Recently, we looked at a few of the top stories in the data center world from 2017. In part two, we conclude our year-end wrapup with a look at Microsoft's Azure outage, Oracle's big cloud deal and more.
A Deeper Look Back at the Biggest Stories of the Past Year
Like the AWS case, human error caused British Airways IT shutdown in May. The engineer disconnected a power supply in a data center near Heathrow airport which caused a power surge and created havoc in the network. The travels of 75,000 passengers were interrupted by the IT failure.
Apple’s Environment Responsibility Report provided information regarding the rapid growth of its leased data center usage. The energy use in its third-party facilities quadrupled in the last four years. In fiscal 2016, it was using 180,205,500 kWh in its colocation facilities compared to 38,552,271 kWh in 2012. The rapid growth of Apple's colocation use shows that large companies cannot rely completely on their own infrastructure to meet demands.
Microsoft added cloud availability zones to its Azure platform. It provides customers another tool to keep their infrastructure operational during outages. Now customers can replicate workloads across zones for higher redundancy. Google Cloud and Amazon AWS already had the capability. Microsoft was playing catchup to keep up with the competition.
In September, Microsoft Azure was out for seven hours due to a problem with the fire suppression system. A false fire alarm caused a series of events that led to the service outage.
Oracle has signed a deal with AT&T to move thousands of large-scale databases to Oracle’s cloud infrastructure. The agreement gives AT&T access to Oracle’s Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Database as a Service (DBaaS) and Software as a Service (SaaS). For its B2B customers, Oracle has also started Blockchain as a Service in its cloud.
In September, Equifax revealed that a data breach has let hackers steal the personal information of 143 million US consumers. It is one of the largest known data breaches. The credit scoring industry can end up collapsing due to this event.
Companies are using machine learning and IoT sensors to create applications for facility monitoring. These applications will allow a limited number of human operators to maintain hyperscale data centers. It can automate processes and make data center management safer. It's still in the early stages because AI training takes time. But expect to see more of machine learning and IoT in the future.
Verizon sold its cloud and managed-hosting services to IBM for $3.6 billion. Iron Mountain, a global leader in storage and information management, bought two large data centers from Credit Suisse, a leading financial company, in Singapore and London for $100 million. Digital Bridge Holdings, LLC, an investment company, acquired Vantage, a leader in data center solutions.
In 2017, data centers saw a lot of human error related outages.
Data center managers should reevaluate their own playbooks and processes to stay safe. Security and breaches were also issues. On the other hand, progress was made through underwater intercontinental high-capacity cable, Google Expresso and machine learning. Hopefully, 2018 will also be a great year for data center progress.