If you run an online business, or your business relies heavily on its online presence, chances are there's nothing that will ruin your day more quickly than a report that your website is down.
For many companies, a website that's inaccessible means missed opportunity, and lost revenue. So it's no surprise that companies large and small spend a lot of time worrying about how to maximize their website uptime.
But is uptime overrated?
On Hacker News, a 2005 post written by 37Signals' David Heinemeier Hansson resurfaced. In it, he wrote:
If you’re Wal-Mart and your credit card processing pipeline stops for 30 minutes during prime time, yes, the world does end. Someone might very well be fired. The business loses millions of dollars. Wal-Mart gets in the news and loses millions more on the goodwill account.
Now what if Delicious, Feedster, or Technorati goes down for 30 minutes? How big is the inconvenience of not being able to get to your tagged bookmarks or do yet another ego-search with Feedster or Technorati for 30 minutes? Not that high. The world does not come to an end. Nobody gets fired.
It's an interesting post to read more than five years later, particularly given that there's some additional evidence to support Hansson's argument. Take Twitter for instance: despite all of the fail whales, Twitter continued to grow and is today still one of the hottest consumer properties on the internet.
At the same time, the internet has evolved considerably since 2005. There are a lot more consumers on the internet, they are more sophisticated, and in many cases, their expectations are higher. That means that companies can't simply expect to go down for the count on a regular basis and survive Twitter-style.
Which raises the question asked back in 2005: how important is uptime, and how much should be invested in maximizing it?
Obviously, there a variety of answers to these two questions. A lot depends on your circumstances. For small and mid-sized organizations operating online, there's good news: there are plenty of affordable things you can do to improve uptime. Here are three.
Outsource the weak links and tough stuff
In my opinion, one of the best ways to maximize your uptime is to offload from your plate the things that are most likely to cause pain. Two examples: email and DNS.
Setting up DNS on your own server is quick and cheap, but I can't count the number of times that I've seen a website become inaccessible not because the website itself was down, but because a DNS server crashed. Providers such as Verisign, DNS Made Easy, easyDNS and even GoDaddy offer DNS hosting packages that even a startup can afford.
Set up monitoring
If your Apache server crashes, your website could be down for hours if nobody else notices. Which highlights a key point: minimizing downtime is often largely dependent on minimizing the time between a website becoming inaccessible and someone who can do something about it finding out.
To minimize this time, set up monitoring so that you're notified within minutes of your site going down. Your hosting company may offer uptime monitoring free of charge, or for a nominal monthly fee. There are also affordable third party solutions, such as Alertbot, Alertra and Wormly.
Hire an on-call server administrator
Many companies don't require a full-time system administrator to be on staff. So hiring one to is an unnecessary expense. At the same time, however, there's a good chance that you can't rely on your freelance web developer(s) to be there to wear the sysadmin hat when your server becomes unresponsive at 2 am in the morning.
The good news is that on-call support doesn't have to cost an arm and a leg. Numerous companies provide outsourced server administration, which typically includes server monitoring and responding to server emergencies automatically.
Due diligence, of course, is a must, and popular hosting message boards can be a good place to find and vet these companies, which can save you a lot of hassle -- and sleep.