Despite Publicized Outages, Cloud Opportunity Remains Real
For sure, the “cringe” word right now in cloud computing is outage. In the past few weeks, a few big ones have generated a lot of public noise, both for the inconvenience they caused online users and the questions they generated for the large companies involved.
In mid April, Amazon’s Elastic Compute Cloud (EC2) service went down for more than two days, as reported here, crashing social news, networking and communication sites Reddit, Foursquare and Hootsuite among others and sparking turbulence in the IT community.
Making matters worse, on May 10 and then again on May 12, Microsoft’s Business Productivity Online Services (BPOS), collaboration tools delivered as a service, went down for about nine hours, the news of which was reported here along with some choice customer responses.
Both Amazon’s and Microsoft’s outages rattled users and prompted reliability questions about the cloud – not only because they happened in the first place, but also because of the prominence of the cloud services affected. Understandably, the downtime episodes prodded users to ask why Amazon’s and Microsoft’s business continuity systems didn’t prevent, or at the very least minimize, the outages.
Indeed, some customers and cloud critics pointed to these incidents and concluded that cloud technology isn’t ready for prime time. But, a closer look shows that it’s not only cloud computing systems that fail occasionally. On-premise IT infrastructures cough and sputter from time-to-time as well.
The difference, of course, is that internal IT outages are much quieter and less public than cloud outages. In both cases, however, the importance of solid disaster recovery and backup solutions to ensure business continuity cannot be overstated.
The first case in point is the mid June connectivity outage that scrubbed United Airlines computers for five hours, a downtime event that stranded thousands of travelers and forced flight cancellations and delays. United’s IT structure isn’t cloud-based, and the airlines subsequently acknowledged that its business continuity systems didn’t operate as expected.
Secondly, during the Amazon outage, Netflix, which hosts its entire video streaming service on Amazon’s cloud, remained available during the downtime event. This wasn’t from happenstance or serendipity, but rather the result of best practices system design in which Netflix constructed its IT to be resilient to various types of failure. As a result, their video streaming systems remained up and operational despite Amazon’s outage.
No IT system is fail safe and no service, whether on-premise or in the cloud, is 100 percent fool proof. Still, while the downtime risk is low, there’s no such thing as an insignificant outage. Businesses should anticipate, or even expect, a certain level of downtime and, with their channel partner trusted advisors, structure disaster recovery systems accordingly.
It’s easy to cringe at the impact of loud and well-publicized cloud services outages, but even with the blaring headlines and raucous blogosphere, we at Ingram Micro aren’t seeing any slowdown in terms of the penetration of cloud services in the market.
In fact, we’re urging our channel partners not to allow the rare public cloud outages to get in their way, or even delay, offering cloud services to their customers. Instead, channel partners should focus on developing a business continuity plan to ensure their customers’ business operations are not impacted by such outages.
Embrace the cloud and capture the value proposition—agility, utility and ubiquity—to better serve your customers and generate more revenue. Building a cloud practice is too exciting and lucrative of an opportunity to put off even for another minute.
Renee Bergeron is VP of managed services and cloud computing at Ingram Micro, overseeing such efforts as Ingram Micro Cloud. Monthly guest blogs such as this one are part of MSPmentor’s annual sponsorship program. Read all of Bergeron’s guest blogs here.