Next to a physical disaster such as a fire or earthquake, many companies fear a data center crash. While no lives are lost, financial, technical and other important information can be lost forever. Data center crashes have happend to three major corporations recently.
As you might suspect, a data center is where a company’s computer systems and servers systems are housed. These systems contain information on company operations as well as telecommunications, security, building operations and other important systems. In general, the larger the company, the larger the data center, which can be as small as a single room or large enough to need several separate buildings.
A lot of design concerns go into designing a data center to allow for security and environmental concerns. They are expensive to build and to maintain. The importance and expense of a data center means that it’s crucial to prevent crashes and to quickly recover from them when they do happen. There are several reasons why data centers can crash:
- Design problems
- Construction problems
- Maintenance problems
- Poor building materials
- Poor equipment
- Human error
Of course, many of these problems can occur simultaneously, so it can be difficult to nail down the exact cause of a crash, especially when it is caused by a small oversight. Here are the stories of three prominent data center crashes that happened recently.
1. Amazon North Virginia Data Center Crash
Amazon, the world’s largest online retailer, experienced a data center crash in its North Virginia facility on October 22, 2012. The crash disabled several websites and mobile apps, including Someecards, Foursquare and Flipboard, which used Amazon’s cloud hosting service.
It took 12 hours to fully restore services, resulting in the probable loss of millions of dollars of sales — right as the crucial holiday shopping season was ramping up. While the company does not seem to have released the cause of that crash, an electrical storm had caused the same center to crash in June. That crash affected Pinterest, Netflix and Instagram for an entire weekend.
2. Twitter Data Center Crash
The prominent social networking micro-blog site was out of commission for about two hours July 27, 2012, when two data centers failed. The company did not announce the cause for the failure, nor the location, but many speculated it may have been caused by an increased amount of “tweets” about the Olympic Games taking place in London at the time.
Twitter officials said two parallel systems failed in an “infrastructure double whammy,” according to Computerworld.com’s account. It was the second data center failure in a month and a half for the company. Previously, on June 21, the microblogging service went down for an hour, started to recover and then failed again.
“The company blamed that outage on a cascading bug, a type of problem that spreads from one software element to others,” according to ComputerWorlds.com.
3. Fisher Plaza, Seattle Data Center Crash
The July 30, 2006 Fisher Plaza crash can be considered the “Titanic” of data center crash. Fisher Plaza, consisting of two facilities with a total of 296,000 sq. ft., is the data center for many companies. It was built to withstand earthquakes and other disasters and to run on its own power.
Apparently, however, it was not built to survive electrical failures. Fisher Plaza’s owners blamed the crash on a “Seattle City Light power outage that triggered a generator start and load transfer event,” according to a press release published on CRMBuyer.com. However, Seattle City Light denied the outage took place.
According to CRMBuyer’s account of the crash, “There had been a minor system change within City Light’s network at the time the outage began, which City Light engineers forensically examined. The engineers speculated that the three-cycle power supply serving Fisher Plaza may have experienced a momentary fluctuation on just one of its cycles. This type of a minor single-cycle fluctuation is not considered an outage or a brownout.
“If Fisher Plaza’s settings were over-sensitive, then a minor fluctuation in one cycle might have been enough to take Fisher Plaza offline. Facilities with their own electrical generation equipment sometimes set their sensitivity levels too high, ostensibly to prevent damage to their own electrical equipment.”
These prominent cases prove that even with the best preparation data center crash can happen. Is your company prepared? Share and Save your documents safe.