Very small businesses frequently depend on computer systems with several potential failure points, where a single failure can cause the network to go down for an uncomfortable length of time.

  • Many small businesses – many of my clients – depend on a single server to hold files and run the company e-mail system. There are backups available if the server goes down, but those backups have to be restored onto working hardware – and it’s not possible to run down to Office Depot and buy a replacement server. Ordering a server, waiting for it to be delivered, and doing the work involved in disaster recovery is time consuming, and few businesses can withstand a week or two without computers and e-mail.
  • Almost everyone relies on a single broadband connection, a DSL line or cable broadband connection. If that line goes down, e-mail goes dead – and for many businesses, that’s the same as shutting the doors. There’s almost never an affordable alternative – few businesses are in a location where both cable and DSL connections are available, T-1 lines are expensive, and it takes expensive equipment to handle two Internet connections and switch from one to the other in the event of an outage. (Having two DSL lines kind of misses the point – if one is down, the other one almost certainly also would be.)

There was a dramatic example in the news of what happens when a network is not redundant. The outage at LAX that left 17,000 international passengers stranded on airplanes for hours was caused by a single malfunctioning network card on a single desktop computer. Here’s the LA Times article that brought this to light. Isn’t that remarkable?

As the cost of a potential outage grows, even a small business will have to start investing in redundant systems. It goes along with our dependence on these interesting machines.