Can you really rely on the cloud? Recent cloud outages have raised questions about the reliability of cloud providers.
At the core, “the cloud” runs on highly scalable, redundant data centers. These data centers are still comprised of the same components of traditional data centers, including servers, storage, networking, cooling, and power. Typically, within a physical region, a cloud provider builds multiple isolated data centers, with isolated utility power, power generation, and cooling to provide resiliency during issues within a single data center. Customers, if they choose, can engineer their cloud solution to take advantage of this redundancy.
Last week, Azure, Microsoft’s cloud, experienced a prolonged outage in one of their data centers. A series of strong storms hit southern Texas near Microsoft Azure’s South-Central Region. Lightning from these storms caused disruption to the electrical utilities and damage to the cooling systems. When the cooling systems were offline, temperatures increased so rapidly that hardware was damaged, including storage, networking and power supplies. This Azure data center was out for several days and during this time their main source of communication to customers was through Twitter. The outage left a lot of businesses frustrated, with their cloud services either down or degraded. While a lot of people are quick to blame Microsoft for the outage, there is a larger problem - Businesses should not rely fully on cloud providers for disaster recovery.
In an article regarding the Azure outage, cloud expert Dave Bermingham stated, "If your only disaster recovery plan is to call, tweet and email Microsoft until the issue is resolved, you just received a rude awakening," he said. "It is up to you to ensure you have covered all the bases.” Bermingham is alluding to the fact that, while cloud systems are highly resilient, they can still experience issues. Because of the possibility of cloud systems experiencing issues, businesses cannot and should not rely solely on the cloud for disaster recovery. Many people may be surprised to discover that their cloud services are not natively backed up by the cloud providers. Businesses need to have a plan in place for situations such as the outage Azure faced last week.
As with anything, there are risks in using the cloud. However, even with these risks, the cloud can be an extremely useful and beneficial element to a business.This blog is not to discourage you from using the cloud, but rather to encourage engineering a resilient cloud solution and to have a comprehensive disaster recovery plan in place that includes your cloud services.
It is also very important to carefully select your cloud provider. Microsoft is not the only provider that has endured outages. Other well-known providers, such as Google, Amazon Web Services, and Rackspace, have also faced outages. To ensure you have the best cloud experience, it is important to take into consideration elements such as the experience of the cloud provider, the services they provide, infrastructure and architecture of the cloud, disaster recovery services provided, data center and power, and customer support availability.
At The IT Company, we not only encourage you to put a lot of thought into choosing a cloud provider and engineering a solid cloud solution, but we also want to encourage you to create a disaster recovery plan that includes your cloud services. This Friday we will be posting a checklist of things to consider when choosing a cloud provider. Be sure to check back on our social media for the list!