Google Blames Outage on Asian Web Traffic Jam

Google experienced slowdowns and outages to its Google News and Gmail services on the afternoon of 14 May.

Users of those services reacted in a variety of ways. On Twitter, which has a search function that allows for a near-real-time survey of what people are microblogging on various topics, users alternately shrugged off the outage or expressed harsher opinions. “I didn’t even know about the outage until [I] saw [the] news,” one wrote, while another opined, “Epic fail whale this morning.”

Users reported that other Google services, such as YouTube, were also showing slower activity.

In a statement sent to eWEEK, a Google spokesperson said:

“Earlier today, Google News was temporarily unavailable for many users, from approximately 3:30 AM until around 7:00 AM, Pacific Time. This issue has now been resolved. We know how important Google News is to our users, so we take issues like this very seriously.”

Service was subsequently restored. In a blog post, Google blamed a computer error that led to a Web traffic jam in Asia.

An error in one of our systems caused us to direct some of our Web traffic through Asia, which created a traffic jam,” Urs Hoelzle, senior vice president of Operations for Google, wrote in the 14 May blog post. “As a result, about 14 percent of our users experienced slow services or even interruptions.”

Hoelzle added, “All planes are back on schedule now.”

Google occasionally experiences short outages. In late February, Gmail was down for users in the United States and the United Kingdom for 2.5 hours. Google executives described that particular incident as a service outage. And in August 2008, Google Gmail and Google Apps experienced nearly 15 hours of downtime.

As enterprise users increasingly migrate to cloud-based services, the cloud’s ability to maintain constant uptime has become a question not only of convenience, but also of revenue. Other cloud services, such as the early test release of Microsoft Azure, have experienced downtime in the past few months of 2009.

However, unless customers and businesses are willing to pay for five-nines reliability, outages of cloud-based services may very well need to be accepted as a part of Web 2.0 life.