Unplug & Recharge//

When Servers Fail

An unexpected look at the power of technology.

Image by Johnny Simon, Quartz.

This is an excerpt from Quartz’s new book The Objects that Power the Global Economy. You may not have seen these objects before, but they’ve already changed the way you live. Each chapter examines an object that is driving radical change in the global economy. This is from the chapter on the server farm, which explores the heart of the internet.

WHEN SERVERS FAIL

  • February 2015

On Feb. 1, an outage in a Fujitsu data center in Perth during a thunderstorm affected the systems of Western Australia’s department of health. The department cares for the state’s population of 2.6 million people and oversees 2.5 million sq km (960,000 sq miles), which is the largest area administered by a single health department in the world. Three months later, Fujitsu announced an AU$10 million (US$8 million) update to the data center.

  • September 2015

On Sep. 20, Amazon Web Services’ DynamoDB database service received too many requests from users. AWS had to pause requests, which affected their cloud service. Clients like Netflix, Medium, Pocket, and IMDb experienced slowdowns and disruptions for about five hours.

  • December 2015

On Dec. 3, a configuration problem in the Microsoft Azure cloud-computing service left users in western and northern Europe without access to their emails for about four hours. The issue caused traffic to be wrongly routed, which in turn caused a service interruption that took down the Office 365 suite.

  • January 2016

On Jan. 14, a maintenance operation caused a power outage in a Verizon data center, which took down JetBlue’s website for about two hours. As a result, 36 JetBlue flights were delayed and four were cancelled.

  • January 2016

On Jan. 28, GitHub’s primary data center experienced a power disruption that caused a quarter of its servers to reboot, which in turn took GitHub’s website down for two hours. GitHub’s library of open-source code is used by 5.8 million active users worldwide.

  • July 2016

On July 27, three burned-out HVAC fans produced smoke and activated the fire alarm inside a data center in Ottawa, Canada, which forced an evacuation and emergency shutdown. The outage downed email servers and left some 50,000 public-service workers cut off from their emails throughout the day.

  • August 2016

On Aug. 8, a small fire in a power supply caused a “massive failure” at Delta’s Technology Command Center in Atlanta, which resulted in a loss of power. Some computers were down for five hours and at least 1,900 flights were cancelled over four days. That small fire cost Delta $150 million.

  • February 2017

On Feb. 28, a typo in code written by an AWS employee resulted in the outage of 54 of the 100 largest online retailers in the world for four hours. Some experienced slow downs, but Express and Lululemon went offline completely. Risk-analytics startup Cyence estimated a loss of $150 million for S&P 500 companies due to the outage.

Cover image by Mathery. 

Check out Quartz’s new book The Objects that Power the Global Economy

Share your comments below. Please read our commenting guidelines before posting. If you have a concern about a comment, report it here.

You might also like...

Well-Being//

The Mental and Physical Toll Of Living Without Power After a Disaster

by Shao Lin, PhD, MD
Community//

Security 101: How to Protect Your Business’s Data

by Joe Resendiz
Community//

Mobile development trends to follow 2019

by Malik Shahmeer

Sign up for the Thrive Global newsletter

Will be used in accordance with our privacy policy.

Thrive Global
People look for retreats for themselves, in the country, by the coast, or in the hills . . . There is nowhere that a person can find a more peaceful and trouble-free retreat than in his own mind. . . . So constantly give yourself this retreat, and renew yourself.

- MARCUS AURELIUS

We use cookies on our site to give you the best experience possible. By continuing to browse the site, you agree to this use. For more information on how we use cookies, see our Privacy Policy.