What’s up with being down?
If you’re a Google, Microsoft and Amazon user, you may recently have had a feeling that the sky was falling, or at least that there was a black cloud over your head. Why?
On Friday, Google services went down.
On Saturday, Microsoft’s Outlook.com service went down.
On Monday, Amazon.com went down.
That’s a lot of high-profile website outages. But the impact of the downtime was not evenly felt.
Google’s outage was amazingly brief: somewhere between 2 – 5 minutes. But the impact was huge. It was not just Google search that was down. Not just Gmail. But all Google services. So many people use these Google services that it was estimated that overall Internet traffic dropped 40% during those few minutes of downtime.
Microsoft’s outage was confined to a few services – Outlook.com, SkyDrive and Contacts – so it affected a fraction of the people impacted by the Google outage. But it lasted for days, not minutes. So if you used Outlook.com for your email, it was a seriously bad outage.
Amazon’s outage was for their main Amazon.com website for U.S. and Canadian customers (not related sites like Zappos.com or other country-specific versions of the Amazon site). Reports about the length of the downtime vary from about 15 minutes to 45 minutes. Not too long, but they may have lost some business as frustrated shoppers turned to other online retailers. And because Amazon.com averages $117,882 in sales every minute, a 40 minute outage could mean $4.72 million in lost sales (of course customers may simply have come back and purchased later).
Google’s brief outage also had real costs. The company is said to make about $108,000 per minute. So their 5-minutes of downtime would mean $545,000 in lost revenue.
So, what does all this mean? High-profile outages often lead to speculation about cyberattacks by individual hackers, hacker groups or even Chinese government hackers. But it doesn’t appear that hackers were behind either of this past week’s outages.
In all likelihood it’s just a high-profile coincidence. But it can be a meaningful coincidence and an important reminder.
To businesses large and small, it should be a reminder that their systems are vulnerable – and not just to headline-grabbing natural disasters. The reality is that it’s not whether your servers will go down (due to natural disasters, human error, equipment failure or other causes) but when. And when it happens, how fast and how well can you recover, and what will the impact to your business be? A prolonged outage or one that causes data loss can be catastrophic for many businesses.
So having a business continuity plan and services such as Cloud Server Backup and Cloud Disaster Recovery are essential to mitigate these very real and very big risks to your business.
And for all our high-speed and dial-up Internet access customers, these outages can be a reminder not to assume your Internet connection is down just because you can’t get to Google.com or Amazon.com.
If you can’t get to the one site you’re trying to visit, it’s always best to test your connection by trying to go to several different, unrelated sites like www.earthlink.net, www.npr.org, www.whitehouse.gov, etc. You may find there’s no problem with your connection. Just a website outage.