Does Your SMB Really Need 99.999 Percent Uptime?

Paul Mah

Some IT pros or managers might have had this experience: Your CIO or CEO comes up to you with the "good" news that he or she has found a vendor or provider who guarantees an uptime of 99.999 percent. Or as a senior executive, you must have heard your fair share of sales pitches exhorting the virtues of the "five nines" of availability, as it is known in the industry.


While no businesses would mind having higher availability for their systems, knowing more about this topic will allow us to separate the marketing fluff from the facts to better determine what our SMB really needs.


Let's take a quick look at the different levels of availability, and what they mean:

  • 99.999 percent availability equates to 5.26 minutes of downtime per year.
  • 99.99 percent availability equates to 52.56 minutes of downtime per year.
  • 99.9 percent availability equates to 8.76 hours of downtime per year.
  • 99.5 percent availability equates to 43.8 hours downtime per year.
  • 99 percent availability equates to 87.5 hours downtime per year.


I derived the above calculations using a simple Excel spreadsheet; you can do the same, or just refer to the comprehensive chart on the Wikipedia page on High Availability here.


Making Sense of the Numbers

These figures are the absolute downtime represented by the various levels of availability, though the figures need to be viewed in the proper context. For example, the loss of connectivity for 8.76 hours (99.9 percent availability) is not so bad when it comes to Internet connectivity. Assuming eight working hours a day with the outages occurring at random, this works out to only about 15 minutes of downtime occurring during office hours per month.


Of course, for a day trader, that 15 minutes without an Internet connection would be unacceptable. On the other hand, even a low availability rate of 99 percent probably would not be catastrophic for an intranet-based message board (5 minutes of downtime during office hours per working day).


High Availability is Expensive

So why is it important to figure out an acceptable level of availability or uptime for your SMB? Well, because achieving a high level of availability can be expensive. The cost grows exponentially with each additional "9" that you demand. This is especially true for SMBs, which usually start off with lesser hardware that can be deployed as redundant or standby units.


Implementing a 99.999 percent rate of Internet availability or higher, for example, will require an SMB to subscribe to another ISP for a redundant Internet connection, as well as to acquire a multi-WAN device from a vendor such as Peplink. And because five minutes of downtime per year doesn't leave much time to make any manual adjustments, another Peplink will have to be acquired to create a high-availability setup for automated failover. And yes, you also will need to purchase the appropriate UPS hardware to power all the hardware mentioned above, plus your network switch and/or your wireless access point.



While the technology does certainly exist to achieve a 99.999 percent uptime, it won't be cheap. As such, senior executives would be prudent to first question the need before committing to the additional cost.


In my next blog, I shall explore some factors that mitigate the need to specify a higher level of availability for your SMB.

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Add Comment      Leave a comment on this blog post
Nov 10, 2010 7:58 PM Terry Terry  says:

Great article, but why go for the 5 nines holy-grail when 4 nines is now so affordable? Take that standby unit, put it into service, marry it with the original server and HA software (not clusters or hot standby for reason below) and get 99.99% availability for a lot less than you think (88 seconds per month or less than 18 min a year using 8hr day assumption).

Also, what really needs to be figured into the equation is how long you are down for and how long it takes to figure out what was lost.  Meaning data in transit or heaven forbid the failed drive. When was the last back-up last nite?

Nov 21, 2012 6:07 PM JannySmith JannySmith  says:
Excellent post. VPS hosting with a higher uptime is very important for us. If the VPS goes down we can lose a lot of money. Reply

Post a comment





(Maximum characters: 1200). You have 1200 characters left.




Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.

Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.