it up.
While many of these systems allow you to create a “set it and forget it” kind of arrangement, it is important to be proactive about your backup system: adjusting it as your needs change, checking it periodically for integrity, keeping an eye out for better methods that may crop up, and so forth. Perhaps the most important point is that no matter how you’re doing it, you need to keep your backups current.
A complete backup plan rides on 3 pillars: local storage, offsite storage, and your ongoing awareness.
It won’t be of much use to you if your latest backup is a month old and you’ve lost something you were working on yesterday. Make a habit of doing regular backups so that the plan you’ve spent all this time researching and establishing is actually able to save your data when the time comes.
Not only that, but you need to test the system! This is one step that is often overlooked, but no backup plan is complete without verification, so once you have everything set up the way you'd like, imagine a disaster scenario and actually try to recover your data from the backups — see what it would entail, how long it might take, if there are any gaps that you need to fill. Only then can you feel totally secure.
When your backup plan consists of a local copy, an offsite copy, and a habit of keeping them both up to date, then you will have attained data security nirvana: ten out of ten squirrels; peace of mind.
How to Train Your Computer: An Example Backup Plan
To wrap things up, I want to offer you a glimpse at my backup plan so that you have an idea of how the information I’ve presented you with might be assembled into a working system.
I work as a composer for media and a writer, so my backup needs are probably similar to those of other media professionals among you. As with all things of this nature though, my setup will not necessarily be good for your uses and doesn’t represent an ideal system for everyone — it is simply an example that may help you think in terms of making your computer and backup system work for you.
Layer 1: Local Backups
The most basic level of backups that I have are a series of local ones, beginning with a full clone of my boot drive, established and maintained via Carbon Copy Cloner on an external drive. I update this clone roughly once a week, because realistically I’m unlikely to have major changes to my filesystem occurring more frequently than that. Both my desktop and laptop are available on separate partitions of the same drive.
I then have a standard backup of my boot and primary data drives via Time Machine. Time Machine’s elegant interface for plucking specific files back from the brink makes it worth me keeping a drive dedicated to it. I keep it turned off while I’m working (because backups running in the background slow down my music work) but have it on every evening to catch up.
Each of my composing gigs tends to result in large folders of recorded audio, session files, mix files, video, and documentation. Projects live on a dedicated folder on my computer, but every three months or so I archive the old ones to a dedicated external portable drive. To make this as easy as possible, I’ve enlisted Hazel’s help.
Essentially, I created a folder called “Archive” on my machine where I drag all old project folders, and I set up a rule so that Hazel monitors the folder for anything coming in. Each time it detects a dropped folder, Hazel compresses it, and appends a date stamp to the name. Then, whenever I plug in the portable drive, it moves the compressed project files off of my computer and onto the external for storage, allowing me to free up some space and keep my projects directory clean and relevant.
This works extremely well for all my music, photography, and even writing work and has encouraged me to establish a more organized and consistent file saving procedure so that I don’t leave things scattered around my drive — a subtle but meaningful perk of having a good backup plan!
Layer 2: Offsite Backups
Because I consider it to be overkill for my needs, I do not have a physical off-site backup in place like a drive in a safety deposit box. I am comfortable enough with the data redundancy offered by other aspects of my plan to conclude that in my case it would be an extra effort without significant benefit.
I’ve got my data flying all around the cloud though. I use Arq to keep a backup of most of my files, including the project files again, all documents and music, and software that is not as easy to re-install as my Mac App Store purchases. Arq is set to perform a backup every evening at around 8:00pm.
On top of that, I have a DropBox Pro account with which I keep all of my documents and day-to-day work files in sync across all my devices, with built-in file versioning and the peace of mind that comes from knowing they’re safe in the cloud and not tied to any particular device.
Layer 3: Keeping it Current
Between the variety of methods I mentioned, you’d think it’s a pain to keep everything in check, but my plan is remarkably hands-off in terms of actual maintenance, which allows me to focus on ensuring that I’m always using the best solution and that I’m prepared to recover from a disaster as quickly as possible.
By now, you’ll have gained a stronger sense of what backup options are available to you and, hopefully, why it’s crucial that you make use of not just one, but two or more methods to keep your data as secure as possible. Don’t become one of the Internet’s sob stories.
Backups are like autosave for your digital life, so make sure you take some time to build a backup plan that meets your needs and lets you focus on creating, exploring, and enjoying your life on the computer.
As this is a huge and important topic, we’d love to hear about how you protect your data and what services and tools you’ve come to trust and admire — share your feedback in the comments!