Digital Earth Weblog

Digital data has become a part of life for a lot of people. Some of that data records moments in time that can never be repeated. Some is an artefact of business. The reason why we do backups is to reduce the risk of losing that data, because:

• We have assigned some personal value to it, or
• The loss of that data would have a cost (money or time).

Everyone knows that they should do backups, but as my dentist says. “You should only floss the teeth that you want to keep.” Does it happen? Usually not until they have lost something important.

Now, before going any further I have to follow a time honoured tradition, when writing about backups, by asking: “You are doing backups, right?”

This question is actually a little dangerous. You can answer “yes”, feel good, and still be quite exposed to data loss. A better way to look at the situation is to ask yourself:

• How could data loss occur, in my situation?
• What are the realistic risks of that data loss happening?
• How valuable is the data?
• What is the cost of data loss?

The answers to those questions depend on your personal situation. However, the answers will give you a much better starting point for working out a backup strategy that works for you.

Data Loss and Risk

I can think of a few ways to lose data:

1. Hardware failure,
2. Malicious acts,
3. Natural and man-made disasters, and
4. Human error

Before going any further, have a think about which of those you have considered to be the highest risk. Make a mental note. Done? OK, let’s keep going.

Hardware Failure

Most people I talk to think that backups are done to protect themselves from hardware failure. Hard disks do suffer from mechanical and electronic failures. A hard disk failure is a catastrophic event, and all data is usually lost.

However, most people also overestimate the risk of hardware failure, and don’t consider other risks. Think of it as similar to how humans overestimate the risk of dying in an aeroplane crash. It does happen, and all on board usually perish, but the chance of dying in a car accident is significantly higher.

Overestimating the risk of hardware failure can actually cause people to underestimate the risk of data loss. That sounds totally counter-intuitive, but bear with me. If a person thinks that their biggest risk of data loss is from hardware failure, but they haven’t ever suffered such an event, their perceived total risk of data loss will be low. Throw in a bit of “it won’t happen to me” and backups might get done, occasionally…maybe.

Malicous Acts

People steal things. People write destructive viruses. People intentionally delete other people’s data. It happens.

Your backup strategy needs to consider these threats and weigh up the risks. If you back up your precious family photos to an external disk drive you have two copies (good). If someone steals both your computer and external disk drive, you are…well…screwed (that’s bad).

If a virus starts deleting files, and can get to the external disk drive because you have it mapped as a drive letter, its the same story. Storing two copies of data on separate devices is probably not good enough.

Good anti-virus protection, a firewall and physical security are important parts of an overall data protection plan, but they will not be discussed further in this post.

Disasters

Disasters happen, at different scales. If your house burns down due to an electrical fault that is a disaster for you. If hundreds of houses get burnt down in a bushfire that’s a disaster for the larger community.

Take your pick:

• Fire
• Flood
• Lightning strike
• Severe storm
• Power surge
• and more

All can wipe out your computer hardware, and possibly your offline storage media.

How many times have you seen images like the one above on TV? People sifting through the ruins of their home looking for family photos and heirlooms, sobbing “It’s all gone”. The risks might be low, but it happens.

Human Error

Humans do silly things. We drop things that don’t bounce well. We delete stuff we need later. We overwrite later versions of files, etc.

Something like 80% of all data loss in corporate environments is due to human error. I’d wager that the ratio is similar in the home and small office, but probably more due to lack of familiarity with the technology than anything. Home users aren’t usually IT professionals, and nor should they need to be to prevent data loss.

When was the last time you accidentally deleted a file that you needed, or wiped out some changes that you had made? I did it two days ago. In a rush to clean up my desktop I put a bunch of files in the rubbish bin. I then emptied it without checking what I was deleting. I lost a file I had been working on. (I had a backup so I only lost the changes I’d made that day.)

Reducing Your Exposure

The first step is to recognise all the risks to your data and make a realistic assessment of them. All the above can happen, but when formulating a backup strategy you should recognise that:

1. You are probably more of a risk to your data than any other factor.
2. Other risks also exist, and should not be ignored.

(What did you think was the highest risk to your data?)

The next step in creating a backup strategy is to consider the value of your data. (That will be the next instalment in this series.)

[Note: The images used in this post were obtained from iStockphoto and are subject to their licence agreement, not the Creative Commons licence used for this website.]